Open AccessError Correction of Weak Current Measurement System Based on Wavelet Denoising and Generalized Regression Neural Network
Author(s) -
Dapeng Wen,
Xiyin Liang,
Maogen Su,
Meng Meng Wu,
Ruilin Chen,
Tianchen Zhang
Publication year - 2021
Publication title -
instrumentation, mesure, métrologie/instrumentation mesure métrologie
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.161
H-Index - 4
eISSN - 2269-8485
pISSN - 1631-4670
DOI - 10.18280/i2m.200205
Subject(s) - mean squared error , artificial neural network , noise reduction , interference (communication) , wavelet , algorithm , observational error , noise (video) , current (fluid) , computer science , error detection and correction , generalization , backpropagation , mathematics , pattern recognition (psychology) , channel (broadcasting) , statistics , artificial intelligence , engineering , telecommunications , electrical engineering , mathematical analysis , image (mathematics)
Aiming at the problems that the weak current signal circuit is susceptible to noise interference and leakage current at the input terminal affects the measurement accuracy, a weak current measurement error correction scheme based on the combination of wavelet threshold denoising and generalized regression neural network (GRNN) was proposed. The scheme was applied to the error correction of multi-channel weak current measurement system based on the ADAS1134 chip: the wavelet threshold denoising was used to preprocess the original current data measured by the system and the current measurement value was corrected after the system measurement error correction model established with GRNN was constructed. Compared with the correction method based on least square method and back propagation neural network (BPNN), this method has many advantages such as high accuracy, anti-interference ability and strong generalization ability. The experimental results showed that RMSE=0.0911 nA, MAE=0.0354 nA, and MAPE=0.0078%, without increasing the complexity of the measurement circuit, which achieved the purpose of correcting the measurement error of the weak current measurement system.