دانلود رایگان ترجمه مقاله سیستم چند منظوره برای نظارت بر توالی های راه اندازی توربین گازی صنعتی (نشریه IEEE 2001)

یک سیستم چند منظوره برای نظارت بر توالی های راه اندازی توربین گازی صنعتی

A Multi Agent System for Monitoring Industrial Gas Turbine Start-up Sequences

Abstract

This paper introduces a novel technique for condition monitoring of gas turbine start-up sequences. The vast amount of data and the complex processes behind on-line fault detection indicate the need for an automated solution. A multi-agent system that views the problem as the interaction of simple independent software entities for effective use of the available data is presented. The overall solution is derived from the combination of partial solutions provided by the components of the multiagent system. As a consequence, data interpretation is achieved by converting the data into appropriate information and combining individual agents’ information, resulting in an automatic fault diagnosis for the engineers. This multiagent system can employ various intelligent system techniques and has been implemented using the ZEUS agent building toolkit.

1 Introduction

BY DEFINITION, condition monitoring is performed when it is necessary to assess the state of a plant item and to determine whether it is malfunctioning, through reasoning and observation [1]. Early diagnosis and fault identification is an important activity for maximizing the plant economy, its operational costs and levels of safety. Engineers have introduced better decision support for complex condition monitoring procedures through the application of centralized intelligent systems by using a variety of artificial intelligence (AI) techniques [2]–[5]. It is now widely recognized that problems due to the functional complexity of condition monitoring can be overcome with architectures that contain many dynamically interacting intelligent distributed modules, called intelligent agents [6]. Each agent is an autonomous system, which processes a selection of input, and the complete interpretation and diagnosis is accomplished through the process of interaction with the other agents.

This paper introduces a real time condition monitoring multi agent system (COMMAS) developed to facilitate the task of monitoring the performance of a gas turbine. The agents collect measurements from the turbine sensors and an existing engine management system (EMS), analyze abnormal measurements and perform fault diagnosis. The aim of this work is to improve the accuracy of the existing monitoring systems by building smaller software components, which contain partial information of the state of the plant being monitored. They interact in a dynamic way to support data fusion, cross sensor corroboration and decision support functions more efficiently. This contrasts with centralized applications, which model the whole plant in a single software system.

Within this paper, Section II discusses the application problem domain of gas turbine condition monitoring. Section III presents the conceptual framework of the proposed multi-agent system and Section IV describes the analysis and design including the information flow between the agents in order to replicate the diagnostic tasks performed by the engineers. Section V presents the implementation of the framework and Section VI outlines the conclusions and some directions for future work.