工业高精度压力变送器非线性温度影响补偿方法研究

Research on compensation method for nonlinear temperature effect of industrial high precision pressure transmitters

温度的大幅度波动会导致工业高精度压力变送器中的材料膨胀或收缩,引起压力变送器的输出值产生漂移。为了保证压力变送器的准确性和稳定性,提出一种基于BP神经网络的智能化温度影响补偿方法。分析温度对压力变动器的非线性影响,计算出零点温度漂移系数和灵敏度系数。通过函数特征化处理标准输出-压力曲线,根据零点温度漂移系数和灵敏度系数计算标准输出-压力曲线的误差,利用三阶多项式反函数对标准输出-压力曲线的误差进行线性化处理误差,用于校正压力变动器输出值。通过正反向传播算法多次训练反向传播神经网络(Back Propagation Neural Network,BP)修正各层间连接权重值,将校正后的压力变动器的输出值输入到已经训练好的网络中,待结果小于预设误差阈值或者达到最大迭代次数时,输出温度影响补偿结果。经模拟实验证明,所提方法可在-20℃~60℃区间内,妥善补偿因温度剧烈变化导致的压力变送器误差值,补偿后最大相对误差仅为0.5%,补偿精度高,效果好。

Large fluctuations in temperature can cause material expansion or contraction in industrial high-precision pressure transmitters,leading to drift in the output value of the pressure transmitter.To ensure the accuracy and stability of pressure transmitters,an intelligent temperature impact compensation method based on BP neural network is proposed.Analyze the nonlinear effect of temperature on the pressure regulator,calculate the zero temperature drift coefficient and sensitivity coefficient,process the standard output pressure curve through function characterization,calculate the error of the standard output pressure curve according to the zero temperature drift coefficient and sensitivity coefficient,and use the third-order polynomial inverse function to linearize the error of the standard output pressure curve to correct the output value of the pressure regulator.The back propagation neural network(BP)is trained many times through the forward Backpropagation to correct the connection weight value between layers,and the output value of the corrected pressure regulator is input into the trained network.When the result is less than the preset error threshold or reaches the maximum number of iterations,the output temperature affects the compensation result.Through simulation experiments,it has been proven that the proposed method can effectively compensate for the error value of the pressure transmitter caused by drastic temperature changes within the range of-20℃-60℃.After compensation,the maximum relative error is only 0.5%,with high compensation accuracy and good effect.