The Study Of Sensor Nonlinear Distortion Correction Based On Artificial Neural Network

Sensor is the core component in the measuring systems, and its accuracy plays a decisive role in the performance of measuring systems. The sensor nonlinear distortion correction is one of the main techniques to enhance the performance of sensors, which has already become an important subject researched at home and abroad at the present time. The main task of this paper is the sensor nonlinear distortion correction. A new nonlinear distortion correction method based on artificial neural network is presented.Sensor nonlinear distortion correction in the tradition is introduced briefly in this paper, and the feasibility of the ANN scheme is demonstrated by theoretical analysis. The model of the nonlinear distortion correction has been established. Compared with the traditional method of nonlinear distortion correction, the new method reduces the complexity of the system. Here the BP network algorithm is used, the calculation process and characteristics of the BP network algorithm is described in detail. The need to pay attention to some issues of the application is analyzed in the process of modeling, such as the choice of hidden nodes, permission error, etc.S3C2410X embedded system as BP network carrier is selected in the project, the construction of the entire hardware system is discussed in detail. Namely the sensor output AC signals are transformed to DC signals by AD637(TRMS-to-DC converter), then the DC signals are converted to digital signals by A/D converter, and corrected by improved BP network on S3C2410X platform, tried to approximate the real value. Finally test data and results are displayed on the LCD. Software system that matches the hardware system is divided into training and testing processes. First of all, BP network training, the number of training samples is 19, and the teacher signals are TRMS value of the output of Agilent 34401A(precision AC voltmeter). Then the outputs of sensor are corrected by trained BP network, and compared with the outputs of sensor without BP network.The results of experiment show that ANN correction method achieves good results by improved BP network to correct sensor nonlinear distortion. The method can also be used for other nonlinear distortion corrections, such as transmission system of Instruments, etc.