No abstract

No abstract

In this paper, a proposed moth-flame optimization (MFO) technique has been investigated for obtaining an accurate simulation of the non-uniform electric field represented by a needle-to-plane gap configuration. The needle electrode is connected to the high-voltage (HV) terminal, while the earthed terminal is connected to the plane electrode. In addition to the non-uniformity of the field, a transverse dielectric barrier has been presented and investigated along the gap with a different thickness and location. The MFO works to optimize the error given by a numerical equation published before for calculating this field problem in the presence of a transverse barrier. This numerical equation was based on a correction coefficient called (β), which is dependant on three values, relative permittivity, barrier location, and barrier thickness. The MFO is working to minimize the error given by β using two new optimization factors in the β equation. To ensure the accurate validation of MFO with a minimum error for field problem simulation, various artificial intelligence (AI) optimization techniques have been compared with the MFO obtained results. The comparative study shows that MFO is more effective, especially at 30% of the gap length from the HV electrode which represents the region of highly non-uniform field along the gap configuration. The numerical results of the field simulation that are held by different types of AI techniques are compared with those obtained from the accurate simulation results using the finite-element method. The value of the error between the numerical and simulation results shows that MFO is the most effective optimization techniques that can be used in the numerical equation to obtain the best value of the correction factor. With MFO, good agreement has been reached between the proposed numerical equation and the accurate simulation values of the electric field problem. INDEX TERMS Artificial intelligence, moth-flame optimization (MFO), particle swarm optimization (PSO), genetic algorithm (GA), dielectric barrier, electric field simulation, finite element method.

The integration of renewable energy sources (RESs) has become more attractive to provide electricity to rural and remote areas, which increases the reliability and sustainability of the electrical system, particularly for areas where electricity extension is difficult. Despite this, the integration of hybrid RESs is accompanied by many problems as a result of the intermittent and unstable nature of RESs. The extant literature has discussed the integration of RESs, but it is not comprehensive enough to clarify all the factors that affect the integration of RESs. In this paper, a comprehensive review is made of the integration of RESs. This review includes various combinations of integrated systems, integration schemes, integration requirements, microgrid communication challenges, as well as artificial intelligence used in the integration. In addition, the review comprehensively presents the potential challenges arising from integrating renewable resources with the grid and the control strategies used. The classifications developed in this review facilitate the integration improvement process. This paper also discusses the various optimization techniques used to reduce the total cost of integrated energy sources. In addition, it examines the use of up-to-date methods to improve the performance of the electrical grid. A case study is conducted to analyze the impact of using artificial intelligence when integrating RESs. The results of the case study prove that the use of artificial intelligence helps to improve the accuracy of operation to provide effective and accurate prediction control of the integrated system. Various optimization techniques are combined with ANN to select the best hybrid model. PSO has the fast convergence rate for reaching to the minimum errors as the Normalized Mean Square Error (NMSE) percentage reaches 1.10% in 3367.50 s.

No abstract