Engine Health Assessment and Prediction Using the Group Method of Data Handling and the Method of Match Matrix: Autoregressive Moving Average
Abstract
This research aims to develop a methodology for engine health assessment and long-term anomaly prediction of the engine. This paper presents methodology for engine health assessment and prediction, which includes the following two steps: (1) engine health assessment and anomaly detection is conducted based on an inductive learning technique called Group Method of Data Handling (GMDH), and (2) engine health prediction is conducted based on the Match Matrix – Autoregressive Moving Average (ARMA) model method. Results from an industry case study illustrate the presented methodology's effectiveness in engine health assessment and prediction. An effective anomaly prediction process capable of providing an early warning of a fault in an engine would be of tremendous value to the user and could result in substantial financial savings.