IET Renewable Power Generation最新文献

This paper presents a bi-objective optimisation approach for grid-connected microgrids, aiming to minimise operational costs and voltage deviation at the connection nodes of distributed energy resources and loads. Existing research typically addresses these objectives separately, and the simultaneous consideration of economic performance and voltage deviation in grid-connected community microgrids with multiple generation resources remains in an early stage of development. To advance the research in this area, a novel mean-guided elite selection genetic algorithm (MGES-GA) is proposed to enhance the balance between convergence and diversity in multi-objective optimisation. The proposed algorithm enhances the selection process by re-evaluating low-performing individuals through gene mixing with elite solutions, thereby preserving diversity and avoiding premature convergence. Comparative analysis of the MGES-GA with the enhanced genetic algorithm, differential evolution with heuristic, and improved differential evolutionary optimisation algorithms demonstrates its superior performance in optimising the economic dispatch of a grid-connected microgrid. In a bi-objective comparison with state-of-the-art algorithms, tested on a modified IEEE European low-voltage test feeder and IEEE 33-bus network, MGES-GA demonstrates its effectiveness in balancing conflicting objectives by producing lower voltage deviations at comparable or lower costs.

To solve the problems of power fluctuations, voltage violations, transformer overloading and restrictions on renewable energy consumption in ports, which are caused by fluctuation of renewables and regenerative braking power, an energy storage system (ESS) control strategy and a nested bi-layer configuration method based on lithium titanate oxide (LTO) batteries are proposed to reduce power fluctuations of the ports main grid, enhancing the integration and utilization of renewable energy. The inner layer is the strategy of ESS to suppress power fluctuations in ports. First, with the goal of maximising the utilisation of regenerative energy of quay cranes, a multiple quay cranes collaborative optimisation operation model is established to reduce the total power fluctuation of ports. Then, an ESS coordinated strategy based on Pontryagin's minimum principle is proposed to reduce the fluctuating power of energy-type ESS. The outer layer is the capacity configuration model of LTO. Based on the power fluctuation of energy-type ESS, a capacity optimisation model is established with the goal of minimising the annual configuration cost of LTO during the project operation period. The inner and outer layers realise the iterative optimisation of energy storage operation strategy and capacity configuration through operation parameters and capacity parameters interaction. The case study results show that the average power fluctuation of each quay crane is reduced by 26.2% under the cooperative operation (three quay cranes). At the same time, the average power fluctuation of energy-type ESS is reduced by 26.3% compared with the traditional adaptive first-order filtering strategy. The average annual cost of LTO is decreased by 17.5% compared to the commonly used lithium iron phosphate battery.

The growing integration of power electronics converters (PECs) and multi-terminal high voltage DC (MT-HVDC) grids within the power system decreases the system's inertia. Conversely, maintaining the voltage level of the MT-HVDC grid is crucial for preserving the overall system's stability. One of the primary challenges in generating virtual inertia for AC grids connected to MT-HVDC grids is the further decline in DC voltage caused by the additional power absorption needed for virtual inertia provision. This indicates that the implementation of virtual inertia negatively impacts DC voltage levels. In order to elucidate this issue, the present study develops a small-signal model of the Cigre-DCS3, incorporating a virtual synchronous generator (VSG). This model aims to analyse the effects of VSG parameters on the stability characteristics of the system under consideration. This analysis reveals a conflicting interaction between the DC voltage droop control loop and the virtual inertia time constant in the VSGs, as the presence of virtual inertia tends to adversely affect the DC-side voltage stability. In response to this challenge, this paper introduces an innovative approach that integrates DC voltage stability considerations into the virtual inertia control loop. This integration aims to improve the dynamic response of VSGs while enhancing overall system reliability. The proposed method incorporates the rate of change of frequency, variations in frequency, and deviations in DC voltage to provide adaptive virtual inertia (AVI). Additionally, the stability of the presented controller is validated through Lyapunov stability analysis. Lastly, the simulation results illustrate the efficiency of the proposed approach in enhancing overall system performance.

The growing integration of power electronics converters (PECs) and multi-terminal high voltage DC (MT-HVDC) grids within the power system decreases the system's inertia. Conversely, maintaining the voltage level of the MT-HVDC grid is crucial for preserving the overall system's stability. One of the primary challenges in generating virtual inertia for AC grids connected to MT-HVDC grids is the further decline in DC voltage caused by the additional power absorption needed for virtual inertia provision. This indicates that the implementation of virtual inertia negatively impacts DC voltage levels. In order to elucidate this issue, the present study develops a small-signal model of the Cigre-DCS3, incorporating a virtual synchronous generator (VSG). This model aims to analyse the effects of VSG parameters on the stability characteristics of the system under consideration. This analysis reveals a conflicting interaction between the DC voltage droop control loop and the virtual inertia time constant in the VSGs, as the presence of virtual inertia tends to adversely affect the DC-side voltage stability. In response to this challenge, this paper introduces an innovative approach that integrates DC voltage stability considerations into the virtual inertia control loop. This integration aims to improve the dynamic response of VSGs while enhancing overall system reliability. The proposed method incorporates the rate of change of frequency, variations in frequency, and deviations in DC voltage to provide adaptive virtual inertia (AVI). Additionally, the stability of the presented controller is validated through Lyapunov stability analysis. Lastly, the simulation results illustrate the efficiency of the proposed approach in enhancing overall system performance.

Renewable energy resources (RERs)-based distributed generations (DGs) such as wind turbines (WTs) and photovoltaics (PVs) are being widely used worldwide, and it is important to consider their impact on the distribution network planning (DNP) problems. In a distribution network, there are many service transformers, and it is vital to consider the impact of RERs on the optimum location and capacity determination of these transformers. Since the shapes of load profiles and RERs generation may highly affect the planning results, in this paper, simultaneous planning of service transformers, PVs and WTs is investigated with respect to different load profiles. Due to the complexity of this problem, the crow search algorithm (CSA), particle swarm optimisation (PSO), differential CSA and a novel method which combines CSA and PSO (CSA-PSO) are applied, and a comparison of the results is performed. To validate the performance of CSA-PSO, a sensitivity analysis is conducted on the key parameter of the algorithm. Over the case study, it is observed that (1) the shape of the load profile and local RER generation significantly influence the placement and capacity of service transformers, (2) compared to a PV system, due to its lower cost and better correlation with the load profile, aWT system is preferable and is installed in the network, and (3) CSA-PSO and differential CSA produce better results than the other investigated methods. Furthermore, over the studied load profiles, the average ratio of the installed WT capacity to the total transformers’ capacity is around 28.3%.

The increasing adoption of distributed energy resources has greatly amplified interest in microgrids, whose effective, reliable and resilient operation relies on the performance of their energy management systems (EMS). These systems ensure the economic operation and maintain load-generation balance. A practical microgrid EMS (M-EMS) incorporates data monitoring, variable forecasting, resource allocation and online supervision to optimise the system while interacting with electricity markets. However, in the inherently uncertain environment of both stand-alone and grid-connected microgrids, variations in key variables can significantly impact the decision-making outcomes of M-EMS. This review paper explores various sources of uncertainties within microgrids, including forecast errors and uncertainties arising from modelling approximations or monitoring inaccuracies. It also provides insights into handling these uncertainties by thoroughly reviewing the pertinent literature and exploring strategies such as analytical methods and AI-based approaches for capturing them. The eventual goal of handling the uncertainties is to enhance system reliability and security through robust energy management solutions. Furthermore, practical measures to mitigate uncertainties are discussed. The practical implementation of these concepts is illustrated through a review of commercially available M-EMS solutions and real-world projects demonstrating their effectiveness in managing energy resources. This paper aims to help both researchers and industry professionals perceive the uncertainties within M-EMS and how to handle them to achieve accurate, optimal solutions and avoid unexpected costs. Emerging trends and future research directions are also outlined.

The increasing adoption of distributed energy resources has greatly amplified interest in microgrids, whose effective, reliable and resilient operation relies on the performance of their energy management systems (EMS). These systems ensure the economic operation and maintain load-generation balance. A practical microgrid EMS (M-EMS) incorporates data monitoring, variable forecasting, resource allocation and online supervision to optimise the system while interacting with electricity markets. However, in the inherently uncertain environment of both stand-alone and grid-connected microgrids, variations in key variables can significantly impact the decision-making outcomes of M-EMS. This review paper explores various sources of uncertainties within microgrids, including forecast errors and uncertainties arising from modelling approximations or monitoring inaccuracies. It also provides insights into handling these uncertainties by thoroughly reviewing the pertinent literature and exploring strategies such as analytical methods and AI-based approaches for capturing them. The eventual goal of handling the uncertainties is to enhance system reliability and security through robust energy management solutions. Furthermore, practical measures to mitigate uncertainties are discussed. The practical implementation of these concepts is illustrated through a review of commercially available M-EMS solutions and real-world projects demonstrating their effectiveness in managing energy resources. This paper aims to help both researchers and industry professionals perceive the uncertainties within M-EMS and how to handle them to achieve accurate, optimal solutions and avoid unexpected costs. Emerging trends and future research directions are also outlined.

Microgrids are increasingly integrating renewable sources like solar and wind to improve sustainability and to reduce reliance on fossil fuels. However, the variable nature of the load demand and the renewable power generation create challenges in maintaining stable and efficient operation of the microgrid. To address these challenges, a distributionally robust chance-constrained model integrated with the grey wolf optimizer is proposed for reliable scheduling of renewable generation and load. This study focuses on a multi-objective energy management model which analyzes eight grid-connected microgrid configurations. In addition, eight diverse optimization algorithms are utilized to compare configurations of these microgrids based on total cost, emissions, and reliability. The investigated results show that the optimal configuration reduces total costs by 47.06% and greenhouse gas emission cost by 78.92% in comparison to the base case, in which the grid alone meets the load. The investigation also confirms that the optimal configuration requires an investment return of 315.75% and a reduction in social welfare cost of 47.063%. In addition, the reliability enhancement is shown by the low values of Expected Energy Not Served, Loss of Load Expectation, Loss of Load Probability, System Average Interruption Duration Index, and Annual Expenditure on Load Interruptions. The optimal configuration is validated on the standard IEEE-33 and IEEE-69 bus systems, and application results are presented to show the benefits and practicality of this work.

Microgrids are increasingly integrating renewable sources like solar and wind to improve sustainability and to reduce reliance on fossil fuels. However, the variable nature of the load demand and the renewable power generation create challenges in maintaining stable and efficient operation of the microgrid. To address these challenges, a distributionally robust chance-constrained model integrated with the grey wolf optimizer is proposed for reliable scheduling of renewable generation and load. This study focuses on a multi-objective energy management model which analyzes eight grid-connected microgrid configurations. In addition, eight diverse optimization algorithms are utilized to compare configurations of these microgrids based on total cost, emissions, and reliability. The investigated results show that the optimal configuration reduces total costs by 47.06% and greenhouse gas emission cost by 78.92% in comparison to the base case, in which the grid alone meets the load. The investigation also confirms that the optimal configuration requires an investment return of 315.75% and a reduction in social welfare cost of 47.063%. In addition, the reliability enhancement is shown by the low values of Expected Energy Not Served, Loss of Load Expectation, Loss of Load Probability, System Average Interruption Duration Index, and Annual Expenditure on Load Interruptions. The optimal configuration is validated on the standard IEEE-33 and IEEE-69 bus systems, and application results are presented to show the benefits and practicality of this work.

In this study, an innovative approach with enhanced recursive feature clustering for load forecasting in smart solar microgrids by integrating density-based spatial clustering (DBSCAN) and radial basis function neural networks (RBFNN) encoder is proposed. Our methodology is based upon novel density-based clustering with feature recursive forwarding RBFNN-LSTM for high-accuracy micro- and macro-feature learning in temporal data. DBSRad-LSTM performance is evaluated using three distinct datasets: Panama electricity consumption, Italy solar electric load, and a custom dataset tailored for smart grid applications. Through rigorous comparative analysis, our DBSRad-LSTM model outperformed traditional machine learning models such as gated recurrent unit (GRU), long short-term memory (LSTM), and convolutional neural networks (CNN) across several metrics. Specifically, DBSRad-LSTM demonstrated superior performance in terms of accuracy, thereby contributing enhanced load forecasting capabilities. The proposed model integrating RBFN linear functionality with the attention of LSTM and DBSCAN clustering to enhance the learning of temporal data outperformed CNN, SVMCNN, and GRU on Panama power consumption, Italy electric load and bespoke datasets, obtaining a higher R² value of 0.89 and much lower MSE 0.015, RMSE 0.123, and MAE of 0.009. Achieving a 9%–25% improvement in error metrics and an average 13% better fit. By offering a distinct clustering-based approach that improves on existing methods, this research makes a substantial contribution to the field of smart grid management and opens the door for more precise and effective energy distribution systems.