FAILURE DIAGNOSIS DEVICE, FAILURE DIAGNOSIS SYSTEM, AND FAILURE DIAGNOSIS PROGRAM

• Main Authors: KODA JUNYA, TAKASU TOSHIKI, KIMURA SHUNSUKE
• Format: Patent
• Language: English; Japanese
• Published: 17.03.2022
• Subjects: BASIC ELECTRIC ELEMENTS; CALCULATING; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRICITY; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSIONOF CHEMICAL INTO ELECTRICAL ENERGY; REGULATING; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

To efficiently diagnose the failure of a fuel cell cogeneration system while reducing the burden on a worker at a site.SOLUTION: When a CGS 40 detects an abnormal operation, a failure diagnosis device 10 acquires operation history data of the CGS 40 corresponding to the abnormality from the CGS 40, selects a failure diagnosis model of a function portion corresponding to an error code indicating the type of abnormality obtained from the acquired operation history data from among failure diagnosis models for function portions that are generated by associating an operation history data group obtained in the past with the function portion of the CGS 40 and performing multivariate analysis or machine learning for the operation history data group, and determine the presence or absence of a failure for each of the function portions of the CGS 40, inputs the acquired operation history data or data obtained by processing the operation history data to the selected failure diagnosis model, and estimates the presence or absence of failure for each of the functional portions of the CGS 40 by using the determination result output from the failure diagnosis model.SELECTED DRAWING: Figure 1 【課題】燃料電池コージェネレーションシステムの故障診断を、作業担当者の現場での負担を軽減させつつ、効率良く行う。【解決手段】故障診断装置１０は、ＣＧＳ４０が動作の異常を検知した場合に、ＣＧＳ４０から、異常に対応するＣＧＳ４０の動作履歴データを取得し、過去に得られた動作履歴データ群を、ＣＧＳ４０の機能部に対応付けて多変量解析又は機械学習することにより生成され、かつ、ＣＧＳ４０の機能部毎に故障の有無を判定する機能部毎の故障診断モデルの中から、取得された動作履歴データから得られる異常の種別を表すエラーコードに対応する機能部の故障診断モデルを選択し、選択された故障診断モデルに対して、取得された動作履歴データ又は動作履歴データを加工して得られるデータを入力し、故障診断モデルから出力される判定結果を用いて、ＣＧＳ４０の機能部毎に故障の有無を推定する。【選択図】図１