【摘要】 为了对地区电网220 kV线损率进行有效的评估,提出了一种基于自适应量子粒子群算法(AQPSO)优化的最小二乘支持向量机(LSSVM)的线损率预测模型。AQPSO在QPSO的基础上引入了遗传算法中的交叉与变异机制以扩大种群多样性,避免算法陷入局部最优。利用AQPSO搜索最佳的LSSVM参数并获取线损率预测结果,通过训练集对预测模型进行训练,然后使用测试集进行实验。最后选择地区电网23条220 kV线路的真实数据进行分析和预测,实验结果表明,文章所提出的AQPSO-LSSVM模型能够更有效地对线损率进行准确预测。更多还原

【Abstract】 In order to evaluate the line loss rate of 220 kV power grid effectively,a line loss rate prediction model based on adaptive quantum particle swarm optimization(AQPSO)and least square support vector machine(LSSVM)is proposed. AQPSO introduces the crossover and mutation mechanism in genetic algorithm on the basis of QPSO to expand the population diversity and avoid the algo-rithm from falling into the local optimum. The model uses AQPSO to search for the best LSSVM parameters and obtain line loss rate prediction results,and trains the prediction model through the training set,and then uses the test set for experiments. Finally,the article selects the real data of 23 lines of 220 kV regional power grid,for analysis and prediction,the experimental results show that the proposed AQPSO-LSSVM model can more effectively predict the line loss rate accurately.更多还原