Method and system for semi supervised deep fault detection for large industrial monitoring system based on time series data using digital twin simulation data

摘要

Problem to be solved: to provide a method for detecting an abnormal operation status of a technical system using a training phase and a monitoring phase.The training phase acquires a first set of time series values generated by a digital twin simulation on a normal status and a second set of time series values measured by a plurality of sensors.The sensor monitors the set of operating parameters, collects a second set at an abnormal status, and detects parameters of the machine learning model to detect normal status and determine from abnormal status.The data samples each include first and second samples from a first set and a third sample from a second set.The monitoring phase calculates an abnormal score value for determining an abnormal status based on the set of multivariate time series values measured by the plurality of sensors and the trained machine learning model, and outputs information about the abnormal status.Diagram