Online Predictive Maintenance Monitoring Adopting Convolutional Neural Networks

Christian Gianoglio,Federico Gallesi,Francesco Guastavino,Edoardo Ragusa,Paolo Gastaldo

doi:10.3390/en14154711

Abstract

Thermal, electrical and mechanical stresses age the electrical insulation systems of high voltage (HV) apparatuses until the breakdown. The monitoring of the partial discharges (PDs) effectively assesses the insulation condition. PDs are both the symptoms and the causes of insulation aging and—in the long term—can lead to a breakdown, with a burdensome economic loss. This paper proposes the convolutional neural networks (CNNs) to investigate and analyze the aging process of enameled wires, thus predicting the life status of the insulation systems. The CNNs training does not require any kind of assumption of how the factors (e.g., voltage, frequency and temperature) contribute to the life model. The experiments confirm that the proposal obtains better estimations of the life status of twisted pair specimens concerning existing solutions, which are based on strong hypotheses about the life model dependency on the factors.

Highlights

Monitoring Adopting ConvolutionalThe online monitoring of high voltage (HV) apparatuses prevents economic losses due to the breakdown of an insulation system [1,2]
The present paper introduces a model for aging assessment based on convolutional neural networks (CNNs), exploiting the ability of CNNs to deal with complex, non-linear problems when input data can be represented as tensors
This paper presented a novel strategy for the aging prediction of electrical insulation systems

Summary

Introduction

The online monitoring of high voltage (HV) apparatuses prevents economic losses due to the breakdown of an insulation system [1,2]. The conditions of an electrical machine are assessed via periodical checks, which cause a temporary disservice and a waste of money. Online predictive maintenance estimates the status of the insulation system without interrupting the normal functioning [3]. Though, should be supported by effective models that can reliably assess aging phenomena. From this perspective, the literature shows that existing solutions still need to be improved. Partial discharges (PDs) are a valuable indicator of the insulation condition [4,5]

Objectives

Methods

Conclusion