太阳总辐射传感器温漂补偿的RB F网络模型

摘要

太阳总辐射传感器的温度漂移是影响精度的主要原因之一。针对太阳总辐射传感器温漂的非线性，建立了太阳总辐射传感器温漂补偿的RBF网络模型，设计了使用24位高精度模数转换器ADS1248和STM32F103VET6组成的采集电路并对太阳总辐射传感器数据进行了补偿。结果表明，RBF神经网络具有很强的逼近非线性函数和自学的能力，能够对太阳总辐射传感器的温漂误差进行修正，补偿后精度有了明显提高。%The temperature drift of total solar radiation sensor is one of the main factors influencing the accuracy. According to the nonlinearity of total solar radiation sensor temperature drift,the RBF network model is estab-lished for the total solar radiation sensor temperature drift compensation.We design acquisition circuit composed of 24 high-precision analog-to-digital converter ADS1248 and STM32F103VET6 and compensate for the total so-lar radiation sensor data.The experimental results show that the RBF neural network has a function of nonlinear approximation and self-learning ability,it is capable of the total solar radiation sensor temperature drift error cor-rection,and the accuracy is obviously improved after compensation.