JPWO2022130098A5 - - Google Patents
Download PDFInfo
- Publication number
- JPWO2022130098A5 JPWO2022130098A5 JP2023534934A JP2023534934A JPWO2022130098A5 JP WO2022130098 A5 JPWO2022130098 A5 JP WO2022130098A5 JP 2023534934 A JP2023534934 A JP 2023534934A JP 2023534934 A JP2023534934 A JP 2023534934A JP WO2022130098 A5 JPWO2022130098 A5 JP WO2022130098A5
- Authority
- JP
- Japan
- Prior art keywords
- local
- computer
- implemented method
- model
- models
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 claims 40
- 230000003068 static effect Effects 0.000 claims 6
- 238000004220 aggregation Methods 0.000 claims 2
- 230000002776 aggregation Effects 0.000 claims 2
- 238000004590 computer program Methods 0.000 claims 2
- 238000011176 pooling Methods 0.000 claims 1
Claims (22)
前記資産の特徴を静的特徴、半静的特徴および動的特徴に区切ることと、
前記静的特徴および前記半静的特徴に基づいて、前記資産のコホートを形成することと、
前記コホートの各々のために、ローカル・サイトでローカル・モデルを生成することと、
各故障タイプのために、前記コホートの各々のための前記ローカル・モデルをローカル・データ上で訓練することと、
訓練された前記ローカル・モデルを中央データベースと共有することと、
複数の前記ローカル・サイトからの複数の前記ローカル・モデルの集約に基づいて、グローバル・モデルを作成することと、
前記複数のローカル・サイトの各々で、
前記コホートの各々のために、前記グローバル・モデルまたは前記ローカル・モデルを選ぶことと、
前記コホートの1つに属する前記資産のうちの1つまたは複数の故障を予測するために、選ばれた前記モデルをローカル・データ上で動作させることと、
を含むコンピュータ実施方法。 A computer-implemented method of predicting asset failure, the computer-implemented method comprising:
dividing the characteristics of the asset into static characteristics, semi-static characteristics and dynamic characteristics;
forming a cohort of assets based on the static characteristics and the semi-static characteristics;
generating a local model at a local site for each of said cohort;
training the local model for each of the cohorts on local data for each failure type;
sharing the trained local model with a central database;
creating a global model based on aggregation of a plurality of said local models from a plurality of said local sites;
At each of the plurality of local sites,
selecting the global model or the local model for each of the cohorts;
operating the selected model on local data to predict failure of one or more of the assets belonging to one of the cohorts;
computer-implemented methods including;
請求項1に記載のコンピュータ実施方法。 The computer-implemented method further includes generating a template model for creating each of the local models for each of the cohorts.
The computer-implemented method of claim 1.
前記複数のローカル・サイトの各々からの前記ローカル・モデルをローカル・モデルのプール内にプールすることと、
前記グローバル・モデルの性能と、ローカル・モデルの前記プールからの前記ローカル・モデルのうちの選択された1つと、を決定することと、
をさらに含む、
請求項1に記載のコンピュータ実施方法。 The computer-implemented method includes:
pooling the local models from each of the plurality of local sites into a pool of local models;
determining the performance of the global model and a selected one of the local models from the pool of local models;
further including,
The computer-implemented method of claim 1.
請求項3に記載のコンピュータ実施方法。 the selected model is selected based on the performance of the global model and the selected one of the local models;
4. The computer-implemented method of claim 3.
請求項3に記載のコンピュータ実施方法。 The computer-implemented method further includes determining a mismatch between the global model and the selected one of the local models.
4. The computer-implemented method of claim 3.
請求項3に記載のコンピュータ実施方法。 The computer-implemented method further includes tuning the selected model based on the local data of the asset to improve performance of the local model.
4. The computer-implemented method of claim 3.
請求項6に記載のコンピュータ実施方法。 The computer-implemented method further includes providing the tuned and selected model to the pool of local models.
7. The computer-implemented method of claim 6.
請求項3に記載のコンピュータ実施方法。 The computer-implemented method further includes updating the global model based on an average of each of the local models in the pool of local models for each of the cohorts.
4. The computer-implemented method of claim 3.
請求項3に記載のコンピュータ実施方法。 each of the local models in the pool of local models is weighted based on the average number of assets of the local site that contributed the local model to the pool of local models;
4. The computer-implemented method of claim 3.
前記資産の特徴を静的特徴、半静的特徴および動的特徴に区切ることと、
前記静的特徴および前記半静的特徴に基づいて、前記資産のコホートを形成することと、
前記コホートの各々のために、ローカル・サイトでローカル・モデルを生成することと、
前記ローカル・モデルを中央データベースと共有することと、
それぞれの複数の前記ローカル・サイトからの複数の前記ローカル・モデルから前記ローカル・モデルのプールを形成することと、
前記複数の前記ローカル・サイトからの前記複数の前記ローカル・モデルの集約に基づいて、グローバル・モデルを作成することと、
前記複数のローカル・サイトの各々で、
前記コホートの各々のために、前記グローバル・モデルのうちの1つおよびローカル・モデルの前記プールから前記複数のローカル・モデルのうちの1つを選ぶことと、
前記コホートの1つに属する前記資産のうちの1つまたは複数の前記故障を予測するために、選ばれた前記モデルをローカル・データ上で動作させることと、
を含むコンピュータ実施方法。 A computer-implemented method for predicting asset failure, the computer-implemented method comprising:
dividing the characteristics of the asset into static characteristics, semi-static characteristics and dynamic characteristics;
forming a cohort of assets based on the static characteristics and the semi-static characteristics;
generating a local model at a local site for each of said cohort;
sharing the local model with a central database;
forming the pool of local models from a plurality of the local models from each of the plurality of local sites;
creating a global model based on aggregation of the plurality of local models from the plurality of local sites;
At each of the plurality of local sites,
selecting, for each of the cohorts, one of the global models and one of the plurality of local models from the pool of local models;
operating the selected model on local data to predict the failure of one or more of the assets belonging to one of the cohorts;
computer-implemented methods including;
請求項10に記載のコンピュータ実施方法。 The computer-implemented method further includes generating a template model for creating each of the local models for each of the cohorts.
11. The computer-implemented method of claim 10.
請求項10に記載のコンピュータ実施方法。 The computer-implemented method further includes determining a mismatch between the global model and the selected model.
11. The computer-implemented method of claim 10.
請求項10に記載のコンピュータ実施方法。 The computer-implemented method further includes tuning the selected model based on local data to improve performance of the local model.
11. The computer-implemented method of claim 10.
請求項13に記載のコンピュータ実施方法。 The computer-implemented method further includes providing the tuned and selected model to the pool of local models.
14. The computer-implemented method of claim 13.
請求項10に記載のコンピュータ実施方法。 The computer-implemented method further includes updating the global model based on an average of each of the local models in the pool of local models for each of the cohorts.
11. The computer-implemented method of claim 10.
請求項15に記載のコンピュータ実施方法。 each of the local models in the pool of local models is weighted based on the average number of assets of the local site that contributed the local model to the pool of local models;
16. The computer-implemented method of claim 15.
前記資産の特徴を静的特徴、半静的特徴および動的特徴に区切ることと、
前記静的特徴および前記半静的特徴に基づいて、前記資産のコホートを形成することと、
前記コホートの各々のために、ローカル・サイトでローカル・モデルを生成することと、
前記コホートの各々のために、グローバル・モデルを作成することと、
複数のローカル・サイトの各々で、
前記コホートの各々のために、前記グローバル・モデルおよび前記ローカル・モデルの1つを選ぶことと、
前記コホートの1つに属する前記資産のうちの1つまたは複数の故障を予測するために、選ばれた前記モデルを前記ローカル・サイトにローカルなデータ上で動作させることと、
を含むコンピュータ実施方法。 A computer-implemented method for asset failure prediction, the computer-implemented method comprising:
dividing the characteristics of the asset into static characteristics, semi-static characteristics and dynamic characteristics;
forming a cohort of assets based on the static characteristics and the semi-static characteristics;
generating a local model at a local site for each of said cohort;
creating a global model for each of said cohorts;
At each of multiple local sites,
selecting one of the global model and the local model for each of the cohorts;
operating the selected model on data local to the local site to predict failure of one or more of the assets belonging to one of the cohorts;
computer-implemented methods including;
前記複数のローカル・サイトのための前記ローカル・モデルの各々を中央データベース内に格納することと、
前記コホートの各々のためのローカル・モデルのプール内の前記ローカル・モデルの各々の平均に基づいて、前記グローバル・モデルを更新することと、
をさらに含む、
請求項17に記載のコンピュータ実施方法。 The computer-implemented method includes:
storing each of the local models for the plurality of local sites in a central database;
updating the global model based on an average of each of the local models in a pool of local models for each of the cohorts;
further including,
18. The computer-implemented method of claim 17.
請求項18に記載のコンピュータ実施方法。 each of the local models in the pool of local models is weighted based on the average number of assets of the local site that contributed the local model to the pool of local models;
19. The computer-implemented method of claim 18.
前記複数のローカル・サイトの各々で、
前記コホートの各々のために、更新された前記グローバル・モデルおよび前記ローカル・モデルの1つを選ぶことと、
前記資産の故障を予測するために、選ばれた前記モデルをローカル・データ上で動作させることと、
をさらに含む、
請求項18に記載のコンピュータ実施方法。 The computer-implemented method includes:
At each of the plurality of local sites,
selecting one of the updated global model and the local model for each of the cohorts;
operating the selected model on local data to predict failure of the asset;
further including,
19. The computer-implemented method of claim 18.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US17/123,088 | 2020-12-15 | ||
US17/123,088 US20220188775A1 (en) | 2020-12-15 | 2020-12-15 | Federated learning for multi-label classification model for oil pump management |
PCT/IB2021/061237 WO2022130098A1 (en) | 2020-12-15 | 2021-12-02 | Federated learning for multi-label classification model for oil pump management |
Publications (2)
Publication Number | Publication Date |
---|---|
JP2023553909A JP2023553909A (en) | 2023-12-26 |
JPWO2022130098A5 true JPWO2022130098A5 (en) | 2024-01-10 |
Family
ID=81941542
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2023534934A Pending JP2023553909A (en) | 2020-12-15 | 2021-12-02 | Federated learning for multi-label classification models for oil pump management |
Country Status (5)
Country | Link |
---|---|
US (1) | US20220188775A1 (en) |
JP (1) | JP2023553909A (en) |
CN (1) | CN116601632A (en) |
DE (1) | DE112021005868T5 (en) |
WO (1) | WO2022130098A1 (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116432040B (en) * | 2023-06-15 | 2023-09-01 | 上海零数众合信息科技有限公司 | Model training method, device and medium based on federal learning and electronic equipment |
Family Cites Families (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6708163B1 (en) * | 1999-02-24 | 2004-03-16 | Hillol Kargupta | Collective data mining from distributed, vertically partitioned feature space |
US8392343B2 (en) * | 2010-07-21 | 2013-03-05 | Yahoo! Inc. | Estimating probabilities of events in sponsored search using adaptive models |
US20170308802A1 (en) * | 2016-04-21 | 2017-10-26 | Arundo Analytics, Inc. | Systems and methods for failure prediction in industrial environments |
US20200364608A1 (en) * | 2019-05-13 | 2020-11-19 | International Business Machines Corporation | Communicating in a federated learning environment |
CN111369042B (en) * | 2020-02-27 | 2021-09-24 | 山东大学 | Wireless service flow prediction method based on weighted federal learning |
CN111382706A (en) * | 2020-03-10 | 2020-07-07 | 深圳前海微众银行股份有限公司 | Prediction method and device based on federal learning, storage medium and remote sensing equipment |
CN111798002A (en) * | 2020-05-31 | 2020-10-20 | 北京科技大学 | Local model proportion controllable federated learning global model aggregation method |
WO2021247448A1 (en) * | 2020-06-01 | 2021-12-09 | Intel Corporation | Federated learning optimizations |
CN111754000B (en) * | 2020-06-24 | 2022-10-14 | 清华大学 | Quality-aware edge intelligent federal learning method and system |
CN111737749A (en) * | 2020-06-28 | 2020-10-02 | 南方电网科学研究院有限责任公司 | Measuring device alarm prediction method and device based on federal learning |
EP4214584A4 (en) * | 2020-08-20 | 2024-05-15 | Hitachi Vantara LLC | Systems and methods for an automated data science process |
US11704942B2 (en) * | 2020-10-29 | 2023-07-18 | Caterpillar Inc. | Undercarriage wear prediction using machine learning model |
US20220138260A1 (en) * | 2020-10-30 | 2022-05-05 | Here Global B.V. | Method, apparatus, and system for estimating continuous population density change in urban areas |
-
2020
- 2020-12-15 US US17/123,088 patent/US20220188775A1/en active Pending
-
2021
- 2021-12-02 JP JP2023534934A patent/JP2023553909A/en active Pending
- 2021-12-02 WO PCT/IB2021/061237 patent/WO2022130098A1/en active Application Filing
- 2021-12-02 DE DE112021005868.1T patent/DE112021005868T5/en active Pending
- 2021-12-02 CN CN202180080228.4A patent/CN116601632A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7413580B2 (en) | Generating integrated circuit floorplans using neural networks | |
US11556690B2 (en) | Generating integrated circuit placements using neural networks | |
JP6758406B2 (en) | Wide and deep machine learning model | |
US8532968B2 (en) | Method of improving the production of a mature gas or oil field | |
WO2020155300A1 (en) | Model prediction method and device | |
Chen | Characterizing growth and form of fractal cities with allometric scaling exponents | |
US20230297553A1 (en) | Relation-enhancement knowledge graph embedding method and system | |
Schirmann et al. | Data-driven models for vessel motion prediction and the benefits of physics-based information | |
US20170206451A1 (en) | Centralized management of predictive models | |
Zhang | Application of data mining and machine learning in management accounting information system | |
JPWO2022130098A5 (en) | ||
CN117631682A (en) | Method and system for determining inspection route of unmanned aerial vehicle of power grid | |
US20220327482A1 (en) | Artificial intelligence for freight estimation | |
CN108229572B (en) | Parameter optimization method and computing equipment | |
CN113887067A (en) | Compact low-permeability reservoir fracturing well pattern gas flooding effect prediction method based on LSTM | |
US20230289721A1 (en) | Systems and methods for probabilistic estimation in tree-based forecast models | |
US12086822B2 (en) | Crisis-recovery data analytics engine in a data analytics system | |
US11897818B2 (en) | Machine learning concrete optimization | |
Galiano et al. | Application of Sleator-Tarjan Dynamic Trees in a Monitoring System for the Arctic Region Based on Remote Sensing Data | |
US20240095424A1 (en) | Alignment Cost for Integrated Circuit Placement | |
CN112990332B (en) | Sub-graph scale prediction and distributed training method and device and electronic equipment | |
US20230196062A1 (en) | Method and system for global explainability of neural networks | |
CN118709832A (en) | Oil reservoir injection and production optimization method, device, equipment and medium | |
EP4409460A1 (en) | Generating integrated circuit placements using neural networks | |
Dash | Design of data scoring model for big data |