Nonlinear Correction Of Sensor Based On Genetic Algorithm And Support Vector Machine

Sensor is an important part in the test system, the quality of the performance and the reliability of the output signal play a vital role in the entire test system. In practical applications, the sensors can be susceptible to many environmental factors, such as temperature, magnetic field, noise and power fluctuations, which reduce the accuracy and result in poor stability of the system. Therefore, the nonlinear correction of the sensors, which improves the performance of the sensors and the accuracy of the test system, expands the measurement range, is of important significance.In order to overcome the disadvantages of several existing methods used in nonlinear correction of the sensor and the problems that the parameters of support vector machine are difficult to determine, a method combined with the strong search ability characteristics of genetic algorithm and based on genetic algorithm and support vector machine is presented. The sensor support vector machine model of the sensor nonlinear correction established. The realization process of genetic algorithm optimizing the parameters of support vector machine is introduced.In the implementation process, CYJ-101type pressure sensor is adjusted with the method of genetic algorithm and support vector machine through Matlab to verify its feasibility. The result of correction is compared with support vector machine and BP neural network in order to confirm the advantages of genetic algorithm and support vector machine. The experiment results show that this method can solve the problem that the parameters of support vector machine and its kernel function are difficult to select and get the optimal parameters compared with the method of support vector machine. Simultaneously, the experiment results show that BP neural network reduces the maximum relative fluctuation from the initial 22.2% to 1.12%.The method of genetic algorithm and support vector machine reduces it to 0.04%,evidently improved the performance of the sensor and achieved better results.The hardware and software of pressure sensor system are designed in this paper. Meanwhile, the method of support vector machine and genetic algorithm is used in the software for sensor nonlinear calibration and compensation. Thus, the method can be realized in practice.