Application of Improved Particle Swarm Optimization Algorithm in Medium and Long Term power Load Combination Forecasting

There are many methods of power load forecasting, but each method has its own inaccurate influence factors, and the result of the single method is relatively large. Medium and long term power load forecasting influence the development of local planning in the future, so the accuracy of prediction is higher. Based on the two order exponential smoothing, regression analysis and grey prediction model, the comprehensive prediction model based on the three single forecasting methods is built. By using the inertia weight of the particle swarm optimization algorithm to determine the weights, the advantage of the improved combination forecast is obtained by comparison. Introduction Power load forecasting is the premise of power system dispatching, safe and stable operation. Accurate power load forecasting can reasonably arrange the operation of the internal engine group of the power grid, and reasonable arrangement of the unit maintenance plans. It is advantageous to improve the scientific planning of regional power grid, and promote the effective utilization of resources to optimize the allocation of regional power energy. At present, the methods of load forecasting are many, and the advantages of each forecasting method are complementary to each other. Through the analysis of the particle motion state, it is concluded that the convergence of the particle swarm can reach the best condition, improve the convergence speed and the global convergence, and reduce the possibility of falling into the local optimum. Particle Swarm Optimization Algorithm Particle swarm optimization (PSO) is a kind of global optimization technology based on biological intelligence, which is inspired by the foraging behavior of birds. The solution space is searched through the mutual sharing information among the particles, and the optimal solution is found. The formula of particle search for PSO algorithm is: Searched through the mutual sharing information among the particles, and the optimal solution is found. The formula of particle search for PSO algorithm is: 1 1 1 2 2 ( ) ( ) n n n n n n i i i i i i i v wv c r P X c r G X + = + × × − + × × − 1 1 n n n i i i X X v + + = + Type: is w Inertial factor, 1 r and 2 r is the random number of random distribution on the interval(0,1). 1i c , 2i c are learning factors, n i x is the space position of the particle n in the iteration of I, n i v velocity of particles, n i p is the individual extreme value generated during the search process. n G is the global extreme. Particle Swarm Optimization Algorithm Influence of inertia weight w on particle swarm optimization embodied in the larger value is advantageous to jump out the local optimum and continue to carry on the overall optimization. The 2nd International Conference on Economics, Management Engineering and Education Technology (ICEMEET 2016) Copyright © 2017, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). Advances in Social Science, Education and Humanities Research, volume 87