Search

• The AES power system intelligent energy scheduling environment is built. • The BI-LSTM-Att action network is propose by combining BI-LSTM with attention mechanism. • The DQN-CE algorithm is proposed by improving the DQN algorithm. • The experiments on an all-electric ferry is conducted using a variety of methods. At present, Ship energy scheduling based on deep reinforcement learning (DRL) is an important research direction in which Deep Q learning algorithm (DQN) has been successfully applied in the field of energy scheduling. In terms of the shortcomings of DQN in all electric ships(AES) energy scheduling, such that, the insufficient performance of using multiplayer perceptron(MLP) as action network and the degradation of convergence performance due to over estimation of Q value, in this paper, we propose a DQN cross entropy(DQN-CE) energy scheduling algorithm combing bi-directional LSTM and attention mechanism(BI-LSTM-Att). Instead of using MLP, firstly, we use BI-LSTM-Att as action network of DQN, and then an improved DRL algorithm called DQN-CE is proposed, that is, we add cross-entropy loss of predicted action of the next time for the target network and the action network to DQN. Finally, we conduct simulation experiments on an all-electric ferry, and the experimental results show that the energy scheduling algorithm proposed in this paper reduces the economic consumption by 4.11%, increases the utilization rate of energy storage system by 24.4%, and reduces the exploration time of agent training by 31.3% compared with the original DQN energy scheduling algorithm. Finally, the effectiveness and superiority of the algorithm were further verified in the new case study. [ABSTRACT FROM AUTHOR]

Accurately capturing data on the external loads that large structural systems endure is crucial for improving the performance of energy equipment. This paper introduces a novel hybrid model-data-driven framework for the dynamic load identification of interval structures, which seamlessly combines finite-element modeling with machine learning techniques. To address potential ill-posed issues in model-driven methods and the interpretability limitations of data-driven methods, we propose a physics-informed neural network. This neural network effectively inverts uncertain modal responses with low data requirements and high predictive performance high by integrating the underlying modal transformation equation into the loss function of a fully connected neural network. To identify the modal loads using predicted modal displacement/acceleration responses, we introduce a pioneering dynamics inversion method. This method modifies the traditional Kalman filter with an assumption of unknown inputs to reduce the sensitivity of load identification process to different noises. In addition, our approach incorporates a subinterval Chebyshev expansion method to adaptively determine the interval boundaries of external loads. The efficiency of the proposed method is assessed through two numerical examples and validated through comparative research against baseline methods. Our findings suggest that this approach enhances precision, robustness, and generalization in dynamic load identification, even when facing challenges such as limited training data, significant noise interference, and non-zero initial conditions. • A novel hybrid model-data-driven dynamic load identification method for interval structures is proposed. • A physics-informed neural network is established to inverse the uncertain modal responses effectively. • Two improved Kalman filter algorithms are developed to identify the modal loads at specified parameters. [ABSTRACT FROM AUTHOR]

Highlights • Co-simulation and building energy models provide a framework to test control algorithms. • Machine learning algorithms are effective to develop predictive optimisation models. • Predictive controls for residential can reduce heating systems costs up to 40%. • Smart predictive control in buildings results in lower carbon footprint up to 39%. • The smart grid dynamically controls dwellings to balance electricity supply/demand. Abstract This paper assesses the performance of control algorithms for the implementation of demand response strategies in the residential sector. A typical house, representing the most common building category in Ireland, was fully instrumented and utilised as a test-bed. A calibrated building simulation model was developed and used to assess the effectiveness of demand response strategies under different time-of-use electricity tariffs in conjunction with zone thermal control. Two demand response algorithms, one based on a rule-based approach, the other based on a predictive-based (machine learning) approach, were deployed for control of an integrated heat pump and thermal storage system. The two algorithms were evaluated using a common demand response price scheme. Compared to a baseline reference scenario, the following reductions were observed: electricity end-use expenditure (20.5% rule-based and 41.8% predictive algorithm), utility generation cost (18.8% rule-based and 39% predictive algorithm), carbon emissions (20.8% rule-based and 37.9% predictive algorithm). [ABSTRACT FROM AUTHOR]

Highlights • Create city-specific energy performance grading model using city disclosure data. • Method to grade buildings that is interpretable, reproducible, scalable, and robust. • Develop a clustering algorithm to grade buildings on performance. • Validated approach demonstrates significant improvements over current methods. Abstract As the effects of anthropogenic climate change become more pronounced, local and federal governments are turning towards more aggressive policies to reduce energy use in existing buildings, a major global contributor of carbon emissions. Recently, several cities have enacted laws mandating owners of large buildings to publicly display an energy efficiency rating for their properties. While such transparency is necessary for market-driven energy reduction policies, the reliance on public-facing energy efficiency grades raises non-trivial questions about the robustness and reliability of methods used to measure and benchmark the energy performance of existing buildings. In this paper, we develop a building energy performance grading methodology using machine learning and city-specific energy use and building data. Leveraging the growing availability of data from city energy disclosure ordinances, we develop the GREEN grading system: a framework to facilitate more accurate, fair, and contextualized building energy benchmarks that account for variations in the expected and actual performance of individual buildings. When applied to approximately 7500 residential properties in New York City, our approach accounts for the differential impact of design, occupancy, use, and systems on energy performance, out-performing existing state-of-the-art methods. Our model and findings reinforce the need for more robust, localized approaches to building energy performance grading that can serve as the basis for data-driven urban energy efficiency and carbon reeduction policies. [ABSTRACT FROM AUTHOR]

The knowledge of the dynamic available charging and discharging power of the battery is a piece of essential information for the safety and longevity of the battery energy storage systems. An accurate real-time prediction of these quantities is very challenging due to the high nonlinearities of battery dynamics. In this paper, an electrochemical model-based online state-of-power prediction algorithm under different time horizons is developed for a safer and more reliable operation of lithium-ion batteries. The safety constraints, which define the safety operation area for the power prediction, are designed based on not only the terminal voltage but also battery internal electrochemical states, i.e., the electrode surface concentration, the electrolyte concentration, and the side reaction overpotential. The algorithm is validated by simulations and experiments under a dynamic load profile, and the dominating constraints in charging and discharging as well as the influence of predictive time horizons on the available battery power are analyzed, providing important information for further researches. Furthermore, the computational speed of the proposed iterative algorithm is improved with the integration of Gaussian process regression by up to 50%. A comparative study with a state-of-the-art equivalent circuit model-based approach highlights the significant benefits of the proposed electrochemical model-based algorithm in operation safety enhancement and battery performance improvement. [Display omitted] • Online power prediction considering internal electrochemical safety constraints. • Computational efficiency improvement by 50% with Gaussian process regression. • Numerical and experimental verification under real-world driving cycles. • Dominating safety constraint analysis for maximum charging and discharging power. • Outperforming state-of-the-art power prediction methods in all time horizons. [ABSTRACT FROM AUTHOR]

In view of increasing concerns about climate change and global warming, various renewable energy sources are widely adopted in power systems to reduce greenhouse gas emissions. However, the widespread adoption of renewable energy sources exacerbates the problem of harmonics pollution. In order to improve power quality, it is urgent to develop a new approach for monitoring harmonics. By leveraging the structured sparsity and spatial sparsity of harmonic sources, this paper proposes a group sparse Bayesian learning method for solving harmonic state estimation problems. The proposed algorithm can employ a small number of measurements less than the number of state variables to pinpoint harmonic sources and to estimate the harmonic magnitudes and angles. Moreover, the proposed method achieves automatic hyperparameter adjustments and solves the system states in complex domain without splitting them into real and imaginary parts. Extensive results on a benchmark IEEE 14-bus system are presented to substantiate the superiority of the proposed method in terms of localization accuracy and harmonic magnitude and phase angle estimation. • A group sparse Bayesian learning method is proposed for harmonic state estimation. • The proposed algorithm achieves automatic hyperparameter adjustments. • The proposed method can directly handle the measurements in the complex domain. • Using a small number of measurements pinpoints the harmonic sources. • The proposed method has high localization precision and small estimation error. [ABSTRACT FROM AUTHOR]

• Propose a novel hybrid model for multi-step ahead daily coal price forecasting. • Decompose original data into multiple modes via variational mode decomposition. • Use long short-term memory network containing attention layer to predict each mode. • Predicted mode results are ensembled by support vector regression model. • Proposed hybrid model is superior to baseline models in coal price forecasting. Accurate and reliable coal price prediction is of great significance to enhance the stability of the coal market. Numerous methods have been developed to improve the prediction performance. However, most of the studies adopt single model for coal price forecasting, and their accuracy and applicability are usually restricted. In this paper, we propose a novel hybrid VMD-A-LSTM-SVR model to achieve accurate multi-step ahead prediction of coal price. The proposed model consists of three valuable strategies. First, variational mode decomposition (VMD) decomposes the original coal price into several relatively regular sub modes to reduce the non-stationarity and uncertainty of the data. Second, the long short-term memory (LSTM) integrated with attention mechanism trains and predicts the decomposed modes individually to better capture the temporal information of historical data. Lastly, a support vector regression (SVR) model ensembles the predicted results of each mode into the final forecasted coal price. The experimental results of three typical coal price datasets demonstrate that the proposed strategies are all valuable for improving the forecasting performance. Moreover, the proposed model outperforms all state-of-the-art baseline models in terms of both model accuracy and stability. Extensive cross-comparisons of performance between models clearly indicate that the proposed hybrid algorithm is more effective and practical for coal price forecasting. [ABSTRACT FROM AUTHOR]

• Extreme learning machine is employed to predict data and filter noised data. • Data shortage and noised data obstacles for parameter extraction are solved. • Eight meta -heuristic algorithms are employed to two solid oxide fuel cell models. • Practical performance of the proposed method is validated in the 5 kW cell stack. • Two constructive recommendations are proposed for future development. A precise, fast, and robust parameter extraction technique of solid oxide fuel cell models is extremely crucial for optimal control and behavior analysis. In this paper, a novel extreme learning machine based method is proposed to extract unknown parameters of solid oxide fuel cell models including electrochemical model and simple electrochemical model. At first, extreme learning machine is applied to overcome two thorny obstacles (e.g., data shortage and noised data) via predicting additional data and updating noised data. Then, both original data collected from a 5 kW solid oxide fuel cell stack and processed data are transferred to effectively guide eight prominent meta-heuristic algorithms for effective parameter extraction. The performance of extreme learning machine is thoroughly investigated in two typical operation conditions through a comprehensive comparison based on various training data. Simulation results validate that the proposed approach can effectively contribute to searching efficient model parameters along with high accuracy, prominent stability, high speed, and great robustness. Particularly, the accuracy of parameter extraction for electrochemical model and simple electrochemical model can be improved by 49.3% and 65.6% at most, respectively. [ABSTRACT FROM AUTHOR]

Controlling heating, ventilation and air-conditioning (HVAC) systems is crucial to improving demand-side energy efficiency. At the same time, the thermodynamics of buildings and uncertainties regarding human activities make effective management challenging. While the concept of model-free reinforcement learning demonstrates various advantages over existing strategies, the literature relies heavily on value-based methods that can hardly handle complex HVAC systems. This paper conducts experiments to evaluate four actor-critic algorithms in a simulated data centre. The performance evaluation is based on their ability to maintain thermal stability while increasing energy efficiency and on their adaptability to weather dynamics. Because of the enormous significance of practical use, special attention is paid to data efficiency. Compared to the model-based controller implemented into EnergyPlus, all applied algorithms can reduce energy consumption by at least 10% by simultaneously keeping the hourly average temperature in the desired range. Robustness tests in terms of different reward functions and weather conditions verify these results. With increasing training, we also see a smaller trade-off between thermal stability and energy reduction. Thus, the Soft Actor Critic algorithm achieves a stable performance with ten times less data than on-policy methods. In this regard, we recommend using this algorithm in future experiments, due to both its interesting theoretical properties and its practical results. • Comparison of four model-free actor-critic reinforcement learning algorithms in a data-centre simulation. • Each of the applied algorithms can learn the task and reduce energy consumption. • The conflict between thermal stability and energy savings is reduced during the training. • The algorithms are robust to changing reward functions and weather conditions. • Soft Actor Critic shows the highest data efficiency and is recommended for applications. [ABSTRACT FROM AUTHOR]

Forecasting the available flexible load provided by electric vehicles would enable electric utilities to make informed decision in utilizing these loads for enhancing the operational efficiency of distribution systems. To overcome the lack of historical loads data at newly-installed EV charging stations, this paper proposes a clustered multi-node learning with Gaussian Process (CMNL-GP) method to fuse data from multiple charging stations and to learn them simultaneously. The proposed method improves the forecasting accuracy in each node by transferring meaningful information among multiple nodes. The proposed method also performs a clustering algorithm within its objective function to obtain within-cluster similarity, since all the nodes may not be related equally, and the nodes within a cluster may have a stronger correlation. To characterize the clustered structures and to transfer the shared information among multiple nodes, different regularization terms are imposed in the objective function of the proposed method. The proposed clustered multi-node learning also utilizes the Gaussian Process for statistical attributes of the residual stochastic process, which refers to the information that may not be shared among multiple nodes and can be node-specific. The proposed method is validated by real-world EV charging stations data in State of Utah, USA, to demonstrate the effectiveness of the proposed algorithm. • A novel method proposed for flexible load forecasting represented by EVs • Flexible load data from similar EV charging stations are learned jointly • The existing multi-task learning algorithm is improved for time-series data • Clear discussions and guidelines for implementing of the proposed method are provided [ABSTRACT FROM AUTHOR]