基于动态双组粒子群的短期负荷预测

Short-term load forecasting based on dynamic bi- group particle swarm

摘要:

为提高电网短期负荷预测的精度，提出一种有效的优化支持向量机参数的算法。该算法首先将初始粒子群适应度排序，然后根据适应度的大小将初始粒子群划分为两组，并同时运用不同的权重进行全局搜索和局部搜索。前期，全局搜索的粒子群数量远多于局部搜索，且使用全局搜索能力强的较大的惯性权重；局部搜索的粒子群使用较小的惯性权重。随着迭代次数的增加，全局搜索的粒子群数量不断减少，局部搜索不断增多，两组粒子数量动态变化。并且引入平均粒距和适应度方差解决粒子群容易陷入局部最优这一问题，最后用改进的动态双组粒子群算法优化最小二乘支持向量机的参数用于短期负荷预测，实验结果表明该方法预测精度更高，可行且有效。

Abstract:

In order to improve the accuracy of short-term load forecasting, this paper presents an efficient algorithm for parameter optimization of Support Vector Machine. The algorithm first sorts the initial particle swarm fitness, and then divides the initial particle swarm into two groups according to the size of the fitness, and simultaneously uses the different weights for global search and local search. The number of global search particles is much larger than that of local search, and the larger global inertia weight is used. The local search particle group uses a smaller inertia weight. With the increase of the number of iterations, the number of global search particles is decreasing, the number of local search is increasing, the number of particles of two groups is dynamically changing. And the average particle size and fitness variance are introduced to solve the problem that the particle group is easy to fall into the local optimum. Finally, the improved dynamic two group particle swarm optimization algorithm is used to optimize the parameters of the least squares support vector machines for short-term load forecasting. The experimental results show that the proposed method has higher prediction accuracy and is feasible and effective.