基于遗传算法的RBF神经网络在铂电阻温度传感器非线性补偿中的应用

Application of RBF Neural Network Based on Genetic Algorithm to Non-linearity Compensation for Platinum Resistance Thermometer

针对铂电阻温度传感器在实际应用中存在非线性问题,提出了基于遗传算法优化径向基函数(RBF)神经网络实现其非线性补偿的方法。分析了非线性补偿原理,设计了RBF神经网络补偿器,并引入遗传算法优化神经网络结构和参数。实验结果表明,所提出的铂电阻温度传感器非线性补偿方法是实用和可行的。

Being aimed at the non-linearity problem existed in practical use of platinum resistance thermometer, the paper presented a method to compensate non-linearity of platinum resistance thermometer based on genetic algorithm optimized radial basis function （ RBF） neural network. The paper analyzed the principle of non-linearity, designed RBF neural network compensator, and led in the optimized structure and parameters of the RBF neural network. The experiment results show that the non-linearity compensation method for platinum resistance is practical and feasible. [ Ch,4 fig. 1 tab. 10ref. ]