基于最小二乘法与RBF神经网络的溶解氧检测系统设计

为克服温度对溶解氧传感器的影响，对极谱型溶解氧检测系统的溶解氧电极激励源、高精度信号采样、软件标定和温度补偿等方面进行研究。通过对极谱型溶解氧传感器工作原理进行分析，设计了极谱型溶解氧传感器检测电路；根据溶解氧电极的温度特性，设计了基于NTC（负温度系数）热敏电阻的硬件温度补偿电路，并利用最小二乘法及RBF神经网络构建了软件温度补偿模型。利用饱和蒸馏水进行温度补偿实验，结果表明：经温度补偿后，该溶解氧检测系统的相对误差及采样波动均在1%以内，大大减小了传感器的非线性误差，测量精度和稳定性均可满足应用要求。

Design of the dissolved oxygen detection system based on the least square method and RBF neural network

A polarographic dissolved oxygen detection system is studied in this paper to understand the dissolved oxygen electrode excitation source, high-precision signal sampling, software calibration, and temperature compensation for overcoming the influence of temperature on the dissolved oxygen sensor during dissolved oxygen detection. Further, a polarographic dissolved oxygen sensor detection circuit is designed by analyzing the working principle of the polarographic dissolved oxygen sensor. Subsequently, a hardware temperature compensation circuit was designed based on the NTC (negative temperature coefficient) thermistor according to the temperature characteristics of the dissolved oxygen electrode. The software temperature compensation model is established using the least squares method and the RBF neural network. The temperature compensation experiments are conducted using saturated distilled water. The experimental results prove that the relative error and sampling fluctuation of the dissolved oxygen detection system are within 1% after temperature compensation, considerably reducing the nonlinear error associated with the sensor. The measurement accuracy and stability can satisfy the application requirements.