Research On Dynamic Compensation Method Of Pressure Sensor Based On Brain Storm Optimization

Shock wave signal is a typical non-stationary random signal,which has the characteristics of wide signal frequency range,fast amplitude change speed,short duration and so on.Therefore,it has high requirements on the dynamic performance of the shock wave test system.In order to solve the problem of dynamic errors due to insufficient dynamic characteristics of the pressure sensor during the shock wave test process,this thesis introduces a brainstorming algorithm suitable for solving multi-extremity problems.The method of directly obtaining the compensation system transfer function is used to dynamically compensate the pressure sensor..The main tasks are as follows:1)Improve the brainstorming algorithm.Aiming at the problem that the original brainstorming algorithm solves complex optimization problems,the global search and local search capabilities are weak,and it is easy to fall into local extreme values.This thesis proposes two improvements to the brainstorming algorithm.The probability parameter that determines the population selection strategy in the original brainstorming algorithm was changed from a fixed value to a monotonically increasing function.The global search and local search of the algorithm were balanced,and the algorithm search ability was enhanced.It also improves the individual fusion method in the algorithm,adds a pre-judgment mechanism,guarantees to improve the influence of high-quality individuals based on the fusion process,and improves the algorithm’s optimization accuracy and later convergence speed.Four typical high-dimensional multimodal test functions were used to test the search ability of particle swarm algorithm,brainstorming algorithm and improved brainstorming algorithm.The test results show that the improved brainstorming algorithm has better optimization accuracy,which verifies the effectiveness of the improvement.2)Apply the improved brainstorming algorithm to the pressure sensor dynamic compensation system.The pressure sensor used in this thesis is a 15 psi sensor from Endevoc.The shock tube test data is used as the input of the sensor dynamic compensation system,and the standard step signal is used as the output.The improved brainstorming algorithm is used to optimize the transfer system parameter combination of the compensation system to obtain The best dynamic compensation model.After the compensation model of the shock tube calibration data,the output signal overshoot is reduced from 54.7% to 9.2%,and the rise time is increased to 12 μs.This compensation model is used to dynamically compensate the measured shock wave data of the artillery muzzle.After compensation,the main characteristic parameters such as the peak pressure of overpressure and the time of positive pressure are more clear and accurate,which effectively suppresses the resonance frequency and improves the test during the weapon test.Precision.