基于IRIME-CNN-DD的硅压阻式压力传感器温度补偿模型OA

Targeting at the specific problem that silicon piezoresistive pressure sensors are sensitive to temperature,an improved rime optimization algorithm(IRIME)is proposed to optimize the temperature compensation model IRIME-CNN-DD for convolutional neural network(CNN)and dendritic network(DD).Firstly,the improved RIME algorithm is applied in the optimization of neural network parameters,i.e.,the position update formula is redefined by using categorical discussion and adaptive Laplace crossover operator strategies,which enhances the algorithm's optimality seeking ability.Secondly,in terms of temperature compensation of silicon piezoresistive pressure sensors,two models,CNN network and DD network,are fused to construct a temperature compensation model;that is,CNN network is used to extract the local features in the dataset,and DD network is used to perform further feature construction for the local features extracted by CNN network,and then the compensation results are output.Finally,in order to validate the model,the measured pressure data from the graduate laboratory of Guangdong Ocean University are brought into the model,and the results show that the root mean square error and the average absolute error of the model are 0.0825 kPa and 0.05844 kPa,respectively.The average absolute error of the model is reduced by 32.89%,35.28%and 38.69%,and the root mean square error is reduced by 30.46%,37.54%and 41.38%compared with PSO-CNN-DD,DE-SVM and GA-BP models.It can be seen that the model can effectively eliminate the influence of temperature on the sensor and significantly improve the detection accuracy of the sensor.