An Autoregressive Fault Model for Condition Monitoring of Electrical Machines in Deep-level Mines

Equipment failure is a serious issue in the mining sector since it affects both reliability and availability. This in turn may result in production losses. Strategies are thus needed to prevent equipment failure. Therefore, in this paper a condition monitoring methodology based on autoregressive fault detection is applied to electrical machines on deep level gold and platinum mines in South Africa. The aim of this methodology is to improve condition-based maintenance strategies for large electrical machines used in harsh mining environments. Improved condition monitoring contributes to improved equipment reliability, ultimately resulting in lower electricity consumption, reduced operational costs and increased production. The autoregressive fault detection model was implemented on two case studies which include two large threephase induction motors. Case Study 1 presents a large disturbance in the temperature of a non-drive end bearing of a multistage centrifugal compressor that was successfully detected by the model. Case Study 2 presents a gradually increasing motor winding temperature of an electric motor powering a multistage centrifugal compressor that was also successfully detected. These two case studies prove that the autoregressive fault detection model is a viable option for South African mines due to its relatively low implementation cost and its capability of automatically detecting faults even within existing alarm and trip limits. Another advantage is that the model can be applied generically to different machinery without the need for knowledge or information about applicable alarm or trip limits.