Simulation of temperature compensation of pressure sensor based on PCA and improved BP neural network

Considering the effects of temperature on output of silicon pressure sensor, this paper proposed a novel method for analyzing temperature compensation of pressure sensor by using the combination of Principal Component Analysis (PCA) and improved BP neural networks. By using PCA to extract the prime information of temperature compensation, the multi-dimensional problem is simplified, the noise error data is eliminated, the neural network is improved and the fault-tolerance capability is enhanced. The results indicate that this method can restrain the effects of temperature on pressure sensors effectively and enhance their stability and accuracy.