RBF神经网络及其在液化气生产数据预测中的应用

RBF Neural Network and Its Application in Predication Soft Measurement of LPG Procreative Data

针对化工生产中响应值测定滞后的问题,应用径向基函数(RBF,RadialBasisFunction)神经网络建立生产数据预测数学模型.将该模型用于液化气生产数据预测,处理结果相对误差均小于5%,表明该模型预测效果良好.

According to determination lag of the chemical industrial data,we have applied RBF(Radial basis function)neural network to erect predicated math-model and applied the math-model in LPG procreative data. The results indicate that the predicated effect is very good and relative error is smaller than five percent.