Search

With recent industrial upgrades, it is essential to transform the current forward supply networks (FSNs) into closed-loop supply networks (CLSNs), which are formed by the integration of forward and reverse logistics. The method chosen in this paper for building reverse logistics is to add additional functions to the existing forward logistics. This process can be regarded as adding reverse edges to the original directed edges in an FSN. Due to the limitation of funds and the demand for reverse flow, we suppose that a limited number of reverse edges can be built in a CLSN. To determine the transformation schemes with excellent robustness against malicious attacks, this paper proposes a multi-population evolutionary algorithm with novel operators to optimize the robustness of the CLSN, and this algorithm is abbreviated as MPEA-RSN. Then, both the generated and realistic SNs are taken as examples to validate the effectiveness of MPEA-RSN. The simulation results show that the index R, introduced to evaluate the robustness of CLSNs, can be improved by more than 95%, and this indicates that (1) the different schemes for adding reverse routes to an FSN can lead to different robustness values, and (2) the robustness of the transformed CLSN to malicious attacks can be significantly improved after optimization by MPEA-RSN. When an FSN is to be transformed into a CLSN, this paper can provide a frame of reference for building a CLSN that is robust to malicious attacks from a network structural perspective.

Non-traditional machining (NTM) has gained significant attention in the last decade due to its ability to machine conventionally hard-to-machine materials. However, NTMs suffer from several disadvantages such as higher initial cost, lower material removal rate, more power consumption, etc. NTMs involve several process parameters, the appropriate tweaking of which is necessary to obtain economical and suitable results. However, the costly and time-consuming nature of the NTMs makes it a tedious and expensive task to manually investigate the appropriate process parameters. The NTM process parameters and responses are often not linearly related and thus, conventional statistical tools might not be enough to derive functional knowledge. Thus, in this paper, three popular machine learning (ML) methods (viz. linear regression, random forest regression and AdaBoost regression) are employed to develop predictive models for NTM processes. By considering two high-fidelity datasets from the literature on electro-discharge machining and wire electro-discharge machining, case studies are shown in the paper for the effectiveness of the ML methods. Linear regression is observed to be insufficient in accurately mapping the complex relationship between the process parameters and responses. Both random forest regression and AdaBoost regression are found to be suitable for predictive modelling of NTMs. However, AdaBoost regression is recommended as it is found to be insensitive to the number of regressors and thus is more readily deployable. Web of Science 9 11 art. no. 2015

This paper studies the production process of a shipbuilding enterprise. The company suffers from long manufacturing cycle, low utilization rate of personnel and an unbalanced production line. To solve these problems, the lean shipbuilding mode, mainly divided into shipbuilding work breakdown, production plan and virtual flow operation in this paper, is put forward, which combines the lean production and modern information management technology with shipbuilding. Supported by the theory of work breakdown structure and task package scheduling, the shipbuilding task package is reasonably divided. The priority of task package manufacturing is determined by calculating the task package manufacturing sequence coefficient, and a reasonable number of operators is calculated to ensure the continuity of segmented manufacturing. After determining the manufacturing priority of the task pack and the number of allocable personnel, the corresponding work can be scheduled. Production planning drives all production activities of the shipbuilding enterprise, and just-in-time production is achieved through the reasonable arrangement of these production plans, thus reducing the waste of personnel and time. Then, the virtual flow operation is carried out, which can achieve high efficiency of flow production and high flexibility of fixed workstation production during the production process of large-scale and heavy-duty products. The virtual assembly production system of the workshop is established according to the characteristics of shipbuilding operation and the actual production situation. On this basis, a lean shipbuilding manufacturing execution system for small and medium-sized shipbuilding enterprises is developed to achieve lean production in a shipbuilding workshop. Through the implementation of the lean shipbuilding mode based on task package scheduling and its manufacturing execution system, compared with the original data, the ship production cycle is reduced to 76.7%, the number of workers is reduced by 16.7% and the production balance rate is up to 81%.

As China proposes to achieve carbon peak by 2030 and carbon neutrality by 2060, as well as the huge pressure on the power grid caused by the load demand of the energy supply stations of electric vehicles (EVs), there is an urgent need to carry out comprehensive energy management and coordinated control for EVs’ energy supply stations. Therefore, this paper proposed a two-step intelligent control method known as ISOM-SAIA to solve the problem of the 24 h control and regulation of green/flexible EV energy supply stations, including four subsystems such as a photovoltaic subsystem, an energy storage subsystem, an EV charging subsystem and an EV battery changing subsystem. The proposed control method has two main innovations and contributions. One is that it reduces the computational burden by dividing the multi-dimensional mixed-integer programming problem of simultaneously optimizing the 24 h operation modes and outputs of four subsystems into two sequential tasks: the classification of data-driven operation modes and the rolling optimization of operational outputs. The other is that proper carbon transaction costs and carbon emission constraints are considered to help save costs and reduce carbon emissions. The simulation analysis conducted in this paper indicates that the proposed two-step intelligent control method can help green/flexible EV energy supply stations to optimally allocate energy flows between four subsystems, effectively respond to peak shaving and valley filling of power grid, save energy costs and reduce carbon emissions.

The advancements in human lifestyle result in growth in daily demands of products, and accordingly, an increased rate of manufacturing. However, the resources on the planet Earth are limited, thus depleting day-by-day. More goods also contribute to more end-of-life (EOL) dumping or even before EOL in some cases. Therefore, an interest in remanufacturing has appeared, and it offers a solution that can solve or perhaps mitigate the risks of consuming more resources and increasing waste. Remanufacturing is a procedure of bringing used products to “like-new” functional status with a matching warranty. However, due to its relative novelty in terms of research field and industry, remanufacturing is poorly understood. People often mix it with other terms such as recycling, reconditioning, or repair. Therefore, in this research, the focus is on the remanufacturing systems’ definition, relevance, main phases, case studies, and solution methods proposed by various researchers. The word ‘remanufacturing’ is clearly described in this paper by differentiating it from alternative green manufacturing initiatives. Both qualitative and quantitative analysis of literature are performed. The quantitative analysis is conducted using a bibliometric method. For quantitative analysis, a systematic approach is utilized for research papers’ selection. The qualitative analysis has been carried out by discussing different aspects of remanufacturing and how the researchers are working on its different domains and phases. The review showed that researchers focused on some phases more as compared with others. Moreover, it is also revealed from the literature that the common solutions methods applied in this domain are optimization techniques. Future research directions are also identified and presented.

In this paper, a control strategy combining the feedback linearization theory and sliding mode variable structure theory is proposed to solve various nonlinear factors, uncertainty of external disturbance and high-precision pressure control problems in the Direct-Drive Volume Control (DDVC) electro-hydraulic servo system. The nonlinear mathematical model of the DDVC electro-hydraulic servo system is established, and the nonlinear factors in the system are accurately linearized by the feedback linearization theory. The uncertainty of external disturbance in the system is compensated by the sliding mode control variable structure theory. The feedback-linearized sliding mode control algorithm proposed in this paper is verified using the DDVC electro-hydraulic servo system experimental platform. The experimental results show that, compared with the classical PID control, the proposed control algorithm can effectively improve the pressure output precision, as well as the dynamic response characteristics, of the DDVC system.

In order to attain high manufacturing productivity, industry 4.0 merges all the available system and environment data that can empower the enabled-intelligent techniques. The use of data provokes the manufacturing self-awareness, reconfiguring the traditional manufacturing challenges. The current piece of research renders attention to new consideration in the Job Shop Scheduling (JSSP) based problems as a case study. In that field, a great number of previous research papers provided optimization solutions for JSSP, relying on heuristics based algorithms. The current study investigates the main elements of such algorithms to provide a concise anatomy and a review on the previous research papers. Going through the study, a new optimization scope is introduced relying on additional available data of a machine, by which the Flexible Job-Shop Scheduling Problem (FJSP) is converted to a dynamic machine state assignation problem. Deploying two-stages, the study utilizes a combination of discrete Particle Swarm Optimization (PSO) and a selection based algorithm followed by a modified local search algorithm to attain an optimized case solution. The selection based algorithm is imported to beat the ever-growing randomness combined with the increasing number of data-types.

The main contribution of the paper is the simplification of the computational process of fuzzy control of a mobile robot controlled by a microcontroller. We present a way to implement this control method with a reduced computation time of control actions and memory demand. Our way to accomplish this, was to replace the fuzzy controller with the approximation of its resulting control surfaces. In the paper, we use the previously presented approximation by the table and describe other methods of approximation of the control area through polynomial and exponential function. We tested all approximation methods in simulations and with a real mobile robot. Based on the measured trajectory of the EN20 mobile robot, we found that approximation through the table is the most accurate in terms of the fuzzy surface but delivers noticeable oscillations of mobile robot control in real conditions. Polynomial and exponential functions fuzzy surface approximations were less accurate than the table, but provide smoother control based on robot trajectories and are much more appropriate in terms of microcontroller implementation due to lower demand on memory.

Tolerance analysis is crucial in every manufacturing process, such as electrical machine design, because tight tolerances lead to high manufacturing costs. A FEM-based solution of the tolerance analysis of an electrical machine can easily lead to a computationally expensive problem. Many papers have proposed the design of experiments, surrogate-model-based methodologies, to reduce the computational demand of this problem. However, these papers did not focus on the information loss and the limitations of the applied methodologies. Regardless, the absolute value of the calculated tolerance and the numerical error of the applied numerical methods can be in the same order of magnitude. In this paper, the tolerance and the sensitivity of BLDC machines’ cogging torque are analysed using different methodologies. The results show that the manufacturing tolerances can have a significant effect on the calculated parameters, and that the mean value of the calculated cogging torque increases. The design of the experiment-based methodologies significantly reduced the calculation time, and shows that the encapsulated FEM model can be invoked from an external system-level optimization to examine the design from different aspects.

A reduced switching components step-up multilevel inverter (RSCS-MLI) is presented in the paper. The basic circuit of the proposed MLI can produce 11 levels in the output voltage with a reduced number of switching components. The other features of the proposed circuit include a low value of voltage stresses and the inherent generation of the voltage levels pertaining to the negative half without the requirement of an H-bridge. Fundamental frequency switching technique, also known as Nearest Level Control (NLC) technique, is implemented in the proposed topology for generating the switching signals. The experimental total harmonic distortion (THD) in the output voltage comes out to be 9.4% for modulation index equal to 1. Based on different parameters, a comparative study has been shown in the paper, which makes the claim of the proposed MLI stronger. An experimental setup is prepared to carry out the hardware implementation of the proposed structure and monitor its performance under dynamic load conditions, which is also used to verify the simulation results. Power loss analysis, carried out by using PLECS software, helps us to gain insight into different losses occurring while operating the inverter. The different results are explained and analyzed in the paper.

This paper is concerned with the real-time optimization (RTO) of chemical plants, i.e., the optimization of the steady-state operating points during operation, based on inaccurate models. Specifically, modifier adaptation is employed to cope with the plant-model mismatch, which corrects the plant model and the constraint functions by bias and gradient correction terms that are computed from measured variables at the steady-states of the plant. This implies that the sampling time of the iterative RTO scheme is lower-bounded by the time to reach a new steady-state after the previously computed inputs were applied. If analytical process measurements (PAT technology) are used to obtain the steady-state responses, time delays occur due to the measurement delay of the PAT device and due to the transportation delay if the samples are transported to the instrument via pipes. This situation is quite common because the PAT devices can often only be installed at a certain distance from the measurement location. The presence of these time delays slows down the iterative real-time optimization, as the time from the application of a new set of inputs to receiving the steady-state information increases further. In this paper, a proactive perturbation scheme is proposed to efficiently utilize the idle time by intelligently scheduling the process inputs taking into account the time delays to obtain the steady-state process measurements. The performance of the proposed proactive perturbation scheme is demonstrated for two examples, the Williams–Otto reactor benchmark and a lithiation process. The simulation results show that the proposed proactive perturbation scheme can speed up the convergence to the true plant optimum significantly.

Integrated energy services will have multiple values and far-reaching significance in promoting energy transformation and serving “carbon peak and carbon neutralization”. In order to balance the supply and demand of power system in integrated energy, it is necessary to establish a scientific model for power load forecasting. Different algorithms for short-term electric load forecasting considering meteorological factors are presented in this paper. The correlation between electric load and meteorological factors is first analyzed. After the principal component analysis (PCA) of meteorological factors and autocorrelation analysis of the electric load, the daily load forecasting model is established by optimal support vector machine (OPT-SVM), Elman neural network (ENN), as well as their combinations through linear weighted average, geometric weighted average, and harmonic weighted average method, respectively. Based on the actual data of an industrial park of Nantong in China, the prediction performance in the four seasons with the different models is evaluated. The main contribution of this paper is to compare the effectiveness of different models for short-term electric load forecasting and to give a guideline to build the proper methods for load forecasting.

Traditional intelligent fault diagnosis methods focus on distinguishing different fault modes, but ignore the deterioration of fault severity. This paper proposes a new two-stage hierarchical convolutional neural network for fault diagnosis of rotating machinery bearings. The failure mode and failure severity are modeled as a hierarchical structure. First, the original vibration signal is transformed into an energy spectrum matrix containing fault-related information through wavelet packet decomposition. Secondly, in the model training method, an adaptive learning rate dynamic adjustment strategy is further proposed, which adaptively extracts robust features from the spectrum matrix for fault mode and severity diagnosis. To verify the effectiveness of the method, the bearing fault data was collected using a rotating machine test bench. On this basis, the diagnostic accuracy, convergence performance and robustness of the model under different signal-to-noise ratios and variable load environments are evaluated, and the feature learning ability of the method is verified by visual analysis. Experimental results show that this method has achieved satisfactory results in both fault pattern recognition and fault severity evaluation, and is superior to other machine learning and deep learning methods.

Low carbon manufacturing has received increasingly more attention in the context of global warming. The flexible job shop scheduling problem (FJSP) widely exists in various manufacturing processes. Researchers have always emphasized manufacturing efficiency and economic benefits while ignoring environmental impacts. In this paper, considering carbon emissions, a multi-objective flexible job shop scheduling problem (MO-FJSP) mathematical model with minimum completion time, carbon emission, and machine load is established. To solve this problem, we study six variants of the non-dominated sorting genetic algorithm-III (NSGA-III). We find that some variants have better search capability in the MO-FJSP decision space. When the solution set is close to the Pareto frontier, the development ability of the NSGA-III variant in the decision space shows a difference. According to the research, we combine Pareto dominance with indicator-based thought. By utilizing three existing crossover operators, a modified NSGA-III (co-evolutionary NSGA-III (NSGA-III-COE) incorporated with the multi-group co-evolution and the natural selection is proposed. By comparing with three NSGA-III variants and five multi-objective evolutionary algorithms (MOEAs) on 27 well-known FJSP benchmark instances, it is found that the NSGA-III-COE greatly improves the speed of convergence and the ability to jump out of local optimum while maintaining the diversity of the population. From the experimental results, it can be concluded that the NSGA-III-COE has significant advantages in solving the low carbon MO-FJSP.

In this paper, the deterministic ensemble Kalman filter is implemented with a parallel technique of the message passing interface based on our in-house black oil simulator. The implementation is separated into two cases: (1) the ensemble size is greater than the processor number and (2) the ensemble size is smaller than or equal to the processor number. Numerical experiments for estimations of three-phase relative permeabilities represented by power-law models with both known endpoints and unknown endpoints are presented. It is shown that with known endpoints, good estimations can be obtained. With unknown endpoints, good estimations can still be obtained using more observations and a larger ensemble size. Computational time is reported to show that the run time is greatly reduced with more CPU cores. The MPI speedup is over 70% for a small ensemble size and 77% for a large ensemble size with up to 640 CPU cores.

This paper studies the distributed secondary control of DC microgrids (MGs) in the case of asynchronous sampling, including both the stability condition and accurate consensus algorithm. The asynchrony means that the update actions of each distributed generation (DG) based on the local information and information received from neighbors are independent of the actions of others at sampled discrete times, which would cause deviation from the accurate convergence and even lead to instability in the worst case. First, a small-signal model of MG installed with secondary voltage control is established to include the individual sampling periods. A stability criterion based on the periodic continuity of sampling instant offset is thus formulated to reveal a stability mapping of multiple sampling. By quantifying the accuracy deviations caused by the asynchrony, an improved ratio consensus strategy is proposed that allows the deviation to be estimated accurately via an auxiliary signal and compensated with respect to the eventual equilibrium to produce an exact solution. Our approach customizes the stability and accuracy for distributed secondary control considering asynchronous sampling in MG, which has been ignored in most existing literature. The effectiveness of the proposed methodology is verified by simulations.

In this paper, the Nonlinear Auto-Regressive with exogenous inputs (NARX) model with parameters of interest for design (NARX-M-for-D), where the design parameter of the system is connected to the coefficients of the NARX model by a predefined polynomial function is studied. For the NARX-M-for-D of nonlinear systems, in practice, to predict the output by design parameter values are often difficult due to the uncertain relationship between the design parameter and the coefficients of the NARX model. To solve this issue and conduct the analysis and design, an improved algorithm, defined as the Weighted Extended Forward Orthogonal Regression (WEFOR), is proposed. Firstly, the initial NARX-M-for-D is obtained through the traditional Extended Forward Orthogonal Regression (EFOR) algorithm. Then a weight matrix is introduced to modify the polynomial functions with respect to the design parameter, and then an improved model, which is referred to as the final NARX-M-for-D is established. The genetic algorithm (GA) is used for deriving the weight matrix by minimizing the normalized mean square error (NMSE) over the data sets corresponding to the design parameter values used for modeling and first prediction. Finally, both the numerical and experimental studies are conducted to demonstrate the application of the WEFOR algorithm. The results indicate that the final NARX-M-for-D can accurately predict the system output of a nonlinear system. The new algorithm is expected to provide a reliable model for dynamic analysis and design of the nonlinear system.

A car side-outer is an iron mold that is applied in the design and safety of the side of a vehicle, and is subjected to a complicated and detailed molding process. The side-outer has three features that make its quality inspection difficult to automate: (1) it is large, (2) there are many objects to inspect, and (3) it must fulfil high-quality requirements. Given these characteristics, the industrial vision system for the side-outer is nearly impossible to apply, and indeed there is no reference for an automated defect-inspection system for the side-outer. Manual inspection of the side-outer worsens the quality and cost competitiveness of the metal-cutting companies. To address these problems, we propose a large-scale Object-Defect Inspection System based on Regional Convolutional Neural Network (R-CNN, RODIS) using Artificial Intelligence (AI) technology. In this paper, we introduce the framework, including the hardware composition and the inspection method of RODIS. We mainly focus on creating the proper dataset on-site, which should be prepared for data analysis and model development. Additionally, we share the trial-and-error experiences gained from the actual installation of RODIS on-site. We explored and compared various R-CNN backbone networks for object detection using actual data provided by a laser-cutting company. The Mask R-CNN models using Res-net-50-FPN show Average Precision (AP) of 71.63 (Object Detection) and 86.21 (Object Seg-mentation), which indicates a better performance than that of other models.

The prevention of cascading failures and large-scale power outages of power grids by identifying weak links has become one of the key topics in power systems research. In this paper, a vulnerability radius index is proposed to identify the initial fault, and a fault chain model of cascading failure is developed with probabilistic attributes to identify the set of fault chains that have a significant impact on the safe and stable operation of power grids. On this basis, a method for evaluating the vulnerability of transmission lines based on a multi-criteria decision analysis is proposed, which can quickly identify critical transmission lines in the process of cascading failure. Finally, the proposed model and method for identifying vulnerable lines during the cascading failure process is demonstrated on the IEEE-118 bus system.

In the development of high-tech industries, graphite has become increasingly more important. The world has gradually entered the graphite era from the silicon era. In order to make good use of high-quality graphite resources, a graphite classification and recognition algorithm based on an improved convolution neural network is proposed in this paper. Based on the self-built initial data set, the offline expansion and online enhancement of the data set can effectively expand the data set and reduce the risk of deep convolution neural network overfitting. Based on the visual geometry group 16 (VGG16), residual net 34 (ResNet34), and mobile net Vision 2 (MobileNet V2), a new output module is redesigned and loaded into the full connection layer. The improved migration network enhances the generalization ability and robustness of the model, moreover, combined with the focal loss function, the superparameters of the model are modified and trained on the basis of the graphite data set. The simulation results illustrate that the recognition accuracy of the proposed method is significantly improved, the convergence speed is accelerated, and the model is more stable, which proves the feasibility and effectiveness of the proposed method.

In bakery production, to perform a processing task there might be multiple alternative machines that have the same functionalities. Finding an efficient production schedule is challenging due to the significant nondeterministic polynomial time (NP)‐hardness of the problem when the number of products, processing tasks, and alternative machines are higher. In addition, many tasks are performed manually as small and medium‐size bakeries are not fully automated. Therefore, along with machines, the integration of employees in production planning is essential. This paper presents a hybrid no‐wait flowshop scheduling model (NWFSSM) comprising the constraints of common practice in bakeries. The schedule of an existing production line is simulated to examine the model and is optimized by performing particle swarm optimization (PSO), modified particle swarm optimization (MPSO), simulated annealing (SA), and Nawaz‐Enscore‐Ham (NEH) algorithms. The computational results reveal that the performance of PSO is significantly influenced by the weight distribution of exploration and exploitation in a run time. Due to the modification to the acceleration parameter, MPSO outperforms PSO, SA, and NEH in respect to effectively finding an optimized schedule. The best solution to the real case problem obtained by MPSO shows a reduction of the total idle time (TIDT) of the machines by 12% and makespan by 30%. The result of the optimized schedule indicates that for small‐ and medium‐sized bakery industries, the application of the hybrid NWFSSM along with nature‐inspired optimization algorithms can be a powerful tool to make the production system efficient., This research has been funded by the EIT Food of the European Institute of Innovation and Technology (EIT), a body of the European Union, the EU Framework Programme for Research and Innovation for the project entitled “Optimization of bakery processes by a computational tool together with consumer feedback to minimize ecological footprint and food waste: Making baking more efficient”.

This paper presents the development of a new Aerodynamic Ball Levitation Laboratory Plant at the Center of Modern Control Techniques and Industrial Informatics (CMCT&amp, II). The entire design process of the plant is described, including the component selection process, the physical construction of the plant, the design of a printed circuit board (PCB) powered by a microcontroller, and the implementation of its firmware. A parametric mathematical model of the laboratory plant is created, whose parameters are then estimated using a nonlinear least-squares method based on acquired experimental data. The Kalman filter and the optimal state-space feedback control are designed based on the obtained mathematical model. The designed controller is then validated using the physical plant.

Recently, with the large scale of power grids and the increase in frequency of extreme weather, the safe and stable operation of power systems is facing great challenges. Therefore, mobile emergency power source (MEPS) are a promising and feasible way to deal with extreme weather and reduce economic losses. However, the current urban power grid and transportation network are closely coupled, and the congested traffic hinders the rapid configuration of MEPSs. Therefore, this paper proposes an MEPS configuration scheme considering real-time traffic conditions. Firstly, the dynamic road traffic index (DRTI) is defined, which can fully describe the dynamic characteristics of traffic. The wavelet neural network (WNN) is used to predict the traffic flow. Then, combined with the knowledge of graph theory, an A-Star algorithm (AS) is used to determine the optimal path. Secondly, the optimal installation location of MEPSs is determined by forward–backward sweep method in distribution network. Finally, the feasibility, accuracy, and time cost of the proposed method are verified by numerical simulations, which can meet the requirements of online application.

By virtue of the steady societal shift to the use of smart technologies built on the increasingly popular smart grid framework, we have noticed an increase in the need to analyze household electricity consumption at the individual level. In order to work efficiently, these technologies rely on load forecasting to optimize operations that are related to energy consumption (such as household appliance scheduling). This paper proposes a novel load forecasting method that utilizes a clustering step prior to the forecasting step to group together days that exhibit similar energy consumption patterns. Following that, we attempt to classify new days into pre-generated clusters by making use of the available context information (day of the week, month, predicted weather). Finally, using available historical data (with regard to energy consumption) alongside meteorological and temporal variables, we train a CNN-LSTM model on a per-cluster basis that specializes in forecasting based on the energy profiles present within each cluster. This method leads to improvements in forecasting performance (upwards of a 10% increase in mean absolute percentage error scores) and provides us with the added benefit of being able to easily highlight and extract information that allows us to identify which external variables have an effect on the energy consumption of any individual household.

The optimal solution of the chemical dynamic optimization problem is the basis of automatic control operation in the chemical process, which can reduce energy consumption, increase production efficiency, and maximize economic benefit. In this paper, a modified sailfish optimizer (MSFO) combined with an equal division method is proposed for solving chemical dynamic optimization problems. Based on the basic sailfish optimizer, firstly, the tent chaotic mapping strategy is introduced to disturb the initialization of sailfish and sardine populations to avoid the loss of population diversity. Secondly, an adaptive linear reduction strategy of attack parameters is proposed to enhance the exploration and exploitation ability of sailfish. Thirdly, the updating formula of sardine position is modified, and the global optimal solution is used to attract all sardine positions, which can avoid the premature phenomenon of the algorithm. Eventually, the MSFO is applied to solve six classical optimization cases of chemical engineering to evaluate its feasibility. The experimental results are analyzed and compared with other optimization methods to prove the superiority of the MSFO in solving chemical dynamic optimization problems.

The application of Computational Fluid Dynamics (CFD) to energy-related problems has increased in the last decades in both renewable and conventional energy conversion processes. In recent years, the application of CFD in the study of hydraulic, marine, tidal, and hydrokinetic turbines has focused on the understanding of the details of the complex turbulent flow and also in improving the prediction of the performance of these devices. There are several complexities involved in the simulation of Vertical Axis Turbine (VAT) for hydrokinetic applications. One of them is the necessity of a dynamic mesh model. Typically, the model used in the simulation of these devices is the sliding mesh technique, but in recent years the fast development of the overset (also known as chimera) mesh technique has caught the attention of the academic community. In the present paper, a comparison between these two techniques is done in order to establish their advantages and disadvantages in the two-dimensional simulation of vertical axis turbines. The comparison was done not only for the prediction of performance parameters of the turbine but also for the capabilities of the models to capture complex flow phenomena in these devices and computational costs.

The supercritical carbon dioxide (SCO2) Brayton cycle, as a substitute for the steam cycle, can be widely used in a variety of power generation scenarios. However, most of the existing SCO2 cycle studies are restricted to basic thermodynamics research, parameter optimizations, system design in different application fields, and even economic analysis. Considering the load variability and control flexibility of the power generation system, the dynamic performance research of the SCO2 cycle is also crucial, but the work done is still limited. Based on the previous studies, Simulink software is used in this paper to develop a dynamic model of the 20 MW-SCO2 recompression cycle, which specifically includes component models that can independently realize physical functions and an overall closed-loop cycle model. A series of comparative calculation are carried out to verify the models and the results are very positive. The SCO2 recompression power system is built with the developed models and the dynamic model runs stably with a maximum error of 0.56%. Finally, the simulation of the dynamic switching conditions of the 20 MW-SCO2 recompression cycle are performed and the analysis results supply instructive suggestions for the system operation and control.

The reduction of maintenance costs in agricultural wireless sensor networks (WSNs) requires reducing energy consumption. At the same time, care should be taken not to affect communication quality and network lifetime. This paper studies a joint optimization algorithm for transmitted power and channel allocation based on deep reinforcement learning. First, an optimization model to measure network reward was established under the constraint of the signal-to-interference plus-noise-ratio (SINR) threshold, which includes continuous power variables and discrete channel variables. Secondly, considering the dynamic changes of agricultural WSNs, the network control is described as a Markov decision process with continuous state and action space. A deep deterministic policy gradient (DDPG) reinforcement learning scheme suitable for mixed variables was designed. This method could obtain a control scheme that maximizes network reward by means of black-box optimization for continuous transmitted power and discrete channel allocation. Experimental results indicated that the studied algorithm has stable convergence. Compared with traditional protocols, it can better control the transmitted power and allocate channels. The joint power and channel optimization provides a reference solution for constructing an energy-balanced network.

A nonlinear offset-free model predictive control based on a dynamic partial least square (PLS) framework is proposed in this paper. A multi-output multi-input system is projected into latent variable space by a PLS outer model. For each latent variable model, the T–S fuzzy model is used to describe the nonlinear characteristics of the system, while the state-space model is used in T–S fuzzy model consequent parameters to describe the dynamic characteristics. A disturbance model is introduced in the state-space model. For model state variables, a state observer is used to compensate for the mismatch of the model. The case study results for the pH neutralization process show that the MPC controller based on this method can guarantee the tracking performance of the nonlinear system without static error.

This paper presents a novel feature extraction and validation technique for data-driven prediction of oxy-fuel combustion emissions in a bubbling fluidized bed experimental facility. The experimental data were analyzed and preprocessed to minimize the size of the data set while preserving patterns and variance and to find an optimal configuration of the feature vector. The Boruta Feature Selection Algorithm (BFSA) finds feature vector’s configuration and the Multiscale False Neighbours Analysis (MSFNA) is newly extended and proposed to validate the BFSA’s design for emission prediction to assure minimal uncertainty in mapping between feature vectors and corresponding outputs. The finding is that the standalone BFSA does not reflect various sampling period setups that appeared significantly influencing the false neighborhood in the design of feature vectors for possible emission prediction, and MSFNA resolves that.

In this paper, we formulate and solve a novel real-life large-scale automotive parts paint shop scheduling problem, which contains color arrangement restrictions, part arrangement restrictions, bracket restrictions, and multi-objectives. Based on these restrictions, we construct exact constraints and two objective functions to form a large-scale multi-objective mixed-integer linear programming problem. To reduce this scheduling problem’s complexity, we converted the multi-objective model into a multi-level objective programming problem by combining the rule-based scheduling algorithm and the adaptive Partheno-Genetic algorithm. The rule-based scheduling algorithm is adopted to optimize color changes horizontally and bracket replacements vertically. The adaptive Partheno-Genetic algorithm is designed to optimize production based on the rule-based scheduling algorithm. Finally, we apply the model to the actual optimization problem that contained 829,684 variables and 137,319 constraints, and solved this problem by Python. The proposed method solves the optimal solution, consuming 575 s.

An integrated analysis is required to evaluate the performance of control algorithms used in power take-off (PTO) systems for floating wave energy converters (FWECs). However, research on PTO systems based on the existing hydraulic device has mainly focused on the input power generation performance rather than on obtaining maximum power through hydraulic device-based electrical load control. The power generation performance is analyzed based on the control variables of the existing torque control algorithm (TCA), however, the amount of power generation for each control variable changes significantly based on the cycle of wave excitation moments. This paper proposes a control algorithm to obtain the maximum power by modeling a hydraulic-device-based integrated FWEC. It also proposes a TCA that can obtain the maximum power regardless of the period of wave excitation moment. The proposed TCA continuously monitors the power generation output and changes the PTO damping coefficient in the direction in which the power generation output can be increased. The proposed TCA increased the output power generation by up to 18% compared to each PTO damping coefficient of the conventional TCA. Thus, the proposed method results in higher power generation regardless of the wave excitation moment cycle and performs better than the existing torque control algorithm.

With the rapid development of solar energy, the photovoltaic (PV) module fault detection plays an important role in knowing how to enhance the reliability of the solar photovoltaic system and knowing the fault type when a system problem occurs. Therefore, this paper proposed the hybrid algorithm of chaos synchronization detection method (CSDM) with convolutional neural network (CNN) for studying PV module fault detection. Four common PV module states were discussed, including the normal PV module, module breakage, module contact defectiveness and module bypass diode failure. First of all, the defects in 16 pieces of 20W monocrystalline silicon PV modules were preprocessed, and there were four pieces of each fault state. When the signal generator delivered high frequency voltage to the PV module, the original signal was measured and captured by the NI PXI-5105 high-speed data acquisition system (DAS) and was calculated by CSDM, to establish the chaos dynamic error map as the image feature of fault diagnosis. Finally, the CNN was employed for diagnosing the fault state of the PV module. The findings show that after entering 400 random fault data (100 data for each fault) into the proposed method for recognition, the recognition accuracy rate of the proposed method was as high as 99.5%, which is better than the traditional ENN algorithm that had a recognition rate of 86.75%. In addition, the advantage of the proposed algorithm is that the mass original measured data can be reduced by CSDM, the subtle changes in the output signals are captured effectively and displayed in images, and the PV module fault state is accurately recognized by CNN.

A Computational fluid dynamic (CFD) model was developed considering three geometries for primary parallel fan stations that have already been developed, implemented, and are currently in operation within Chilean mines. To standardize the comparison, the same primary fan was used in all the simulations with a unique set of settings (speed, blade angle, and density). The CFD representation was used to determine the operating point per configuration and compare the performances in terms of airflow and pressure delivered. This approach allowed ranking primary fan station geometry based on resistance curve and energy consumption of the fan. This paper presents the results obtained through the CFD simulations and the corresponding primary fans operating points of each configuration: symmetrical branches (SB), overlap branches (OB), and run around (RA) bypass. The RA configuration was identified as the best-performing station geometry on the lowest frictional and shock pressure losses, highest airflow delivery, and lowest energy cost. The results are discussed, considering pressure, velocity, and vector contours to understand the fluid dynamics phenomena occurring inside the station. The capital cost involved in the development of each primary parallel station was considered in the analysis in addition to the energy cost to determine the economic configuration over time.

This paper proposes a neural network-based process fault diagnosis system with Andrews plot for information pre-processing to enhance the performance of online process fault diagnosis. By using features extracted from Andrews plot as the inputs to a neural network, as a classifier, the diagnosis speed and reliability are improved. A method for determining the important features in the Andrews function is proposed. The proposed fault diagnosis system is applied to a simulated continuous stirred tank reactor process and is compared with two conventional neural network-based fault diagnosis systems: scheme B where the monitored measurements are directly fed to a neural network after scaling and scheme C where the monitored measurements are converted to qualitative trend data before feeding to a neural network. Of all the considered faults, the proposed fault diagnosis system diagnosed the abrupt faults on average 5.45 s and 2.66 s earlier than schemes B and C, respectively and diagnosed the incipient faults on average 13.82 s and 5.09 s earlier than schemes B and C, respectively. The results reveal that Andrews plot method utilized in online process monitoring has a great potential in industrial process monitoring.

The emission of greenhouse gasses is a major environmental problem, and efforts are being made worldwide in various ways to encourage producers to reduce their emissions. There is a need to incorporate environmental measures into process design and synthesis, as pollution prevention is a higher priority than waste management, and in this way, more sustainable solutions can also be achieved. One possibility is to introduce a CO2 tax, the value of which is very uncertain in the future. This paper demonstrates how the CO2 tax affects the optimal results of synthesizing chemical processes using mixed-integer nonlinear programming (MINLP). It was found that the tax increase promotes the use of better-quality raw materials and more efficient process units. Energy consumption and emissions are reduced and economic performance deteriorates. A multi-period, two-stage stochastic approach with recourse is suitable to incorporate the uncertainty of the CO2 tax in the MINLP process synthesis and gives better results than a simpler deterministic approach. In the case of the heat exchanger network synthesis, the costs obtained with the stochastic approach were 5% lower, and the emissions 7% lower than with a deterministic approach.

This paper studies the dynamics of a two-stage gear transmission system in both the normal state and the fault state with tooth breakage. The torsional vibration model of the two-stage parallel shaft gear was developed by using the lumped parameter method. The time-varying meshing stiffness of the gear transmission system is described by Fourier series which is determined by the periodical meshing characteristics of the gears with both the single-tooth and the double-tooth contacts. By introducing the pulse into the regular time-varying meshing stiffness, the tooth breakage existing in the gear transmission system is mimicked. Based on the numerical simulation of the developed dynamic model, both the time domain analysis and the frequency domain analysis of the gear transmission system under both the normal condition and the tooth breakage are compared accordingly. The influence of the tooth breakage on the dynamic characteristics of the gear transmission system is analyzed comprehensively. Furthermore, based on the developed test bench of a two-stage gear transmission system, the experimental research was carried out, and the experimental results show great agreements with the results of numerical simulation, and thus the validity of the developed mathematical model is demonstrated. By comparing the periodic motion with the chaotic motion, the fault identification for the gear transmission system is verified to be tightly related to its vibration condition, and the control of the vibration condition of the gear transmission system as periodic motion is of great significance to the fault diagnosis.

This paper puts forward an optimal quick-response variable structure control with a single-phase sine-wave inverter application, which keeps harmonic distortion as low as possible under various conditions of loading. Our proposed solution gives an improvement in architecture in which a quick-response variable structure control (QRVSC) is combined with a brain storm optimization (BSO) algorithm. Notwithstanding the intrinsic resilience of a typical VSC with respect to changes in plant parameters and loading disruptions, the system state convergence towards zero normally proceeds at an infinitely long-time asymptotically, and chattering behavior frequently takes place. The QRVSC for ensuring speedy limited-time convergence with the system state to the balancing point is devised, whilst the BSO will be employed to appropriately regulate the parametric gains in the QRVSC for the elimination of chattering phenomena. From the mix of both a QRVSC together with a BSO, a low total harmonic distortion (THD) as well as a high dynamic response across different types of loading is generated by a closed-loop inverter. The proposed solution is implemented on a practicable single-phase sine-wave inverter under the control of a TI DSP (Texas Instruments Digital Signal Processor). It has experimentally shown the simulation findings as well as the mathematical theoretical analysis, displaying that both quick transient reaction as well as stable performance could be obtained. The proposed solution successfully inhibits voltage harmonics in compliance with IEEE 519-2014’s stringent standard of limiting THD values to less than 5%.

In industrial piping systems, turbomachinery, heat exchangers etc., pipe bends are essential components. Computational fluid dynamics (CFD), which is frequently used to analyse the flow behaviour in such systems, provides extremely precise estimates but is computationally expensive. As a result, a computationally efficient method is developed in this paper by leveraging machine learning for such computationally expensive CFD problems. Random forest regression (RFR) is used as the machine learning algorithm in this work. Four different fluid flow characteristics (i.e., axial velocity, x-velocity, y-velocity and z-velocity) are studied in this work. The accuracy of the RFR models is assessed by using a number of statistical metrics such as mean-absolute error (MAE), mean-squared-error (MSE), root-mean-squared-error (RMSE), maximum error (Max.Error) and median error (Med.Error) etc. It is observed that the RFR models can produce considerable cost reductions in computing by surrogating the CFD model. Minor loss in estimation accuracy as compared to the CFD models is observed. While the magnitude of intricate flow characteristics such as the additional vortices are correctly predicted, some error in their location is observed.

This paper analyzed the problems associated with physical exergy splitting, and based on this, presented a new splitting method. This new method splits the physical exergy into three parts according to the feasible working ways, i.e.,: the direct, indirect, and adaptive exergy. The computational method and the physical meaning of the three exergy parts were presented in detail in terms of graphic representation and mathematical derivation. Then, it was applied to the exergy analysis of a thermal power cycle. The results show that compared with the conventional method which splits the physical exergy into thermal and mechanical parts, the current exergy splitting method can better represent the change rule of the working ability of the real working stream in the cycle and the influence of some operation parameters, such as the turbine inlet temperature, on the real working ability. The study suggests that the new method can make the exergy analysis more helpful and guidable in its applications.

There are cases where the output of renewable eappennergy (RE) is curtailed due to an increase in the share of RE. Typically, wind power (WP) is curtailed due to oversupply and low loads at midnight. However, there are cases where the output of WP is limited during the daytime due to the increase in the share of photovoltaics (PV). In the current electricity market, as the share of PV is increased, the curtailments of WP will be increased further, which will add to the difficulties experienced by wind farm operators. This paper proposes a supervisory power coordination scheme. The main purposes are to prevent the penetration of extra power from REs into the grid, thus, the curtailments can be prevented. In order to make it feasible, the proposed scheme is to design a grid-connected microgrid system to be operated only in response to loads and virtual power plant (VPP) requests. The effectiveness of the proposed scheme was verified by simulation studies conducted in the MATLAB/Simulink environment. The verification was conducted based on the voltage criteria, such as the AC voltage regulation between ±6% of the rated AC voltage, the DC voltage regulation between ±10% of the rated DC voltage, the power balance according to variations in the loads, and VPP requests for power. The simulation showed that the proposed scheme is feasible and justifiable, not only to mitigate the power curtailment problem but also to apply different system configurations.

Constrained optimization problems (COPs) are widely encountered in chemical engineering processes, and are normally defined by complex objective functions with a large number of constraints. Classical optimization methods often fail to solve such problems. In this paper, to solve COPs efficiently, a two-phase search method based on a heat transfer search (HTS) algorithm and a tandem running (TR) strategy is proposed. The main framework of the MHTS–TR (Multiple HTS) method aims to alternate between a feasible search phase that only examines feasible solutions, using the HTS algorithm, and an infeasible search phase where the treatment of infeasible solutions is relaxed in a controlled manner, using the TR strategy. These two phases play different roles in the search process, the former ensures an intensified optimum in a relevant feasible region, whereas the latter is used to introduce more diversity into the former. Thus, the ensemble of these two complementary phases can provide an effective method to solve a wide variety of COPs. The proposed variant was investigated over 24 well-known constrained benchmark functions, and then compared with various well-established metaheuristic approaches. Furthermore, it was applied for solving a chemical COP. The promising results demonstrate that the MHTS–TR approach is applicable for solving real-world COPs.

The application of scene recognition in intelligent robots to forklift AGV equipment is of great significance in order to improve the automation and intelligence level of distribution centers. At present, using the camera to collect image information to obtain environmental information can break through the limitation of traditional guideway and positioning equipment, and is beneficial to the path planning and system expansion in the later stage of warehouse construction. Taking the forklift AGV equipment in the distribution center as the research object, this paper explores the scene recognition and path planning of forklift AGV equipment based on a deep convolution neural network. On the basis of the characteristics of the warehouse environment, a semantic segmentation network applied to the scene recognition of the warehouse environment is established, and a scene recognition method suitable for the warehouse environment is proposed, so that the equipment can use the deep learning method to learn the environment features and achieve accurate recognition in the large-scale environment, without adding environmental landmarks, which provides an effective convolution neural network model for the scene recognition of forklift AGV equipment in the warehouse environment. The activation function layer of the model is studied by using the activation function with better gradient performance. The results show that the performance of the H-Swish activation function is better than that of the ReLU function in recognition accuracy and computational complexity, and it can save costs as a calculation form of the mobile terminal.

In this paper, internal division point genetic algorithm (IDP-GA) was proposed to lessen the computational burden of multi-variable multi-objective optimization problem using finite element analysis such as optimal design of electric bicycles. The IDP-GA could consider various objectives with normalized weighted sum method and could reduce the number of function calls with novel crossover strategy and vector-based pattern search method. The superiority of the proposed algorithm was verified by comparing performances with conventional optimization method at two mathematical test functions. Finally, the applicability of the IDP-GA in practical electric machine design was verified by successfully deriving an improved design of electric bicycle propulsion motor.

Manufacturers are eager to replace the human inspector with automatic inspection systems to improve the competitive advantage by means of quality. However, some manufacturers have failed to apply the traditional vision system because of constraints in data acquisition and feature extraction. In this paper, we propose an inspection system based on deep learning for a tampon applicator producer that uses the applicator’s structural characteristics for data acquisition and uses state-of-the-art models for object detection and instance segmentation, YOLOv4 and YOLACT for feature extraction, respectively. During the on-site trial test, we experienced some False-Positive (FP) cases and found a possible Type I error. We used a data-centric approach to solve the problem by using two different data pre-processing methods, the Background Removal (BR) and Contrast Limited Adaptive Histogram Equalization (CLAHE). We have experimented with analyzing the effect of the methods on the inspection with the self-created dataset. We found that CLAHE increased Recall by 0.1 at the image level, and both CLAHE and BR improved Precision by 0.04–0.06 at the bounding box level. These results support that the data-centric approach might improve the detection rate. However, the data pre-processing techniques deteriorated the metrics used to measure the overall performance, such as F1-score and Average Precision (AP), even though we empirically confirmed that the malfunctions improved. With the detailed analysis of the result, we have found some cases that revealed the ambiguity of the decisions caused by the inconsistency in data annotation. Our research alerts AI practitioners that validating the model based only on the metrics may lead to a wrong conclusion.

This paper presents an algorithm for efficient uncertainty quantification (UQ) in the presence of many uncertainties that follow a nonstandard distribution (e.g., lognormal). Using the polynomial chaos expansion (PCE), the algorithm builds surrogate models of uncertainty as functions of a standard distribution (e.g., Gaussian variables). The key to build these surrogate models is to calculate PCE coefficients of model outputs, which is computationally challenging, especially when dealing with models defined by complex functions (e.g., nonpolynomial terms) under many uncertainties. To address this issue, an algorithm that integrates the PCE with the generalized dimension reduction method (gDRM) is utilized to convert the high-dimensional integrals, required to calculate the PCE coefficients of model predictions, into several lower-dimensional ones that can be rapidly solved with quadrature rules. The accuracy of the algorithm is validated with four examples in structural reliability analysis and compared to other existing techniques, such as Monte Carlo simulations and the least angle regression-based PCE. Our results show our algorithm provides accurate UQ results and is computationally efficient when dealing with many uncertainties, thus laying the foundation to address UQ in complex control systems.

In this paper, a new hybrid algorithm based on two meta-heuristic algorithms is presented to improve the optimization capability of original algorithms. This hybrid algorithm is realized by the deep ensemble of two new proposed meta-heuristic methods, i.e., slime mold algorithm (SMA) and arithmetic optimization algorithm (AOA), called DESMAOA. To be specific, a preliminary hybrid method was applied to obtain the improved SMA, called SMAOA. Then, two strategies that were extracted from the SMA and AOA, respectively, were embedded into SMAOA to boost the optimizing speed and accuracy of the solution. The optimization performance of the proposed DESMAOA was analyzed by using 23 classical benchmark functions. Firstly, the impacts of different components are discussed. Then, the exploitation and exploration capabilities, convergence behaviors, and performances are evaluated in detail. Cases at different dimensions also were investigated. Compared with the SMA, AOA, and another five well-known optimization algorithms, the results showed that the proposed method can outperform other optimization algorithms with high superiority. Finally, three classical engineering design problems were employed to illustrate the capability of the proposed algorithm for solving the practical problems. The results also indicate that the DESMAOA has very promising performance when solving these problems.

Many ports worldwide continue to expand their capacity by developing a multiterminal system to catch up with the global containerized trade demand. However, this expansion strategy increases the demand for container exchange between terminals and their logistics facilities within a port, known as interterminal transport (ITT). ITT forms a complex transportation network in a large port, which must be managed efficiently given the economic and environmental implications. The use of trucks in ITT operations leads to the interterminal truck routing problem (ITTRP), which has been attracting increasing attention from researchers. One of the objectives of truck routing optimization in ITT is the minimization of empty-truck trips. Selection of the transport order (TO) based on the current truck location is critical in minimizing empty-truck trips. However, ITT entails not only transporting containers between terminals operated 24 h: in cases where containers need to be transported to a logistics facility within operating hours, empty-truck trip cost (ETTC) minimization must also consider the operational times of the transport origin and destination. Otherwise, truck waiting time might be incurred because the truck may arrive before the opening time of the facility. Truck waiting time seems trivial, but it is not, since thousands of containers move between locations within a port every day. So, truck waiting time can be a source of ITT-related costs if it is not managed wisely. Minimization of empty-truck trips and truck waiting time is considered a multiobjective optimization problem. This paper proposes a method of cooperative multiagent deep reinforcement learning (RL) to produce TO truck routes that minimize ETTC and truck waiting time. Two standard algorithms, simulated annealing (SA) and tabu search (TS) were chosen to assess the performance of the proposed method. The experimental results show that the proposed method represents a considerable improvement over the other algorithms.

A bidirectional resonant converter is presented and verified in this paper for an electric vehicle battery charger/discharger system. The presented circuit can achieve forward and backward power operation, low switching losses on active devices, and wide output voltage operation. The circuit structure of the presented converter includes two resonant circuits on the primary and secondary sides of an isolated transformer. The frequency modulation approach is adopted to control the studied circuit. Owing to the resonant circuit characteristic, active devices for both forward (battery charge) and backward (battery discharge) power operation can be turned on at zero voltage switching. In order to implement a universal battery charger for different kinds of electric vehicle applications, the DC converter is demanded to have a wide output voltage range capability. The topology morphing between a full bridge resonant circuit and half bridge resonant circuit is selected to obtain high- and low-output voltage range operations so that the 200–500 V output voltage range is realized in the presented resonant converter. Compared to the conventional bidirectional converters, the proposed can be operated under a wide voltage range operation. In the end, a 1 kW laboratory prototype circuit is built, and experiments are provided to demonstrate the validity and performance of the presented bidirectional resonant converter.

As the object detection dataset scale is smaller than the image recognition dataset ImageNet scale, transfer learning has become a basic training method for deep learning object detection models, which pre-trains the backbone network of the object detection model on an ImageNet dataset to extract features for detection tasks. However, the classification task of detection focuses on the salient region features of an object, while the location task of detection focuses on the edge features, so there is a certain deviation between the features extracted by a pretrained backbone network and those needed by a localization task. To solve this problem, a decoupled self-attention (DSA) module is proposed for one-stage object-detection models in this paper. A DSA includes two decoupled self-attention branches, so it can extract appropriate features for different tasks. It is located between the Feature Pyramid Networks (FPN) and head networks of subtasks, and used to independently extract global features for different tasks based on FPN-fused features. Although the DSA network module is simple, it can effectively improve the performance of object detection, and can easily be embedded in many detection models. Our experiments are based on the representative one-stage detection model RetinaNet. In the Common Objects in Context (COCO) dataset, when ResNet50 and ResNet101 are used as backbone networks, the detection performances can be increased by 0.4 and 0.5% AP, respectively. When the DSA module and object confidence task are both applied in RetinaNet, the detection performances based on ResNet50 and ResNet101 can be increased by 1.0 and 1.4% AP, respectively. The experiment results show the effectiveness of the DSA module.