基于迭代局部搜索和自适应粒子群优化的SVM短期负荷预测

为了能够进一步的提高船舶电力负荷的预测精度,针对SVM模型在负荷预测中存在的参数选取问题,该研究提出了一种新的参数优化算法:基于迭代局部搜索和自适应粒子群优化的组合算法。自适应粒子群算法提高了传统粒子群算法的收敛速度,引入的迭代局部搜索思想,配合新的极值评价标准能够很好的解决粒子群算法容易陷入局部最优的问题。仿真结果表明,利用新的参数优化算法使得SVM预测模型的精度得到了很大的提高。

SVM short-term load forecasting based on the iterated local search and adaptive particle swarm optimization

In order to further improve the prediction accuracy of ship power load, this paper introduces a new parameter optimization algorithm aiming at the parameter selection of SVM: combination algorithm based on the iterated local search and adaptive particle swarm optimization. Adaptive particle swarm algorithm improves the traditional particle swarm algorithm convergence speed, iterated local search and the new extreme value evaluation criteria can well solve the problem that the particle swarm algorithm is easy to fall into local optimum. The simulation results show that the precision of SVM is improved a lot using the new parameters optimization algorithm.