This paper implements two novel meta-heuristic algorithms, including the Coati optimization algorithm (COA) and War strategy optimization (WSO) for determining the optimal solutions to the optimal power flow problem incorporating the use of wind turbines (WTs). Two objective functions are considered in this study, including minimizing the entire electricity generation expenditure (EEGE) with the value point effect and minimizing the voltage fluctuation index (VFI). IEEE 30-bus system is chosen to conduct the whole study and validate the efficiency of the two applied methods. Furthermore, DFIG WTs are used in grids with varying power output and power factor ranges. The comparison of the results obtained from the two methods in all case studies reveals that WSO is vastly superior to COA in almost all aspects. In addition, the positive contributions of WTs to the EEGE and VFI while they are properly placed in the grid are also clarified by using WSO. As a result, WSO is acknowledged as a highly effective search method for dealing with such optimal power flow (OPF) problems considering the presence of renewable energy sources.
Sub-synchronous oscillation (SSO) is considered a stability issue related to turbine-generator systems, posing negative impacts on system integrity. The high penetration of wind energy has increased the urgency for this phenomenon. The analyses in this paper reveal that high compensation levels and low wind speeds are the two primary factors driving SSO, categorized as the induction generator effect (IGE), whereas the control parameters of the rotor-side converter (RSC) adversely contribute to the sub-synchronous control interaction (SSCI) phenomenon. Various solutions have been explored to suppress the SSO, which are divided into grid side and generation side approaches. However, relying on a single solution for either side does not provide the optimal damping effectiveness under critical operating conditions. Therefore, this paper proposes optimizing the internal control parameters of the RSC using meta-heuristic algorithms, including Particle Swarm Optimization (PSO), Cuckoo Search Algorithm (CSA), and Ant Colony Optimization (ACO). These tuned parameters are combined with a grid side solution based on a thyristor-controlled series capacitor (TCSC) or a generation side solution involving a supplementary damping controller (SDC) added to the RSC to improve SSO damping. This combination offers advantages such as low investment costs, simplicity, and high effectiveness. Both time-domain simulations and small-signal analyses have been carried out to demonstrate the capability of the improved approach, as they significantly enhance the damping ratio across low wind speed ranges and wide compensation levels. This study was conducted using MATLAB software based on the IEEE first benchmark, which was connected to the IEEE 14-bus system.
This study proposes a new approach for optimizing the penetration level of photovoltaic units (PVUs) to achieve both engineering and economic benefits in a standard distribution power system. The Mirage Search Optimization (MSO) and the Pied Kingfisher Optimizer (PKO) are applied to minimize the total active power loss (TRPL) in the IEEE 69-node system. Two cases are considered: Case 1, where PVUs inject only active power, and Case 2, where PVUs inject both active and reactive power. The results demonstrate that PKO outperforms MSO and several metaheuristic algorithms reported in the literature. In Case 2, the optimal PVU penetration level of 67.17% significantly reduces TRPL compared with Case 1. The effectiveness of this optimized penetration level is further evaluated by comparing it with four other penetration levels: 25%, 50%, 75%, and 100%. PKO is then used to optimize the 24 h energy cost considering load variation and dynamic PV generation during four months of the year, including December, September, June, and March, which are ordered by increasing solar radiation across seasons under Vietnam’s climatic conditions. The results show that although the 75% penetration level slightly reduces the energy purchasing cost compared with the optimal level, it requires higher power capacity. Therefore, the optimized penetration level of 67.17% provides a balanced solution for reducing power losses while maintaining economic efficiency.
This paper proposes the application of artificial intelligence to forecast the generation capacity of wind power plants by processing data through noise reduction and filtering. It subsequently employs Long-Short Term Memory (LSTM) and Gated Recurrent Unit (GRU) neural networks for training, testing, and evaluation. Processing the initial data will help minimize noise and reduce the data space. The study focuses on preprocessing methods and selecting the appropriate neural network between LSTM and GRU. The initial data processing will assess the similarity through the Spearman rank correlation coefficient. The data used in the paper is taken from local wind turbines. The processed data will be input into the neural network for evaluation based on Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Percent Error (PE), and training time. The entire network simulation and evaluation process is performed using MATLAB software. The simulation results show the feasibility and suitability of the GRU network model combined with noise filtering methods, bringing high accuracy and less training time compared to the LSTM network. Specifically, the analysis contrasting the GRU network with the optimized dataset and the LSTM with the unprocessed dataset is more effective than a difference in RMSE of 15.592 and MAPE of 611.047%. Moreover, the time difference between the GRU and LSTM networks with the same dataset has a much earlier time difference from 6 to 28 seconds.