Forecasting of the electrical actuators condition using stator's current signals

International Conference on Mechanical Engineering, Automation and Control Systems 2016

机械制造;无线电电子学;计算机科学

Kruglova, T.N.^1 ; Yaroshenko, I.V.^2 ; Rabotalov, N.N.^3 ; Melnikov, M.A.^3

Platov South-Russian State Technical University, 132, Prosveschenia Str., Novocherkassk, Rostov Region

346428, Russia^1

Diagnostic Systems and Complexes L.L.C., 132, Warsaw ave., Moscow

117519, Russia^2

HV Measurement Complexes and Systems L.L.C., 44/11 Balakovo, Trnavskaya str., Saratov Region

413864, Russia^3

关键词: Diagnostic functions;    Electrical actuators;    Forecasting methods;    Fourier transformations;    Hierarchical structures;    Neural network techniques;    Operating cycle;    Operational cycle;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/177/1/012140/pdf
DOI  :  10.1088/1757-899X/177/1/012140

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

This article describes a forecasting method for electrical actuators realized through the combination of Fourier transformation and neural network techniques. The method allows finding the value of diagnostic functions in the iterating operating cycle and the number of operational cycles in time before the BLDC actuator fails. For forecasting of the condition of the actuator, we propose a hierarchical structure of the neural network aiming to reduce the training time of the neural network and improve estimation accuracy.