JP7708834B2 - メンテナンス意思決定のための強化学習方法及びコンピュータ可読媒体 - Google Patents
メンテナンス意思決定のための強化学習方法及びコンピュータ可読媒体Info
- Publication number
- JP7708834B2 JP7708834B2 JP2023191931A JP2023191931A JP7708834B2 JP 7708834 B2 JP7708834 B2 JP 7708834B2 JP 2023191931 A JP2023191931 A JP 2023191931A JP 2023191931 A JP2023191931 A JP 2023191931A JP 7708834 B2 JP7708834 B2 JP 7708834B2
- Authority
- JP
- Japan
- Prior art keywords
- decision maker
- actions
- rul
- equipment
- estimator
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/092—Reinforcement learning
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0283—Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G06N3/0455—Auto-encoder networks; Encoder-decoder networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/045—Explanation of inference; Explainable artificial intelligence [XAI]; Interpretable artificial intelligence
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Medical Informatics (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing And Monitoring For Control Systems (AREA)
Description
与えられたウィンドウTについて、それぞれの与えられた時間kで:
Claims (6)
- 設備の予知保全のための予知保全支援システムのオフライン強化学習方法であって、
前記予知保全支援システムは、
過去の観測、過去のアクション、関連する報酬、および関連する予測アクションに関連する履歴を格納するデータベースと、XAIユニットと、残存耐用年数(RUL)推定器と、決定器訓練エンジンと、GUIと、報酬計算器と、意思決定器と、を備え、
前記報酬計算器によって、ユーザからの指示に従って動作する複数の機器の内部センサのセンサデータが入力され、期待される将来の収益期待値を出力し、
前記意思決定器は、
前記センサデータとして過去の観測結果と、過去のアクションと、前記期待される将来の収益期待値と、を入力し、
予測された次のアクションと、前記意思決定器の意思決定器モデルの信頼度スコアと、を出力し、
前記XAIユニットは、前記データベースと前記意思決定器とにアクセスし、人間が読める予測された次のアクションの説明を出力し、
前記RUL推定器によって、
前記センサからの過去の観測結果を入力し、前記意思決定器に残存耐用年数の推定値を前記意思決定器に出力し、
前記データベースによって、時間毎に、前記複数の機器のそれぞれの、センサデータと、収益と、修理コスト、報酬とを格納し、
前記決定器訓練エンジンによって、前記データベースのデータを取得し、前記RUL推定器のトレーニングし、
前記決定器訓練エンジンによって、さらに、
前記信頼度スコアを閾値と比較し、
前記信頼度スコアがしきい値を下回る場合、最近の観察よりも時間的に新しく観測された観察と、最近のアクションよりも時間的に新しく観測されたアクションを入力として、前記意思決定器モデルを再トレーニングする
オフライン強化学習方法。 - 請求項1に記載のオフライン強化学習方法であって、
前記RUL推定器への入力として、最近の観測値を供給し、
前記RUL推定器からの出力として、機器の推定残存耐用年数を生成し、
生成された機器の推定残存耐用年数は、次のアクションを生成する際の前記意思決定器
の前記意思決定器モデルの入力として使用される
オフライン強化学習方法。 - 請求項2に記載のオフライン強化学習方法であって、
生成された前記機器の推定残存耐用年数を、前記GUIに表示する
オフライン強化学習方法。 - 機器の予知保全のための命令を記憶した、非一時的なコンピュータ可読媒体であって、
前記命令は、以下の処理をコンピュータに実行させる、
ユーザからの指示に従って動作する複数の機器の内部センサのセンサデータが入力され、期待される将来の収益期待値を出力させ、
前記センサデータとして過去の観測結果と、過去のアクションと、前記期待される将来の収益期待値と、を入力させ、
予測された次のアクションと、前記コンピュータの意思決定器の意思決定器モデルの信頼度スコアと、を出力させ、
過去の観測、過去のアクション、関連する報酬、および関連する予測アクションに関連する履歴を格納するデータベースと前記意思決定器とにアクセスし、人間が読める予測された次のアクションの説明を出力させ、
前記センサからの過去の観測結果を入力させ、前記意思決定器に残存耐用年数の推定値を前記意思決定器に出力させ、
時間毎に、前記複数の機器のそれぞれの、センサデータと、収益と、修理コスト、報酬とを格納させ、
過去の観測、過去のアクション、関連する報酬、および関連する予測アクションに関連する履歴を格納するデータベースのデータを取得させ、残存耐用年数(RUL)推定器をトレーニングさせて、
さらに、
前記信頼度スコアを閾値と比較させ、
前記信頼度スコアがしきい値を下回る場合、最近の観察よりも時間的に新しく観測された観察と、最近のアクションよりも時間的に新しく観測されたアクションを入力として、前記意思決定器モデルを再トレーニングさせる
コンピュータ可読媒体。 - 請求項4に記載のコンピュータ可読媒体であって、
前記命令は、
残存耐用年数推定器(RUL推定器)への入力として、最近の観測値を供給させ、
前記RUL推定器からの出力として、機器の推定残存耐用年数を生成させ、
生成された機器の推定残存耐用年数は、次のアクションを生成する際の前記意思決定器モデルの入力として使用させる
コンピュータ可読媒体。 - 請求項5に記載のコンピュータ可読媒体であって、
前記命令は、
グラフィカル・ユーザー・インターフェース(GUI)にモデル出力と機器の推定残存耐用年数を表示する
コンピュータ可読媒体。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US18/102,388 US20240255939A1 (en) | 2023-01-27 | 2023-01-27 | Reinforcement learning system for maintenance decision making |
| US18/102388 | 2023-01-27 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP2024106961A JP2024106961A (ja) | 2024-08-08 |
| JP7708834B2 true JP7708834B2 (ja) | 2025-07-15 |
Family
ID=89619823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2023191931A Active JP7708834B2 (ja) | 2023-01-27 | 2023-11-10 | メンテナンス意思決定のための強化学習方法及びコンピュータ可読媒体 |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US20240255939A1 (ja) |
| EP (1) | EP4407523A1 (ja) |
| JP (1) | JP7708834B2 (ja) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022181453A1 (ja) | 2021-02-26 | 2022-09-01 | 京セラ株式会社 | 熱デバイス |
| US12259968B2 (en) * | 2022-02-11 | 2025-03-25 | Microsoft Technology Licensing, Llc | Detecting anomalous post-authentication behavior for a workload identity |
| US20240346283A1 (en) * | 2023-04-14 | 2024-10-17 | Kyndryl, Inc. | Explainable classifications with abstention using client agnostic machine learning models |
| CN119579140B (zh) * | 2024-11-14 | 2025-09-26 | 重庆大学 | 一种基于强化学习的风电机舱装备质量评估与维修决策方法 |
| CN120013530B (zh) * | 2025-04-21 | 2025-06-20 | 杭州市北京航空航天大学国际创新研究院(北京航空航天大学国际创新学院) | 剩余寿命不确定下的飞机强化学习预测性维修决策方法 |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190384257A1 (en) | 2018-06-13 | 2019-12-19 | Hitachi, Ltd. | Automatic health indicator learning using reinforcement learning for predictive maintenance |
| US20200265331A1 (en) | 2019-02-20 | 2020-08-20 | Accenture Global Solutions Limited | System for predicting equipment failure events and optimizing manufacturing operations |
| US20210323167A1 (en) | 2018-08-27 | 2021-10-21 | 3M Innovative Properties Company | Learning framework for robotic paint repair |
| US20220004182A1 (en) | 2020-07-02 | 2022-01-06 | Nec Laboratories America, Inc. | Approach to determining a remaining useful life of a system |
| US20220188181A1 (en) | 2020-12-15 | 2022-06-16 | International Business Machines Corporation | Restricting use of selected input in recovery from system failures |
| US20220391670A1 (en) | 2020-10-14 | 2022-12-08 | UMNAI Limited | Explanation and interpretation generation system |
| US20220398460A1 (en) | 2021-06-09 | 2022-12-15 | UMNAI Limited | Automatic xai (autoxai) with evolutionary nas techniques and model discovery and refinement |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12576515B2 (en) * | 2020-07-28 | 2026-03-17 | Gdm Holding Llc | Off-line learning for robot control using a reward prediction model |
| US12228905B2 (en) * | 2021-12-10 | 2025-02-18 | Applied Materials, Inc. | Eco-efficiency monitoring and exploration platform for semiconductor manufacturing |
-
2023
- 2023-01-27 US US18/102,388 patent/US20240255939A1/en active Pending
- 2023-11-10 JP JP2023191931A patent/JP7708834B2/ja active Active
-
2024
- 2024-01-15 EP EP24151917.2A patent/EP4407523A1/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190384257A1 (en) | 2018-06-13 | 2019-12-19 | Hitachi, Ltd. | Automatic health indicator learning using reinforcement learning for predictive maintenance |
| US20210323167A1 (en) | 2018-08-27 | 2021-10-21 | 3M Innovative Properties Company | Learning framework for robotic paint repair |
| US20200265331A1 (en) | 2019-02-20 | 2020-08-20 | Accenture Global Solutions Limited | System for predicting equipment failure events and optimizing manufacturing operations |
| US20220004182A1 (en) | 2020-07-02 | 2022-01-06 | Nec Laboratories America, Inc. | Approach to determining a remaining useful life of a system |
| US20220391670A1 (en) | 2020-10-14 | 2022-12-08 | UMNAI Limited | Explanation and interpretation generation system |
| US20220188181A1 (en) | 2020-12-15 | 2022-06-16 | International Business Machines Corporation | Restricting use of selected input in recovery from system failures |
| US20220398460A1 (en) | 2021-06-09 | 2022-12-15 | UMNAI Limited | Automatic xai (autoxai) with evolutionary nas techniques and model discovery and refinement |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2024106961A (ja) | 2024-08-08 |
| EP4407523A1 (en) | 2024-07-31 |
| US20240255939A1 (en) | 2024-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7708834B2 (ja) | メンテナンス意思決定のための強化学習方法及びコンピュータ可読媒体 | |
| US11042145B2 (en) | Automatic health indicator learning using reinforcement learning for predictive maintenance | |
| JP7603807B2 (ja) | ラベルなしセンサデータを用いた産業システム内の稀な障害の自動化されたリアルタイムの検出、予測、及び予防に関する、方法または非一時的コンピュータ可読媒体 | |
| US11231703B2 (en) | Multi task learning with incomplete labels for predictive maintenance | |
| EP3407267A1 (en) | Deep learning network architecture optimization for uncertainty estimation in regression | |
| US11500370B2 (en) | System for predictive maintenance using generative adversarial networks for failure prediction | |
| US20210064983A1 (en) | Machine learning for industrial processes | |
| US20200327450A1 (en) | Addressing a loss-metric mismatch with adaptive loss alignment | |
| US20260073293A1 (en) | Real time detection, prediction and remediation of machine learning model drift in asset hierachy based on time-series data | |
| WO2023191787A1 (en) | Recommendation for operations and asset failure prevention background | |
| JP2025517109A (ja) | データ駆動型アプローチを通したセンサ故障のリアルタイム検出、予測、及び修復 | |
| JP2023048962A (ja) | 強化学習ポリシを訓練する方法 | |
| Liu et al. | Residual useful life prognosis of equipment based on modified hidden semi-Markov model with a co-evolutional optimization method | |
| US12608595B2 (en) | Method for reducing bias in deep learning classifiers using ensembles | |
| Karagiorgou et al. | Unveiling trends and predictions in digital factories | |
| US20230196088A1 (en) | Fan behavior anomaly detection using neural network | |
| CN118475438A (zh) | 过程中指令层面的错误原因的识别 | |
| WO2022039748A1 (en) | Systems and methods for an automated data science process | |
| WO2022177567A1 (en) | Dependability assessment framework for railroad asset management | |
| EP4506874A1 (en) | Method for generalized and alignment model for repair recommendation | |
| US12619887B2 (en) | Systems and methods for an automated data science process | |
| US20250272176A1 (en) | Artificial intelligence-based system and method for determining potential issues occurred in equipments by analyzing data using a root cause analysis engine | |
| US20250172933A1 (en) | Method and controller for generating a predictive maintenance alert | |
| Khamrakulov et al. | Probabilistic Inference Algorithm for Fault Diagnosis in Industrial Control Systems | |
| JP2026508319A (ja) | 深層学習に基づく人物固有洞察生成器 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20231110 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20231114 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20240827 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20240905 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20241217 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20250108 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20250415 |
|
| A521 | Request for written amendment filed |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20250512 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20250701 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20250703 |
|
| R150 | Certificate of patent or registration of utility model |
Ref document number: 7708834 Country of ref document: JP Free format text: JAPANESE INTERMEDIATE CODE: R150 |