IDENTIFICATION AND COMPENSATION OF A CAPACITIVE DIFFERENTIAL PRESSURE SENSOR BASED ON SUPPORT VECTOR REGRESSION USING PARTICLE SWARM OPTIMIZATION

Capacitive Differential Pressure Sensors (CDPSs) are highly utilized in industry. However, the accuracy of CDPSs is limited because of the adverse effects of ambient temperature on their output characteristics. In this paper, the effect of temperature on a CDPS output is identified and compensated for using a Support Vector Machine for Regression (SVR) method. To achieve a better performance, a Particle Swarm Optimization (PSO) method is employed to optimize the parameters of SVR. Also, a test bench is designed and implemented to obtain data under real environmental conditions. The experimental results obtained verify the performance of modeling and compensation for the non-linear behavior of the CDPS based on SVR using PSO. Simulation results show that the proposed identifier and compensator estimates and compensates the output accurately. Finally, the performances of the proposed methods are also compared with those of Artificial Neural Network (ANN) techniques.