基于神经网络的硫化仪温控改造

Temperature Control of Vulcanization Based on Neural Network

硫化仪是一种广泛配备使用的橡胶检测仪器,其传统机型多采用固定参数的PID控制,存在升温曲线多震荡及控温速度较慢的问题;并且对于固定参数的传统控制方法来说,在不同环境条件下的控制匹配效果不同。现采用PSO粒子群算法改进BP神经网络,用神经网络计算、控制PID参数进而控制模腔。实验结果表明:采用BP-PID控制的硫化仪模腔达到同样150℃的设定温度,比传统PID控制的模腔加热速度提高36.5%;且针对不同设定温度,此控制方法升温曲线波动平均降低22%,具有很好的普适性。

The Vulcanizer is a widely used rubber testing instrument.The traditional models of vulcanizers mostly use PID control with fixed parameters,which has the problems of many oscillations in the heating curve and slow temperature control speed,and for the control method with fixed parameters,the control matching effect under different environmental conditions is different.The PSO particle swarm algorithm is now used to improve the BP neural network,and the neural network is used to calculate and control the PID parameters and then control the mold cavity.The experimental results show that the mold cavity of the curing apparatus controlled by BP-PID reaches the same set temperature of 150℃,which is 36.5%higher than the heating speed of the mold cavity controlled by traditional PID,and for different set temperatures,the heating curve of this control method fluctuates evenly.It is reduced by 22%,which has good universality.