A New Modified Artificial Bee Colony Algorithm And Its Applications On Economic Load Dispatch Problems

Economic load dispatch(ELD) of power system is a classic optimizing problem of power system, power supply enterprises can use this optimizing technology to maximize their benefits, which is of great practical significance. Along with the environmental problems caused by SO2, NOX and other pollutants from coal combustion in power generating units is becoming more and more serious, the serious environmental problems cause people strong concern. Therefore, in the study of economic load dispatch of power system problems, researchers not only consider the cost of power generation, but also consider the cost of environmental pollution in the process of power generation. This is so-called environmental economic load dispatch(EELD) problem in power system.Artificial bee colony(ABC) algorithm is a swarm intelligence optimizing technology based on simulating the foraging behavior mechanism and internal labor division of the bee colony. It has incurred widespread concern in academic areas since it was proposed by Turkey scholar Karaboga at 2005. ABC algorithm has turned out to be one of the powerful evolutionary algorithms with the advantages such as simple operation, less control parameters, high precision and strong robustness, and ABC has been successfully used in many fields now.In this paper, based on the domestic and foreign research of power system load dispatch, the coal consumption characteristic of generator have been analyzed. Incorporating the coal consumption characteristics and pollution cost, the mathematical models of the single objective economic load dispatch problem and multi-objective environmental economic load dispatch problem have been established. Aiming at the shortcomings of artificial bee colony algorithms, such as the low convergence rate at the early stage and the poor searching ability at the late stage, the scheme is proposed to dynamically adjust the population distribution between onlooker bees and employed bees. Based on the standard artificial bee colony algorithm, the quality of scout bee is improved by weighted average of abandoned nectar has been proposed by analyzing the effect of the employed bee in the searching of the algorithm. This is Artificial Bee Colony Algorithm based on Dynamic Adjustment of Population Distribution(DA-ABC). The performance of DA-ABC is validated by the standard test functions, single-objective economic load dispatch examples of 3 units, 13 units and 40 units, and multi-objective environmental economic load dispatch problem of 6 units. The optimization results show that the DA-ABC algorithm has the advantages such as high convergence rate, strong ability of exploring and searching at the later stage, and effectively avoids the premature phenomenon.