基于自适应粒子群优化LSSVM的变压器油中溶解气体浓度预测

Predicting method for dissolved gas in transformer oil based on improved self-adaptive PSO algorithm and LSSVM

为了实现油浸式变压器油中溶解气体的预测,进而达到变压器状态预警、降低事故发生率的目的,本文将自适应粒子群优化算法(IDPSO)和最小二乘支持向量机(LSSVM)相结合,建立变压器油中溶解气体预测模型。利用IDPSO算法基于种群的并行搜索策略特点来自适应迭代搜索最优的目标函数值,寻找LSSVM模型中的参数ξ、C和σ的最优取值,克服了应用传统支持向量机算法进行预测时凭主观经验选择参数对模型泛化能力和预测性能的影响。利用国网陕西省电力公司某变电站采集变压器油色谱数据进行实例验证,结果表明基于IDPSO优化的LSSVM算法具有较好的模式跟踪性能,且能够有效提高变压器油中溶解气体预测的预测精度。

In order to predict the condition of the dissolved gas in transformer oil and achieve the function of early warning and avoiding accidents,the improved self-adaptive PSO algorithm and least squares support vector machine are combined to establish the prediction model of transformer parameters.Firstly,the improved self-adaptive PSO algorithm is applied to acquire the optimal value of the characteristics in the objective function objective function by the parallel search strategy.After setting reasonable value of parametersξ,C andσ,LSSVM(least square support vector machine)is excellent in prediction and generalization ability.The experimental results show that the model proposed performs well both in tracks and prediction.