Economic Load Dispatching Based On Chaos Genetic Algorithm And Fuzzy Decision

Market-orientation of electric power is the trends of global power industry development and innovation as well. With the performance of electric market reform and the greater difference between maximum and minimum of load in power network, it has great value to research the economic load dispatching (ELD) on the background of three scientific research projects.Chaos genetic algorithm and fuzzy decision (CGAFD) which can obtain optimal unit commitment (UC) and load distribution (LD) simultaneously is presented. The advantages of improved priority list, heuristic genetic algorithm, chaotic optimization, and fuzzy decision is combined.In order to solve the problem of LD, first of all, economic LD model is presented. And then, the LD model based on speediness is combined, and optimal LD model based on economy and speediness via having dimensionless disposal to aim functions is presented. Economy and speediness index can be changed rapidly by adjusting weight coefficient. Finally, the characteristic of parallel searching of genetic algorithm (GA) is used to solve the LD problem. Meanwhile, searching at the best point's neighborhood by using chaotic optimization can avoid GA trapping into local minimum, and effectively accelerate the speed of convergence.In order to solve the problem of UC, firstly, the model is presented, which contains the cost of unit start and stop and kinds of constraints, such as power balance constraint, the minimum running time and minimum halting time, start and stop times, etc. Furthermore, UC is determined by heuristic GA on the basis of using improved priority list to confirm the unit sequence in different periods of time. Meanwhile, the crossover rate and mutation rate of GA is controlled by fuzzy decision.This thesis uses MATLAB to design algorithm programs of ELD, and the algorithm is used in several examples. Simulation results show that not only the algorithm tackles various constraints of ELD very well, but also it can largely reduce the number of unfeasible solutions, and greatly improve convergence rate. Moreover, there is no special demand on the fuel-cost curve, which ensures the precision of the solution.It is feasible to solve the problem of optimal UC and LD in the environment of electric power market by CGAFD. Moreover, all kinds of economic indicators can be computed, which is helpful to take part in the competition of electric market for power generation enterprise.