Search

At present, there’s a lack of combustion visualization in the combustion control of heating boilers. To understand the combustion of coal in the furnace, only experienced workers can observe it through visual inspection. Using infrared thermal imaging technology to monitor the combustion can realize combustion visualization. This paper analyzed and solved two problems: the installation position and number of infrared cameras, and the infeasible of using infrared cameras observing the combustion condition in the furnace through heat-resistant glass. Monitored parameters such as oxygen content, furnace temperature and smoke exhaust temperature, and monitored the concentration of PM, NOx, and SO2 in the main atmospheric pollutants in the flue gas. After calculation, the air leakage coefficient when the inspection doors are opened for observation is 0.04. This value still includes the sum of air leakage from coal hopper, furnace door, grate side seal, peep holes and other parts. The monitored average emission concentration of PM decreased by 16.28%, from which we can concluded that the use of thermal imaging technology to monitor the combustion in the furnace is conducive to emission reduction. The application of thermal imaging technology implementation of coalfired boiler combustion visualization is feasible. [ABSTRACT FROM AUTHOR]

It is of great importance to obtain the exact location and severity of slagging deposits on the waterwall surface of an operational boiler to avoid aimless soot-blowing and reduce steam consumption. In this paper, an effective waterwall surface temperature monitoring method is proposed to determine the slagging locations. It has been noted that the temperature difference of the waterwall surface before and after soot-blowing varied with the waterwall location, with more than 80 °C covered with slag and less than 20 °C found clean. According to this, a slagging temperature index was developed to describe the severity of slagging deposits on the waterwall surface. Results indicated that the process of slagging deposit growth included four stages, with the slagging temperature fluctuating in the range of about 90–110 °C in stage III, followed by a rapid drop below 60 °C in stage IV. Furthermore, a digital image monitoring system was used to validate the slag growth process and study the relationship between deposit thickness growth and area expansion. This novel approach provides automated and accurate guidance for each soot blower around the furnace, which reduces soot-blowing steam consumption and avoids serious slagging on the waterwall surface. [ABSTRACT FROM AUTHOR]

Ammonia (NH 3), as a potential carbon-free alternative fuel, can be blended into coal-fired boiler to achieve significant pollution reduction and carbon reduction, but there are concerns about high NOx emissions due to high nitrogen content. According to the characteristics of coal/NH 3 co-combustion, a dual-fuel co-combustion model with strong adaptability and high accuracy was established in this study through Chemkin software to study the influence of different injection positions and blending ratios on combustion characteristics and NOx generation process. Then, the co-combustion model was applied to the three-dimensional CFD calculation process of a 330 MWe front-fired boiler, and the combustion characteristics, NOx distribution and reaction process were calculated when cal. 20% NH 3 was blended in the primary air. The results show that when cal. 20% NH 3 is blended, the change of NO content mainly occurs in ignition zone and flame zone, and the transformation behavior of N in NH 3 is optimized to a 15-step elementary reaction; The temperature distribution in the furnace is similar, and the average temperature at the furnace outlet decreases from 1033 °C to 988 °C, while NH 3 have a preferential combustion reaction with air than coal, resulting in a decrease in the burnout rate of coal; The NOx concentration at the furnace outlet decreases from 355 mg/Nm3 to 281 mg/Nm3, which is 20.85% lower than that under the pure coal burning condition, and the variation range of O 2 concentration and unburned NH 3 concentration is small. • The NO transformation behavior was optimized to a 15-step elementary reaction. • The change of NO content mainly occurred in the IZ and FZ when blending cal. 20% NH 3. • The burnout rate of pulverized coal was reduced by blending NH 3. • NOx concentration at the furnace outlet reduced by 20.85% when blending cal. 20% NH 3. [ABSTRACT FROM AUTHOR]

In this paper, a novel model with a parallel structure is proposed to predict NOx emissions from coal‐fired boilers by using historical operational data, coal properties, and convolutional neural networks. The model inputs are processed and passed into three parallel subnetworks with well‐designed building blocks. The features learned by the three subnetworks are fused and used to predict NOx emissions from a 330‐MW pulverized coal‐fired utility boiler. A comprehensive comparison of different prediction models based on deep learning algorithms shows that the prediction model proposed in this paper outperforms other prediction models in terms of root mean square error criteria. The results show that the parallel structure is key to obtaining accurate predictions while reducing model complexity. This suggests that the model's performance can be improved by designing the model architecture. [ABSTRACT FROM AUTHOR]

The variations in the boiler operation conditions have a great effect on the combustion characteristics and the pollutant formation in furnaces. This work aims to investigate the effects of operational parameters on NOx formation and its distribution in furnaces using the numerical simulation method to obtain the optimum control strategy for reducing NOx emissions. The numerical simulation models of pulverized coal combustion in furnaces involving flow, heat transfer, combustion and NOx formation are established. Taking a 600 MW supercritical opposed firing pulverized coal boiler as the study object, a full-scale three-dimensional physical model of the boiler is constructed with Gambit software. On this basis, the pulverized coal combustion and the NOx formation under various boiler loads are numerically simulated using the software of Ansys Fluent 2021R1, and the accuracy and the reliability of the models established are verified by comparing the simulation data with the field test data. According to the combustion numerical simulation of 128 groups of operating conditions, the effects of boiler load, the air rate and the air temperature on combustion and NOx formation have been emphatically investigated. The simulation results indicate that the formation of NOx and the NOx concentration distribution are mainly affected by the oxygen concentration and the temperature in the furnace. Especially, the effects of the variation in the excess air coefficient, the over-fire air (OFA) ratio, the primary air ratio and the internal secondary air ratio on NOx concentration distribution vary greatly. When the air temperature increases the overall NOx concentration in the furnace increases, and the influence of the secondary air temperature and the OFA temperature is greater than that of the primary air temperature. Large amounts of simulation data are a necessary data source for further study on the NOx prediction model at the economizer outlet, which can improve the prediction ability and the generalization ability of the NOx prediction model. [ABSTRACT FROM AUTHOR]

This paper presents a small and efficient model for predicting NOx emissions from coal-fired boilers. The raw data collected are processed by the min–max scale method and converted into a multivariate time series. The overall model's architecture is mainly based on building blocks consisting of separable convolutional neural networks and efficient channel attention (ECA) modules. The experimental results show that the model can learn good representations from sufficient data covering different operation conditions. These results also suggest that ECA modules can improve the model's performance. The comparative study shows our model's strong performance compared to other NOx prediction models. Then, we demonstrate the effectiveness of the model proposed in this paper in terms of predicting NOx emissions. [ABSTRACT FROM AUTHOR]

The advancement of co-processing solid wastes in coal-fired boilers is significant for waste recycling and contributes to the sustainable development of the coal-fired power industry. However, concerns over the emission of dioxins during co-processing have prompted a comprehensive investigation into the dioxin emission properties. In this study, the PCDD/F emission concentrations of seven coal-fired boilers, including three pulverized coal boilers and four circulating fluidized bed boilers were examined. The results indicate that co-processing solid wastes in coal-fired boilers did not lead to an increase in the mass concentration of dioxins in either the flue gas or solid samples, and the international toxic equivalents (I-TEQ) of dioxins in the flue gas complied with prevailing emission standards (0.1 ng I-TEQ/Nm3) in China, proving that coal-fired boilers co-processing would not raise the emission risk of dioxins. The types of waste during co-processing had minimal effect on the I-TEQ of dioxins. A significant proportion of PCDD/Fs was observed in the ash samples, while only 13.0–25.7% and 0.7–6.8% of dioxins were distributed in the boiler slag and the flue gas, respectively. The emission factor of dioxins under the blank conditions ranged from 0.009 to 0.327 ng I-TEQ/kg-coal, while it ranged from 0.015 to 0.129 ng I-TEQ/kg-(coal+waste) under the co-processing conditions. The reduction of emission factor under co-processing condition could be attributed to the significant decrease of dioxins I-TEQ. [Display omitted] • Co-processing wastes in boilers did not lead to an increase in the I-TEQ of dioxins. • The type of waste during co-processing had minimal effect on the I-TEQ of dioxins. • A significant proportion of PCDD/Fs was observed in the ash samples. • The emission factor of dioxins ranged from 0.015 to 0.129 ng I-TEQ/kg-(coal + waste). [ABSTRACT FROM AUTHOR]

The production of nitrogen oxides (NOx) in coal-fired boiler combustion has been found as a significant source of environmental pollution. Flue gas denitrification is a standard NOx control technology for small- and medium-sized coal-fired boilers. Achieving steady-state control in flue gas denitrification can be challenging since coal-fired boiler systems have complexity and significant delay. A model based on a learning-based K-nearest neighbor (KNN) query mechanism created for NOx output soft prediction is proposed in this study. First, a knowledge base in the proposed model is established through spatial division in accordance with the previous combustion parameters. Moreover, the clusters are established based on the output NOx values. Next, the domain of values of combustion parameters for the respective cluster is obtained. Second, the optimal cluster is selected using the knowledge base for an input vector q with new combustion parameters ( q 1 , q 2 , ... , q n). Lastly, the K tuples in the cluster the closest to the values of the input vector q are adopted to predict the output NOx value of q. The predicted NOx value can serve as a feedforward signal to control the output of the reductant for accurate denitrification. As revealed by the experimental results, the proposed practical model, capable of conducting the prediction in a sub-second time, is highly competitive with existing techniques. Furthermore, a deep learning algorithm (DLA) is designed, whereas it underperforms the KNN model. [ABSTRACT FROM AUTHOR]

Copyright of Clean Coal Technology is the property of Clean Coal Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

• A segmented model of ultra supercritical boiler water-cooled wall is established for dynamic simulation. • The uneven heat flux distribution on the water-cooled wall is considered. • The step response characteristics of different flow loops are compared under different load conditions. • The influence of combustion instability on the water-cooled wall is studied under different load conditions. The dynamic characteristics of water-cooled wall are of great significance to the evaluation of the safety and flexibility for the boiler during peak shaving process. However, few dynamic studies considered the uneven heat flux distribution on the water-cooled wall, let alone the variable heat load caused by the combustion instability. In this paper, a segmented dynamic model is developed for the water-cooled wall of a 660 MW ultra supercritical boiler. The model is validated under steady and dynamic working conditions, respectively. Then, the step responses of the steam temperature in different flow loops under different load conditions are comparatively studied under the disturbances of single-parameter changes. In the cases of 10 % step decrease of either the feedwater temperature, the heat load or the feedwater mass flowrate, the outlet steam temperature is influenced greatly by the step decreased feedwater temperature under 100 % BMCR condition, while it is impacted most by the step decreased feedwater mass flow under 50 % THA condition. Under these disturbances, the outlet steam temperature of the tube with higher heat loads has larger variation and longer response time. In addition, the simulated combustion instability by the sinusoidal fluctuating heat load on the water wall is employed to explore the influences of the flame fluctuation amplitude and frequency to the water wall. The maximum steam temperature rises with the increase in amplitude but drops with the frequency of flame fluctuation. The thresholds for the flame fluctuation amplitude and frequency are obtained under different load conditions. [ABSTRACT FROM AUTHOR]

Environmental issues have received significant global attention, and achieving ultra-low emission has become a crucial goal for thermal power plant. Pollutant prediction is essential to achieving ultra-low emission control. Due to the frequent adjustment of operating conditions, the coal-fired power generation process is a typical non-stationary process with complex changes along the time direction. A single model cannot effectively portray the complex non-stationary characteristics and model mismatch may occur in the prediction, which brings new challenges for modeling coal-fired SO 2 emission. Therefore, the condition-driven category boosting (CatBoost) soft sensor model is proposed for real-time prediction of SO 2 concentration. First, the thermal process time series is rearranged into different data slices. These data slices vary along the direction of the condition indicator, which can characterize the process characteristics on different slices. Second, the load-based sequential condition division (LSCD) algorithm is designed to realize the division of boiler operation process. The characteristics of process are effectively captured by variational auto-encoder (VAE), and the data slices with similar process characteristics are merged into the same mode. Non-stationary process along the time direction is reorganized into different condition modes along the load direction. Finally, the high-resolution local CatBoost models under different modes are constructed to achieve fine-grained prediction of SO 2 concentration. The effectiveness of the proposed condition-driven model is verified by an industrial example of 1000 MW coal-fired boiler. The prediction results show that the proposed model can effectively divide the operation conditions and achieve accurate prediction of SO 2 concentration. In the typical models in SO 2 predicting, the proposed model has better performance with a determination coefficient of 0.92. • Propose a condition-driven model to predict the SO 2 concentration in coal-fired boiler under frequent load variations. • Design a load-based sequential condition division (LSCD) algorithm to divide non-stationary thermal processes. • Develop high-resolution local CatBoost models for different modes of operation conditions. • Compared with the typical methods, the proposed condition-driven CatBoost model has better prediction performance. [ABSTRACT FROM AUTHOR]

Copyright of Clean Coal Technology is the property of Clean Coal Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Clean Coal Technology is the property of Clean Coal Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Clean Coal Technology is the property of Clean Coal Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Clean Coal Technology is the property of Clean Coal Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

• Three typical monotonicity relationships in boiler NOx emissions are embedded into PINN. • Synthetic samples are added to ensure the physical consistency. • The proposed PINN outperforms conventional models in terms of accuracy, interpretability and generalization. • PINN maintains desired monotonic behaviors even under unknown operating conditions. Nitrogen oxides (NO x) are major pollutants emitted from coal-fired power plants. Accurately predicting the NO x concentration at the boiler outlet is crucial for optimizing unit control and reducing NO x emissions. Machine learning (ML) methods have been widely employed to predict NO x emissions from coal-fired boilers. However, the interpretability and generalizability of ML models are constrained by the absence of physical insights. To overcome this limitation, we introduce a physics-informed neural network (PINN) that embeds three typical monotonic relationships into the NO x prediction model by combining regression error, regularization, and physical inconsistency for joint training. The results for a 660-MW ultra-supercritical boiler reveal that the developed PINN method outperforms the other three traditional pure-regression ML methods (RF, DNN, LSSVR) regarding prediction accuracy. Moreover, additional field experiments demonstrate that the PINN model exhibits robust generalization capabilities when predicting NO x emissions under new operating conditions. This study highlights the benefits of using reliable physical information to improve the scientific interpretability and extrapolability of an ML model. [ABSTRACT FROM AUTHOR]

A novel approach for the dynamic prediction of boiler NO x is proposed that integrates the graph convolutional network (GCN) with the gated recurrent unit (GRU) model, referred to as GC-GRU model. The operational status of boiler parameters is treated as graph-structured data, and the GCN is used to extract the spatial characteristics of boiler parameters which will then be taken as the input of the GRU that considers the dynamic characteristics of boiler operation. The GC-GRU model can simultaneously take into consideration the spatio-temporal characteristics of boiler NO x to realize accurate prediction of boiler NO x. This study presents a complete procedure for boiler NO x prediction ranging from the selection of key feature variables for model input to the optimization of model hyperparameters. Specifically, the MIC analysis is used to extract the key boiler parameters that are closely associated with NO x emission. The Genetic Algorithm (GA) is proposed to optimize the model hyperparameters (GA-GC-GRU model). The results show that the GA-GC-GRU model exhibits superior performance because of its ability to capture the spatio-temporal characteristics of boiler NO x. Specifically, the test MAPE of the model is 2.41%, reduced by more than 25% compared to those of SVM, DNN, and LSTM models. • Developed a GC-GRU model for the dynamic prediction of coal-fired boiler NO x under load-cycling conditions. • Model can simultaneously take into consideration the spatio-temporal characteristics of boiler NO x. • Applied MIC analysis to extract the key boiler parameters in their original forms. • Used genetic algorithm to explore the optimal hyperparameter configuration of the GC-GRU model. • Model shows superiority over SVM, DNN, and LSTM models. [ABSTRACT FROM AUTHOR]

The temperature field is a critical factor for ensuring the safe combustion and energy conservation in boilers. However, an effective method for reconstructing the temperature field near the water wall is still under exploration. In this paper, a method for online reconstruction of the temperature field distribution near the water wall in a 330 MW tangentially fired coal boiler is proposed, which progresses from a well-established model for the entire furnace. A two-branch fusion network for transfer learning (TBFN-TL) method is proposed, incorporating additional key parameters, heat flux, during the transfer process to enhance the effectiveness. A Bayesian hierarchical neural architecture search (BHNAS) method is proposed to optimize the configuration of the hidden layers in building neural networks. Compared with computational fluid dynamics (CFD) results, the mean absolute percentage error (MAPE) of the reconstruction results for the entire furnace model, traditional transfer learning methods, and the proposed TBFN-TL are 11.57%, 5.738%, and 2.052%, respectively, demonstrating a significant enhancement. The proposed BHNAS method extends the search optimization space, obtaining more excellent configurations for the hidden layer nodes. The proposed methods have significant implications for temperature field reconstruction, the field of transfer learning, and the optimization of hidden layer configurations. • Developed a model for real-time temperature field reconstruction near furnace walls. • Proposed TBFN-TL, enhancing transfer learning with key target domain features. • BHNAS method introduced for optimizing hierarchical neural network configurations. [ABSTRACT FROM AUTHOR]

• First industrial scale NH 3 -coal cofiring testing under high cofiring ratios. • Demonstrate the technical feasibility of NH 3 cofiring in large-scale boilers. • Find optimal SOFA ratio in the range of 15–20 % for NO x and fuel burnout. • Find optimal boiler O 2 in the range of 3.0–3.7 % for NO x and NH 3 slip control. This paper reports the first industrial scale testing on ammonia (NH 3) cofiring in a 40 MW th coal-fired boiler under high ammonia cofiring ratio (E N H 3 , percent of thermal input) ranging from 0 % to 25 %. A full-scale cofiring swirl burner is designed to inject NH 3 gas with pulverized coal through the primary air pipe of the burner. The effects of ammonia cofiring ratio and key boiler operating parameters on boiler NO x emission and fuel burnout are examined. The results show that the boiler can establish stable ignition and combustion under all ammonia cofiring conditions in this study. The burnout of coal under ammonia cofiring conditions are better than those of coal combustion cases. Overfire air (SOFA) has significant influences on boiler NO x emission and burnout of coal and NH 3. It is found that there exists optimal SOFA ratio in the range of 15–20 % within which both boiler NO x emission and NH 3 slip can be well controlled. Under 20 % SOFA ratio, boiler NO x emission tends to increase and then decrease with the increase of E N H 3 reaching the peak value under E N H 3 = 5 %. As a result, the NO x emission can be reduced to the level lower than that of coal combustion case under E N H 3 > 20 %. Boiler excess O 2 also shows significant influences on boiler NO x emission and NH 3 slip. With the decrease of boiler O 2 , boiler NO x emission decreases significantly while NH 3 slip increases rapidly. It is found that there exists optimal boiler O 2 in the range of 3.0–3.7 % within which both boiler NO x emission and NH 3 slip can be kept at relatively low levels. Injection of NH 3 from the side wall of furnace is also tested in this study. The NO x emission shows similar trends with the change of E N H 3 in the range of 0–20 %. However, above this range the NO x emission tends to increase with further increase of E N H 3 . The results of this paper demonstrate the technical feasibility of NH 3 cofiring in large-scale coal-fired boilers under high cofiring ratios and provide important guidance to the operation of boilers. [ABSTRACT FROM AUTHOR]

• A hybrid dewatering and air-preheating system mounted to coal-fired boiler was investigated. • New attainable ideal and unavoidable cases for combustion process were identified. • Improved potential of combustion system was conducted as 74.4%. • The critical dewatered moisture content was found as 20%. • A saving of about 29.54% in fuel consumption was determined. In this study, a parametric advanced exergy analysis was conducted for the waste heat recovery of a pulverized coal-fired power plant. In this aim, a new attainable ideal and the technically available unavoidable case for the combustion process were identified. Under the identified improved potential term, different unavoidable cases were parametrically investigated for the hybrid dewatering and air preheating process to improve the combustion process. In this aim, different removed moisture contents, air preheating conditions, and effectiveness of both internal and external heat exchangers were thermodynamically evaluated in terms of the combustion process. The improved potential was calculated as 74.4% for the whole combustion system. The critical moisture removal ratio was calculated as 20%. The effects of rehabilitation on improved potential were found as 2.80 points per 1% of effectiveness increase for the self (internal) heat exchangers of the boiler, 1.68 points per 1% of effectiveness increase for HE-I and HE-II (external heat exchangers), and 0.62 points per 1% increase of dewatered moisture ratio. Besides, a saving of about 29.54% was determined in fuel consumption for the most appropriate unavoidable condition. [Display omitted] [ABSTRACT FROM AUTHOR]

Copyright of Coal Science & Technology (0253-2336) is the property of Coal Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

• A model of biomass gasification coupled with coal-fired boiler is proposed. • The performance of the coupled system at various loads is studied. • The coupled system efficiency reaches 84% at optimum conditions for various boiler loads. • The maximum reductions of NO and SO 2 emissions reach 5.99% and 16.80% respectively. • An efficient utilization method for straw in a coal-fired power plant is provided. The utilization of agricultural wastes in existing pulverized coal power plants is an attractive option to alleviate environmental pollution and reduce over-exploitation of fossil fuels. A coupled system model of biomass gasification coupled to a coal-fired boiler is established in Aspen Plus and successfully validated by experimental data. A 20 t/h straw gasifier operates at the rated capacity and the straw gas is introduced to the boiler running at different loads. The co-firing ratio increases with the reduction of boiler load. Results indicate that the main parameters, such as furnace combustion temperature, flue gas temperature, and NO and SO 2 emission decrease with the reduction of boiler load. Compared to pure coal combustion, co-firing can reduce the furnace combustion temperature and increase the flue gases temperature. More importantly, the coal consumption, and NO and SO 2 emissions are reduced at all loads, especially at lower loads. The excess air ratio should be adjusted to obtain the optimum combustion performance in the furnace, but there is still a slight drop of around 0.2% in boiler efficiency when co-firing. Meanwhile, the coupled system efficiency at various loads can reach slightly more than 84% under optimum conditions. [ABSTRACT FROM AUTHOR]

Nowadays, much attention is paid to the elimination of heat sources poisoning our environment. The air pollution is the most dangerous form of pollution of the environment, due to its direct impact on all living organisms, coverage of large areas and easy movement of pollutants. Therefore, air protection is one of the priority directions of each voivodeship. Hence, many programmes are being developed to promote the replacement of the old heating devices with the new ones and the installation of renewable energy conversion devices. There is no doubt that the equipment which emits the most harmful compounds into the atmosphere includes the old boilers fired with coal, wood or wood chips. In winter, chimneys with black circular smoke are a frequent sight. This problem can only be solved by connecting houses to the public heating network or by removing the old stoves. In this publication, an analysis of the profitability of boiler replacement with coal was carried out in terms of the material and environmental benefits. It is proposed to replace this boiler with a pellet boiler, an eco-pea boiler and a gas condensing boiler. The boilers were selected on the basis of the calculated thermal output, which is equal to 18.56 kW. [ABSTRACT FROM AUTHOR]

• An on-line combustion optimization framework for coal-fired boiler is proposed. • The boiler dynamic combustion model with self-adaptive capability is constructed. • Domain maximum substitution strategy is introduced into the decomposition-based multiple target evolutionary algorithm (D/MOEA) to enhance the optimization power of D/MOEA. • Mutation strategy-based case-based reasoning (MS-CBR) is proposed to gain the combustion optimization adjustment strategy online. • The thermal efficiency could be increased by 0.588 %, and NOx emission could be reduced 14.183 mg/m3, and the difference in operating load reached 0.696 MW. It is important to gain the on-line combustion adjustment strategy under ensuring the running working condition of boiler unchanged. Because it can not only significantly improve the whole combustion environment of boiler, but can ensure the stability and safety of boiler in optimization. This article proposes a kind of new-style on-line combustion optimization framework for boiler combustion system, and this study is to quickly acquire the combustion adjustment strategies to improve the whole combustion environment of boiler under ensuring the operating working condition of boiler unchanged. First, improved culture algorithm based on improved particle swarm optimization and double mutation evolution strategy (CIPSODM) and long-short term memory neural network (LSTM) are combined to establish the boiler dynamic combustion model with self-adaptive capability. The improvement of CIPSODM involves two aspects: introducing clustering analysis to the population space of culture algorithm and adding the double mutation thought to the belief space of culture algorithm. And then domain maximum substitution strategy (DMS) is introduced into the decomposition-based multiple-objective evolutionary algorithm (DMS- D/MOEA) to obtain a great deal of combustion optimization cases offline, and an optimization cases base is established by gathering all optimization cases. Whereafter, mutation strategy is added to the case-based reasoning (MSCBR) to online obtain the combustion optimization adjustment strategy with the same running working condition as the current running working condition of the boiler for stable and safety combustion. Meanwhile, to illustrate the availability of the combination of CIPSODM-LSTM, DMS-D/ MOEA and MSCBR, several different on-line optimization approaches are adopted to acquire the combustion adjustment strategies online for comparison. The experiment results illustrated that the thermal efficiency could be increased by 0.588 % and NOx emission could be reduced 14.183 mg/m3 using proposed online combustion optimization approaches. Furthermore, the difference in operating load between adjustment strategy and current boiler reached the minimum value 0.696 MW in all optimization results. Consequently, proposed the on-line combustion optimization framework can gain the on-line combustion adjustment strategies to maximize thermal efficiency and decrease NOx emission so as to improve the whole running environment of boiler. [ABSTRACT FROM AUTHOR]

The flower pollination algorithm (FPA) is a nature-inspired optimization algorithm. To improve the solution quality and convergence speed of FPA, we proposed a novel flower pollination algorithm (NFPA) which is a hybrid algorithm based on original FPA and wind driven optimization algorithm. Simulation experiments demonstrate that NFPA has better search performance on classical numerical function optimizations compared with other the state-of-the-art optimization methods. In addition, the NFPA is adopted to optimize parameters of fast learning network to build thermal efficiency model of a 330 MW coal-fired boiler and a well-generalized model is obtained. Experimental results show that the tuned fast learning network model by NFPA has better prediction accuracy and generalization ability than other combination models. [ABSTRACT FROM AUTHOR]

[Display omitted] • A 3D combustion temperature prediction method, GDNN, uses CFD simulations to achieve online reconstructions was proposed. • A temperature field correction method employs actual measurement points was proposed to correct the prediction results. • Optimal parameters for the proposed prediction and correction method were selected through a parameter analysis. • The proposed GDNN and correction method jointly yield the lowest error, demonstrating their real-world effectiveness. Predicting furnace temperature distribution is vital for coal-fired boiler safety. Existing methods, including finite element calculations and three-dimensional (3D) reconstruction still face limitations. A 3D combustion temperature field prediction method, GDNN, which utilizes offline-computed computational fluid dynamics (CFD) simulation results for online reconstructions of entire boiler, was proposed. GDNN method leverages the knowledge of the temperature field acquired by the base neural network model and Gaussian processes. Furthermore, a temperature field correction method is introduced, which employs intermediate variables of the GDNN model and measured values from temperature sensors to establish a correction model for the entire predicted temperature field. We compared GDNN's effectiveness with four well-established algorithms: Extreme Learning Machine (ELM), Least Absolute Shrinkage and Selection Operator (LASSO), Deep Neural Network(DNN), and Radial Basis Function (RBF) network, by also substituting these algorithms as the base model in our proposed method. The experimental results demonstrate that the proposed prediction method exhibits the highest performance, and the correction method effectively improves the overall results. The optimal parameters for predicting and correcting 3D furnace temperature field results were determined through experimental comparison, and the proposed method was applied to a 350 MW boiler, achieving an error of 2.41%, proving its real-world effectiveness. [ABSTRACT FROM AUTHOR]

To achieve low-NO x emission in the coal-fired industrial boiler, the radial air staging method is proposed. The location of the over-fired air (OFA) is an essential parameter as it affects the combustion stability, NO x generation, burnout, and slagging trend. These are related to the gas-particle flow such as the recirculation region, gas mixture, and gas-particle mixture. Therefore, in this paper, taking a coal-fired industrial boiler with radial air staging method as an object, in a cold-modeling furnace, the gas-particle flow characteristics under various OFA locations (L D of 1.29, 1.43, 1.72, and 2.00) are studied. When OFA is radially arranged, the OFA jet diffuses towards the furnace center. As the L D increases, the size of the central recirculation region increases, thereby improving combustion stability. When the L D is 2.00, the mixing of the OFA jet and PCC jet obviously is delayed. Moreover, an increase in L D leads to more returned particles in the furnace center, promoting the reduction reaction of NO x. In addition, as the L D increases from 1.29 to 1.72, the gas-particle mixing uniformity is enhanced, which is beneficial to the uniform temperature distribution. Comprehensively, the L D of 1.43–1.72 is recommended when radial air staging is adopted in coal-fired industrial boilers. • The radial air staging method accelerates the mixing of OFA and the main airflow while reducing the mixing intensity. • With the increase of L D , the obstructing effect by radial OFA on the diffusion of the swirling main airflow is weakened. • The increased L D facilitates gas recirculation in the main combustion zone, which is conducive to stable combustion. • The mixing uniformity of GPF increases with L D in general, promoting the uniformity of temperature distribution. [ABSTRACT FROM AUTHOR]

The flexibility of the coal-fired boiler depends on the control system, which is designed according to its dynamic characteristics. In this paper, a detailed dynamic model of a supercritical coal-fired boiler is developed in Dymola, and the necessary control models are built. The model is validated against the design values under various conditions with a maximum error of 2.19%. The model is further validated by comparing dynamic simulation results with operating data. Then, the dynamic behavior of the boiler is analyzed under the single- and two-parameter disturbances. When the feedwater temperature, feedwater flow, and coal quantity are all disturbed by −5%, the maximum deviations of the live steam temperature are −21 °C, +32 °C, and −34 °C, respectively. Compared to the simultaneous change of feedwater and fuel, the maximum value of the live steam temperature is reduced by 21.56 °C when the coal quantity is changed 100 s before the feedwater flow. Besides, a first-order inertia compensation element is developed ahead of the feedwater flow command, which effectively improves the performance of the water-fuel ratio control. As the load-variation rate increases, the deviation in the transient process of the thermodynamic parameters increases, and the time required for stabilization lengthens. • A detailed dynamic model of a supercritical coal-fired boiler is developed. • The model is validated for dynamic performance using measured plant data. • The response time and maximum deviation of main parameters are evaluated. • Adding a first-order inertia element before feedwater improves water-fuel control. [ABSTRACT FROM AUTHOR]

The flexible operation of coal-fired boilers deteriorates the heat transfer capacity of tubes, leading to localized overheating, tube cracking, and even boiler failure. This study presents a digital twin (DT) for detecting the metal temperature anomaly of a coal-fired boiler by predicting the metal temperature time series using data-driven methods. A data-based model adopting an extreme gradient boosting algorithm tuned by a particle swarm optimization algorithm is proposed to predict the metal temperature with an output sequence of 6 min. The predicted metal temperature is then transformed into an interval based on statistical distribution and confidence level to improve the robustness of the model. Moreover, a histogram is used to determine the benchmark of metal temperature anomaly with respect to varying operating conditions. The proposed DT is applied to a 600-MW on-duty supercritical coal-fired boiler. Results indicate that the proposed DT can accurately predict metal temperature and is useful for detecting metal temperature anomalies in advance. •PSO-XGBoost works well in predicting the tube temperature as a multi-step time-series. •The model prediction is expressed a temperature interval can avoid false or missing alarms of detected anomaly. •Proposed digital twin can accurately predict metal temperature and detect metal temperature anomalies 6 min in advance. [ABSTRACT FROM AUTHOR]

The development of industry has led to the big data revolution in manufacturing industry, which has greatly improved production quality and efficiency. A accuracy prediction model is established by using deep learning method, especially short and long term memory (LSTM). However, the application field of LSTM is usually very limited, and the interpretability of LSTM is weak. In this paper, a novel bidirectional DiPLS based LSTM algorithm (BiDiPLS-LSTM) is proposed to solve this problem. First, DiPLS is used to process the forward time series data to obtain the most predictable dynamic latent variable (DLV) of the target variable. At the same time, DiPLS method is also used to process the reverse time series data and extract the reverse DLV. Then, the forward and reverse dynamic features of the same location will form new input data, and this new input data will be input into the LSTM network to realize the prediction of time series data. In order to verify the effectiveness of the proposed BiDiPLS-LSTM method, the real 660 MW coal-fired boiler process data is used, and the emission of oxides of nitrogen (NO x) is regarded as the output. The results show that the extracted dynamic features can be used to build a very accuracy model, and the results outperform the traditional LSTM method, the PLS-LSTM method, the unidirectional DiPLS-LSTM method, and traditional RNN method in the comparison. For the first time, we used the bidirectional dynamic features of the input data as the real input to predict the output, which will obtain a good prediction effect. What is more, the use of dynamic latent variables made the result more explanatory. • The receive field of LSTM is usually very limited, and the interpretability is weak. • DiPLS can extract dynamic features between input and output. • DiPLS can improve the interpretability of LSTM extraction. • a novel bidirectional DiPLS based LSTM algorithm is proposed in this paper. • Application results confirm the effectiveness and reliability of the proposed method. [ABSTRACT FROM AUTHOR]

The coal-fired boiler (CFB) often suffers from corrosion, slagging, fouling, and clinkering; among them, clinkering is one of the critical problems, which is addressed in very few literature. Most of the existing research does not focus on clinkering prediction for 500 MW CFBs, and the features are selected based on the operators' experience. Hence, to overcome human errors and improve the safety and reliability of thermal power plants, we propose a data-driven-based predictive maintenance method, dClink , that uses automated feature selection for clinkering prediction from real-time process parameters of a 500 MW CFB. An automatic filter-wrapper-based feature selection model is proposed to select and rank the most important process parameters for clinkering. Root-cause analysis of clinkering is also performed so that the vital parameters can be controlled to prevent clinkering. The Support Vector Machine detected clinkering-fault with 99.8% accuracy. Further, the designed Decision Tree Regression model predicted clinkering with Root Mean Squared Error of 0.0103 and Mean Absolute Error of 0.0107 during the testing phase. The method has been validated using randomly selected data from the given dataset. It has been proved that the proposed method can detect and accurately predict clinkering to prevent accidents, low efficiency, and financial loss. • Architecture to efficiently detect and predict clinkering in 500 MW pulverized CFB. • Automatic selection of the features with the highest effect on clinkering. • Root cause analysis to effectively control and prevent the clinkering occurrence. • Decision tree-based prediction model has high accuracy with 1.033 RMSE. [ABSTRACT FROM AUTHOR]

Big data and time series are typical features of modern industrial process data. Effective time series modeling methods are required for ensuring the normal and stable operation of industrial processes. However, the existing time series prediction methods usually cannot take important historical data features long ago into consideration. In this paper, we proposed a new attention-mechanism based dynamic-inner partial least squares long short-term memory (ADiPLS-LSTM) method to predict the time series data. First, the DiPLS method is applied to extract the dynamic features in the selected data group, and the attention-mechanism is used then to define the importance of the responded features. The product of the extracted dynamic features and the attention-mechanism results is regarded as the input of LSTM to predict the future time information. The proposed method uses both the recent data information and important data information in a long run, which helps to obtain more accurate prediction results. The effectiveness of the ADiPLS-LSTM method is verified by the data of a 660MW coal-fired boiler. Compared with BPNN, LSTM, and DiPLS-LSTM, the proposed method also advances in time series data prediction and interpretability. • The receive field of LSTM is usually very limited, and the interpretability is weak • DiPLS can extract dynamic features between input and output • The existing prediction methods cannot take important historical data features. • A new novel ADiPLS-LSTM method is proposed in this paper. • Application results confirm the effectiveness and reliability of the proposed method. [ABSTRACT FROM AUTHOR]

Co-incineration of sludge in the existing coal-fired utility boilers is a promising sludge disposal approach, but study on the combustion characteristics and the economic impacts of co-firing sludge in full-scale utility boilers is in absence. In this study, a series of full-scale field experiments and numerical experiments were conducted to study the coal-sludge co-combustion characteristic in a 100 MW coal-fired utility boiler. The effects of sludge blending ratio and moisture content in sludge are concerned. Subsequently, the economic impacts of co-firing sludge is presented. The results demonstrate that the combustion characteristics is acceptable when co-firing 10% sludge. The ignition property and the flame stability are significantly affected when the mass content of co-fired sludge is over 10%. Decreasing the moisture content in sludge is beneficial to the combustion characteristics but the improvements are limited. Co-firing sludge is profitable for power plants as the savings from coal and sludge disposal allowance is quite substantial. Considering both the combustion characteristics and the economic impacts, co-firing 10% sludge with moisture content ranging from 40% to 56% is optimal. The annual profit from co-firing sludge can reach 4.2 million to 5.5 million Chinese yuan (CNY), which is remarkable for the power plant. [ABSTRACT FROM AUTHOR]

A numerical evaluation using mesh and models verified by comparison with experimental results was performed to investigate the effects of swirl configuration and fuel-rich environment on the flame phenomenon, uniformity of coal particles, and combustion characteristics in a 16-kW th coal-fired boiler. In addition, the suitability of the application of existing technologies to coal-fired power generation to reduce pollutant emissions and enhance the combustion efficiency was evaluated by calculating environmental costs based on the life cycle assessment method. The coal particles of the co-swirling flame were more uniformly dispersed than those of the counter-swirling flame because of the generation of the inner recirculation zone and exhaust tube vortex. Therefore, the burnout zone was evenly formed inside the boiler. Counter-swirling flames exhibit proportional results in terms of the increase or decrease in NO x emissions and unburned coal particles (UCPs) with swirl intensity. However, co-swirling flames have a synergistic effect that can reduce the NO x and UCPs emissions up to 102.4 ppm (@6%O 2) and 1.23 g/m3, respectively. Considering the external environmental costs for NO x , SO 2 , CO 2 , and UCPs emissions for all fuel-rich environments, the application of air staging technologies can reduce the environmental costs from $0.003 to $0.015 per day. • Numerical evaluation was performed in a coal-fired boiler using deNO x technologies. • Effects of swirl configuration on three major NO x emissions were investigated. • Co-swirl flame was effective in reducing NO x emissions and unburned coal particles. • Fuel-rich environment was applied to evaluate the combustion characteristics. • Air staging was beneficial for the environmental cost and operational goals. [ABSTRACT FROM AUTHOR]

[Display omitted] • Condensable particulate matter accounts for over 50% of the total PM emissions. • NH 4 +-rich water-soluble inorganic salts are major CPM constituents. • SO 2 and NO x control technologies affect CPM emissions significantly. • Excessive use of NH 3 in NO x control technologies increases the production of CPM. Different control technologies to reduce emissions of sulfur oxides and nitrogen oxides from coal-fired boilers have been adopted globally, leading to significant reductions in these gas emissions, especially over the past several years in China. Comparatively, little attention has been paid to the potential effects of these control technologies on particulate matter (PM) emissions, especially condensable particulate matter (CPM), which is harmful to the atmosphere and human health. In this study, we measured filterable and condensable PM emissions in five coal-fired commercial-scale boilers installed with SNCR or SCR for NO x control and wet, semidry, or dry flue gas desulfurization (FGD) for SO 2 emission controls. Water-soluble ionic components in PM were analyzed. The experimental results show that the CPM accounts for 55.9% to 95.1% of total particulate matter (TPM) and is composed primarily of water-soluble SO 4 2- and NH 4 +, with (NH4) 2 SO 4 being the dominant CPM aerosols. Experimental results also reveal that the flue gas emission control technology significantly affects CPM emission from coal-fired boilers. Since the production of (NH4) 2 SO 4 -rich CPM results from the use of ammonia solution for NO x reduction, boilers emit more CPM with SNCR than with SCR because the former operates with a higher ammonia/NO x ratio than the latter. Additionally, the CPM emission is related to the FGD technology installed. Boilers installed with wet FGD emit the lowest CPM, followed by semidry and dry FGDs, because of their decreased capabilities of dissolving and absorbing water-soluble components. [ABSTRACT FROM AUTHOR]

• Thermodynamic calculation model of a 600 MW boiler co-firing coal and NH 3 was established. • Co-firing NH 3 decreased furnace heat load but hardly changed flue gas temperature at furnace outlet. • Co-firing NH 3 raised the exhaust flue gas temperature and decreased boiler efficiency. • Heat absorption of steam can be satisfied with no modifications of heating surfaces. • Tube wall temperatures of superheaters were changed little by NH 3 co-firing. Ammonia (NH 3) co-firing provides a promising route to reducing CO 2 emissions of coal-fired boilers. The thermodynamic calculation model of a 600 MW supercritical coal-fired boiler was established to investigate the effects of NH 3 co-firing on the heat transfer, safety, and economy of the boiler. Co-firing 50 % of NH 3 (on the lower heat value basis) reduced the emissivity and temperature of the flame, and decreased the heat load of the furnace by 6.2 %-7.1 %. Nevertheless, the heat absorption of both the main steam and reheat steam can still be satisfied by adjusting the ratio of flue gas in two vertical flues without modifications of the heating surfaces. With the NH 3 ratio rising from 0 % to 50 %, the temperature of flue gas at the furnace outlet grew by only 7.0 °C. Besides, the tube wall temperatures of the platen superheater and the high temperature superheater increased by < 3.0 °C, ensuring the safety of the heating surfaces. At boiler rated load, low temperature corrosion of air pre-heaters was hardly aggravated by co-firing NH 3 of 50 %, but it became more severe at lower boiler loads. Compared with single coal combustion, the temperature of exhaust flue gas rose by 6.2–13.9 °C when NH 3 ratio was 50 %, and the boiler efficiency decreased by 0.7 %-1.2 %. The high price of NH 3 greatly increased the running cost of the boiler, but a higher CO 2 tax can highlight the economic benefits of NH 3 co-firing. [ABSTRACT FROM AUTHOR]

• Field measurements of sludge co-combustion on a 600 MW coal-fired boiler were conducted. • Three ash particles modes of PM 10 can clearly be observed. • The molten Fe Al-silicate and Fe-Ca-Si-Al mineral species were formed during co-combustion. • Sticky particles contributed to the partitioning of ultrafine particles into coarser particles. Co-combustion of sewage sludge and coal in coal-fired power plants may be a reasonable alternative for intractable sewage sludge. The co-combustion experiments were conducted on a 600 MW subcritical coal-fired boiler to discuss the fly ash partitioning and PM 10 emission characteristics. The sewage sludge was sent to the boiler through coal pulverizer with blending ratios of 5 and 10 wt% based on the coal mass. The results showed that the particle size distributions (PSDs) of PM 10 present the typical two-peak distribution. The mass yields of PM 1 dramatically decreased by 57.83 % and 61.67 % for the 5 % and 10 % blending ratios, respectively, compared with single coal combustion, corresponding to the increase in PM 1-10 emissions. The reduced PM 1 appears to be transferred toward PM 1-10 during co-combustion. The Fe Al-silicate and the Fe-Ca-Si-Al species were formed by the co-combustion of sludge and coal, which were derived from the active Fe-rich fine particles in sludge and mainly mullite and Ca-Si-Al found in coal ash. These transformations can be intensified by the high temperature, intense turbulent flow fields and the consequent frequent contact and collision of minerals in the large capacity boiler to form molten aluminosilicates. Therefore, the condensable Ca and S vapor from coal combustion and fine Si and Al adhere to the surface of molten Fe Al-silicate to form PM 1+ , which is the main reason for PM 1 reduction during co-combustion. In addition, Fe-rich and Ca-rich flocculants in sludge contribute to PM 1-10 largely, while some of the fine Al-Si particles are partitioned from PM 1 to PM 1-10 as captured by the molten aluminosilicates, both contributing to the increased PM 1-10 emission after co-combustion. [ABSTRACT FROM AUTHOR]

This paper focuses on modeling and reducing NO X emissions for a coal-fired boilers with advanced machine learning approaches. The novel ELM (extreme learning machine) model was introduced to model the correlation between operational parameters and NO X emissions of the boiler. Approximately ten days of real data from the SIS (supervisory information system) of a 700 MW coal-fired power plant were acquired to train and verify the ELM-based NO X model. Based on the NO X model, HS (harmony search) algorithm was then employed to optimize the operational parameters to finally realize NO X emission reduction. The modeling results indicated that the ELM model was more precise and faster in modeling NO X emissions than the popular artificial neural network and support vector regression. The searching process of HS was convergent and consumed only 0.7 s of CPU (Central Processing Unit) time on a personal computer. 16.5% and 19.3% NO X emission reductions for the two selected cases were achieved according to the simulation result. Additionally, the simulation result was experimentally justified, which demonstrated that the experimental results corresponded well with the computational: the experimental NOX reduction percentages were 14.8% and 15.7%, respectively. The proposed integrated method was capable of providing desired and feasible solutions within 1 s. [ABSTRACT FROM AUTHOR]

Fast learning network (FLN) is a novel double parallel forward neural network, which proves to be a very good machine learning tool. However, some randomly initialed weights and biases may be non-optimal performance parameters. Therefore, for the problem, this paper proposes a double linear fast learning network (DLFLN), in which all weights and biases are divided into two parts and each part is determined by least squared method. DLFLN is employed to model the combustion characteristics of a 330 MW coal-fired boiler and is combined with an optimization algorithm to tune the operating parameters of the boiler to achieve the combustion optimization objective. Experimental results show that, compared with extreme learning machine and FLN, although the DLFLN is assigned much less hidden neural nodes, the DLFLN could achieve much better generalization performance and stability under various operational conditions; in addition, the effect of the combustion optimization is very satisfactory. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) is widely used in complex industrial problems, especially the online-sequential extreme learning machine (OS-ELM) plays a good role in industrial online modeling. However, OS-ELM requires batch samples to be pre-trained to obtain initial weights, which may reduce the timeliness of samples. This paper proposes a novel model for the online process regression prediction, which is called the Recurrent Extreme Learning Machine (Recurrent-ELM). The nodes between the hidden layers are connected in Recurrent-ELM, thus the input of the hidden layer receives both the information from the current input layer and the previously hidden layer. Moreover, the weights and biases of the proposed model are generated by analysis rather than random. Six regression applications are used to verify the designed Recurrent-ELM, compared with extreme learning machine (ELM), fast learning network (FLN), online sequential extreme learning machine (OS-ELM), and an ensemble of online sequential extreme learning machine (EOS-ELM), the experimental results show that the Recurrent-ELM has better generalization and stability in several samples. In addition, to further test the performance of Recurrent-ELM, we employ it in the combustion modeling of a 330 MW coal-fired boiler compared with FLN, SVR and OS-ELM. The results show that Recurrent-ELM has better accuracy and generalization ability, and the theoretical model has some potential application value in practical application. [ABSTRACT FROM AUTHOR]

Reliability of a system varies under different operating conditions like temperature, humidity, dust, maintenance facilities, maintenance personnel skill, operator skill, etc., called “covariates.” Under poor operating environments the reliability decreases. Therefore, for precise estimation of system reliability, the influence of various covariates is included. This article aims to investigate the effect of five covariates on the reliability characteristics of a coal-fired boiler using the Weibull proportional hazard model, a semiparametric reliability model. The results from the case study indicate that there is a substantial decrease in reliability. A preventive maintenance program based on this calculated reliability will be safer to use. [ABSTRACT FROM PUBLISHER]

Copyright of Ecological Chemistry & Engineering S / Chemia i Inzynieria Ekologiczna S is the property of Sciendo and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Abstract: This paper presents a novel artificial neural network with a very fast learning speed, all of whose weights and biases are determined by the twice Least Square method, so it is called Least Square Fast Learning Network (LSFLN). In addition, there is another difference from conventional neural networks, which is that the output neurons of LSFLN not only receive the information from the hidden layer neurons, but also receive the external information itself directly from the input neurons. In order to test the validity of LSFLN, it is applied to 6 classical regression applications, and also employed to build the functional relation between the combustion efficiency and operating parameters of a 300WM coal-fired boiler. Experimental results show that, compared with other methods, LSFLN with very less hidden neurons could achieve much better regression precision and generalization ability at a much faster learning speed. [Copyright &y& Elsevier]

Highlights: [•] Off-design coal having a low ash fusion temperature caused excessive slagging. [•] Impacts of boiler operating settings on FEGT and slagging were studied. [•] Sootblowers were characterized and optimized for mitigating slagging. [•] A strategy was developed for slowing the deposition rate down. [ABSTRACT FROM AUTHOR]

• A new multi-purpose system design powered by waste heat for H 2 production was conducted. • Optimal operating conditions of high-temperature electrolyzer were determined. • An energy and exergy efficiency increase of about 2% in the boiler was obtained. • An energy efficiency increase between 15.78 and 16.53% in the whole system was obtained. • An exergy efficiency increase between 20.43 and 21.16% in the whole system was obtained. A multi-generation system including power generation, domestic hot water, and H 2 production was designed for the waste recovery of a 150 MW coal-fired power plant. A high-temperature electrolyzer was integrated into the present system to produce H 2. The second product of O 2 from the electrolyzer was used to enrich the combustion process. The required power for the electrolyzer was obtained from a bottoming organic Rankine cycle (ORC). In this regard, a supercritical ORC was designed in which cyclopentane was used as the working fluid. The required heat of the electrolyzer was obtained by a heat exchanger located in the combustion chamber to reach the required temperature level. The designed system was analyzed by the energy and exergy method for different temperatures and current densities. It is determined that it is available to produce H 2 in an amount of 0.0739 to 0.0762 kmol/s for the optimal current densities at every single handled temperature. The energy efficiency of the overall system was determined in the range of 27.19% and 27.36%. The exergy efficiency of the overall system was determined in the range of 35.05% and 35.25%. An increase of 15.78–16.53% in energy efficiency was achieved. This increase was achieved as 20.43–21.16% in the exergy efficiency. [ABSTRACT FROM AUTHOR]

Abstract: The teaching–learning-based optimization (TLBO) is a new efficient optimization algorithm. To improve the solution quality and to quicken the convergence speed and running time of TLBO, this paper proposes an ameliorated TLBO called A-TLBO and test it by classical numerical function optimizations. Compared with other several optimization methods, A-TLBO shows better search performance. In addition, the A-TLBO is adopted to adjust the hyper-parameters of least squares support vector machine (LS-SVM) in order to build NOx emissions model of a 330MW coal-fired boiler and obtain a well-generalized model. Experimental results show that the tuned LS-SVM model by A-TLBO has well regression precision and generalization ability. [Copyright &y& Elsevier]

Abstract: This paper focuses on NO x emission prediction and operating parameters optimization for coal-fired boilers. Support Vector Regression (SVR) model based on CGA (Conventional Genetic Algorithm) was proposed to model the relationship between the operating parameters and the concentration of NO x emission. Then CGA and two modified algorithms, the Quantum Genetic Algorithm (QGA) and SAGA (Simulated Annealing Genetic Algorithm), were employed to optimize the operating parameters of the coal-fired boiler to reduce NO x emission. The results showed that the proposed SVR model was more accurate than the widely used Artificial Neural Network (ANN) model when employed to predict the concentration of NO x emission. The mean relative error and correlation coefficient calculated by the proposed SVR model were 2.08% and 0.95, respectively. Among the three optimization algorithms implemented in this paper, the SAGA showed superiority to the other two algorithms considering the quality of solution within a given computing time. The SVR plus SAGA method was preferable to predict the concentration of NO x emission and further to optimize the operating parameters to achieve low NO x emission for coal-fired boilers. [Copyright &y& Elsevier]

Abstract: Real operation data of power plants are inclined to be concentrated in some local areas because of the operators’ habits and control system design. In this paper, a novel least squares support vector machine (LSSVM)-based ensemble learning paradigm is proposed to predict NO x emission of a coal-fired boiler using real operation data. In view of the plant data characteristics, a soft fuzzy c-means cluster algorithm is proposed to decompose the original data and guarantee the diversity of individual learners. Subsequently the base LSSVM is trained in each individual subset to solve the subtask. Finally, partial least squares (PLS) is applied as the combination strategy to eliminate the collinear and redundant information of the base learners. Considering that the fuzzy membership also has an effect on the ensemble output, the membership degree is added as one of the variables of the combiner. The single LSSVM and other ensemble models using different decomposition and combination strategies are also established to make a comparison. The result shows that the new soft FCM-LSSVM-PLS ensemble method can predict NO x emission accurately. Besides, because of the divide and conquer frame, the total time consumed in the searching the parameters and training also decreases evidently. [Copyright &y& Elsevier]

Abstract: In this paper, we propose a new combination modeling method whose structure consists of three components: extreme learning machine (ELM), adaptive neuro-fuzzy inference system (ANFIS) and PS-ABC which is a modified hybrid artificial bee colony algorithm. The combination modeling method has been proposed in an attempt to obtain good approximations and generalization performances. In the whole model, ELM is used to build a global model, and ANFIS is applied to compensate the output errors of ELM model to improve the overall performance. In order to obtain a better generalization ability and stability model, PS-ABC is adopted to optimize input weights and biases of ELM. For stating the proposed model validity, it is applied to set up the mapping relation between the boiler efficiency and operational conditions of a 300WM coal-fired boiler. Compared with other combination models, the proposed model shows better approximations and generalization performances. [Copyright &y& Elsevier]