Search Results (5)

With the rapid development of radio technology and its widespread application in the military field, the electromagnetic environment in which radar communication operates is becoming increasingly complex. Among them, human radio interference makes radar countermeasures increasingly fierce. This requires radar systems to have strong capabilities in resisting electronic interference, anti-radiation missiles, and radar detection. However, array antennas are one of the effective means to solve these problems. In recent years, array antennas have been extensively utilized in various fields, including radar, sonar, and wireless communication. Many evolutionary algorithms have been employed to optimize the size and phase of array elements, as well as adjust the spacing between them, to achieve the desired antenna pattern. The main objective is to enhance useful signals while suppressing interference signals. In this paper, we introduce the dandelion optimization (DO) algorithm, a newly developed swarm intelligence optimization algorithm that simulates the growth and reproduction of natural dandelions. To address the issues of low precision and slow convergence of the DO algorithm, we propose an improved version called the chaos exchange nonlinear dandelion optimization (CENDO) algorithm. The CENDO algorithm aims to optimize the spacing of antenna array elements in order to achieve a low sidelobe level (SLL) and deep nulls antenna pattern. In order to test the performance of the CENDO algorithm in solving the problem of comprehensive optimization of non-equidistant antenna array patterns, five experimental simulation examples are conducted. In Experiment Simulation Example 1, Experiment Simulation Example 2, and Experiment Simulation Example 3, the optimization objective is to reduce the SLL of non-equidistant arrays. The CENDO algorithm is compared with DO, particle swarm optimization (PSO), the quadratic penalty function method (QPM), based on hybrid particle swarm optimization and the gravity search algorithm (PSOGSA), the whale optimization algorithm (WOA), the grasshopper optimization algorithm (GOA), the sparrow search algorithm (SSA), the multi-objective sparrow search optimization algorithm (MSSA), the runner-root algorithm (RRA), and the cat swarm optimization (CSO) algorithms. In the three examples above, the SLLs obtained using the CENDO algorithm optimization are all the lowest. The above three examples all demonstrate that the improved CENDO algorithm performs better in reducing the SLL of non-equidistant antenna arrays. In Experiment Simulation Example 4 and In Experiment Simulation Example 5, the optimization objective is to reduce the SLL of a non-uniform array and generate some deep nulls in a specified direction. The CENDO algorithm is compared with the DO algorithm, PSO algorithm, CSO algorithm, pelican optimization algorithm (POA), and grey wolf optimizer (GWO) algorithm. In the two examples above, optimizing the antenna array using the CENDO algorithm not only results in the lowest SLL but also in the deepest zeros. The above examples both demonstrate that the improved CENDO algorithm has better optimization performance in simultaneously reducing the SLL of non-equidistant antenna arrays and reducing the null depth problem. In summary, the simulation results of five experiments show that the CENDO algorithm has better optimization ability in the comprehensive optimization problem of non-equidistant antenna array patterns than all the algorithms compared above. Therefore, it can be regarded as a strong candidate to solve problems in the field of electromagnetism. Full article

In this paper, a novel hybrid Maximum Power Point Tracking (MPPT) algorithm using Particle-Swarm-Optimization-trained machine learning and Flying Squirrel Search Optimization (PSO_ML-FSSO) has been proposed to obtain the optimal efficiency for solar PV systems. The proposed algorithm was compared with other well-known methods viz. Perturb & Observer (P&O), Incremental Conductance (INC), Particle Swarm Optimization (PSO), Cuckoo Search Optimization (CSO), Flower Pollen Algorithm (FPA), Gray Wolf Optimization (GWO), Neural-Network-trained Machine Learning (NN_ML), Genetic Algorithm (GA), and PSO-trained Machine Learning. The proposed algorithm was modelled in the MATLAB/Simulink environment under different operating conditions, for example, with step changes in temperature, solar irradiance, and partial shading. The proposed algorithm improved the efficiency up to 0.72% and reduced the settling time up to 76.4%. The findings of the research highlight that PSO_ML-FSSO is a potential approach that outperforms all other well-known algorithms tested herein for solar PV systems. Full article

Accurate short-term load forecasting is of great significance to the safe and stable operation of power systems and the development of the power market. Most existing studies apply deep learning models to make predictions considering only one feature or temporal relationship in load time series. Therefore, to obtain an accurate and reliable prediction result, a hybrid prediction model combining a dual-stage attention mechanism (DA), crisscross grey wolf optimizer (CS-GWO) and bidirectional gated recurrent unit (BiGRU) is proposed in this paper. DA is introduced on the input side of the model to improve the sensitivity of the model to key features and information at key time points simultaneously. CS-GWO is formed by combining the horizontal and vertical crossover operators, to enhance the global search ability and the diversity of the population of GWO. Meanwhile, BiGRU is optimized by CS-GWO to accelerate the convergence of the model. Finally, a collected load dataset, four evaluation metrics and parametric and non-parametric testing manners are used to evaluate the proposed CS-GWO-DA-BiGRU short-term load prediction model. The experimental results show that the RMSE, MAE and SMAPE are reduced respectively by 3.86%, 1.37% and 0.30% of those of the second-best performing CSO-DA-BiGRU model, which demonstrates that the proposed model can better fit the load data and achieve better prediction results. Full article

Recent advances in electric grid technology have led to sustainable, modern, decentralized, bidirectional microgrids (MGs). The MGs can support energy storage, renewable energy sources (RESs), power electronics converters, and energy management systems. The MG system is less costly and creates less CO₂ than traditional power systems, which have significant operational and fuel expenses. In this paper, the proposed hybrid MG adopts renewable energies, including solar photovoltaic (PV), wind turbines (WT), biomass gasifiers (biogasifier), batteries’ storage energies, and a backup diesel generator. The energy management system of the adopted MG resources is intended to satisfy the load demand of Basra, a city in southern Iraq, considering the city’s real climate and demand data. For optimal sizing of the proposed MG components, a meta-heuristic optimization algorithm (Hybrid Grey Wolf with Cuckoo Search Optimization (GWCSO)) is applied. The simulation results are compared with those achieved using Particle Swarm Optimization (PSO), Genetic Algorithms (GA), Grey Wolf Optimization (GWO), Cuckoo Search Optimization (CSO), and Antlion Optimization (ALO) to evaluate the optimal sizing results with minimum costs. Since the adopted GWCSO has the lowest deviation, it is more robust than the other algorithms, and their optimal number of component units, annual cost, and Levelized Cost Of Energy (LCOE) are superior to the other ones. According to the optimal annual analysis, LCOE is 0.1192 and the overall system will cost about USD 2.6918 billion. Full article

Solar irradiance forecasting is an inevitable and most significant process in grid-connected photovoltaic systems. Solar power is highly non-linear, and thus to manage the grid operation efficiently, with irradiance forecasting for various timescales, such as an hour ahead, a day ahead, and a week ahead, strategies are developed and analysed in this article. However, the single time scale model can perform better for that specific time scale but cannot be employed for other time scale forecasting. Moreover, the data consideration for single time scale forecasting is limited. In this work, a multi-time scale model for solar irradiance forecasting is proposed based on the multi-task learning algorithm. An effective resource sharing scheme between each task is presented. The proposed multi-task learning algorithm is implemented with a long short-term memory (LSTM) neural network model and the performance is investigated for various time scale forecasting. The hyperparameter estimation of the proposed LSTM model is made by a hybrid chicken swarm optimizer based on combining the best features of both the chicken swarm optimization algorithm (CSO) and grey wolf optimization (GWO) algorithm. The proposed model is validated, comparing existing methodologies for single timescale forecasting, and the proposed strategy demonstrated highly consistent performance for all time scale forecasting with improved metric results. Full article