应用PSO算法改进Elman神经网络的双压凝汽器真空预测

Application of PSO algorithm-modified Elman neural network in vacuum prediction for dual-pressure condensers

为实现对凝汽器真空的优化控制,引入一种采用粒子群优化(PSO)算法改进的Elman神经网络,建立双压凝汽器真空预测模型,提出对双压凝汽器高、低压侧真空分别进行预测计算,将该模型应用于某600 MW机组的双压凝汽器真空预测,并与普通算法改进的Elman神经网络的预测结果进行比较。结果表明:采用PSO算法改进的Elman神经网络对双压凝汽器高、低压侧真空预测的收敛速度更快、精确度更高,是一种行之有效的双压凝汽器真空预测模型。

An Elman neural network optimized by particle swarm optimization（ PSO） algorithm was introduced to establish the vacuum prediction model for dual-pressure condensers. The calculation model which can forecast both the high pressure and low pressure side of the dual-pressure condensers was proposed. Moreover,the above model was applied in dual-pressure condenser in a 600 MW unit and the results were compared with that predicted by the common algorithm-modified Elman neural network. The results show that,this Elman neural network optimized by PSO algorithm has faster convergence speed and higher accuracy,which is a feasible model for vaccum prediction in dual-pressure condensers.