JP2022069608A5 - - Google Patents
Download PDFInfo
- Publication number
- JP2022069608A5 JP2022069608A5 JP2022041688A JP2022041688A JP2022069608A5 JP 2022069608 A5 JP2022069608 A5 JP 2022069608A5 JP 2022041688 A JP2022041688 A JP 2022041688A JP 2022041688 A JP2022041688 A JP 2022041688A JP 2022069608 A5 JP2022069608 A5 JP 2022069608A5
- Authority
- JP
- Japan
- Prior art keywords
- data
- state
- plant
- item
- map data
- 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.)
- Granted
Links
- 238000010801 machine learning Methods 0.000 claims 22
- 238000013523 data management Methods 0.000 claims 20
- 238000007726 management method Methods 0.000 claims 18
- 239000004568 cement Substances 0.000 claims 16
- 238000000034 method Methods 0.000 claims 11
- 230000006870 function Effects 0.000 claims 6
- 235000008733 Citrus aurantifolia Nutrition 0.000 claims 4
- 235000011941 Tilia x europaea Nutrition 0.000 claims 4
- 238000009825 accumulation Methods 0.000 claims 4
- 239000004571 lime Substances 0.000 claims 4
- 238000011156 evaluation Methods 0.000 claims 3
- 239000002994 raw material Substances 0.000 claims 2
Claims (23)
前記学習用データ管理部に蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを構築するモデル構築部と、を備え、
前記学習用データ管理部は、少なくとも二つの前記状態項目同士の間で、前記データの変動時刻のずれを縮小するように、当該二つの状態項目間で時間をずらして前記マップデータを補正する、プラント管理サーバ。 For each cell specified by the combination of the time interval and the state item in the two-dimensional space of the time axis including a plurality of time intervals arranged in a time series and the item axis including a plurality of state items related to the operation state of the plant. A training data management unit that stores training data that combines map data including data and teacher data indicating the state of the plant corresponding to the map data.
A model building unit for constructing a model for estimating the state of the plant by machine learning based on the learning data accumulated in the learning data management unit is provided .
The learning data management unit corrects the map data by shifting the time between the two state items so as to reduce the deviation of the fluctuation time of the data between at least two state items. , Plant management server.
前記学習用データ管理部に蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを構築するモデル構築部と、を備え、 A model building unit for constructing a model for estimating the state of the plant by machine learning based on the learning data accumulated in the learning data management unit is provided.
前記項目軸は、少なくとも、第1の状態項目と、第2の状態項目と、前記第2の状態項目に比較して前記第1の状態項目に対する相関の強い第3の状態項目とを含み、 The item axis includes at least a first state item, a second state item, and a third state item that has a stronger correlation with the first state item than the second state item.
前記学習用データ管理部は、前記第3の状態項目が前記第1の状態項目及び前記第2の状態項目の間に位置するように前記マップデータを補正する、プラント管理サーバ。 The learning data management unit is a plant management server that corrects the map data so that the third state item is located between the first state item and the second state item.
前記学習用データ管理部に蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを構築するモデル構築部と、を備え、 A model building unit for constructing a model for estimating the state of the plant by machine learning based on the learning data accumulated in the learning data management unit is provided.
前記項目軸は、いずれかの前記状態項目の微分値を示す状態項目を含む、プラント管理サーバ。 The item axis is a plant management server that includes a state item that indicates the derivative of any of the state items.
前記学習用データ管理部に蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを構築するモデル構築部と、を備え、 A model building unit for constructing a model for estimating the state of the plant by machine learning based on the learning data accumulated in the learning data management unit is provided.
前記プラントは、セメントキルンを含み、 The plant contains a cement kiln and
前記項目軸は、前記セメントキルン内の温度に関する状態項目と、前記セメントキルン内のガスの濃度に関する状態項目と、を含み、 The item axis includes a state item relating to the temperature in the cement kiln and a state item relating to the concentration of gas in the cement kiln.
前記教師データは、前記セメントキルン内におけるフリーライムの量に関する評価値を含む、プラント管理サーバ。 The teacher data includes a rating value for the amount of free lime in the cement kiln, a plant management server.
前記モデル構築部は、前記学習用データ選択部によりいずれかの前記学習用データが除外された場合に、残った前記学習用データに基づく機械学習により、前記推定用モデルを再構築するように構成されている、請求項1~4のいずれか一項記載のプラント管理サーバ。 The model building unit is configured to reconstruct the estimation model by machine learning based on the remaining learning data when any of the learning data is excluded by the learning data selection unit. The plant management server according to any one of claims 1 to 4.
前記マップデータに対応する前記プラントの状態を示す教師データを取得するデータ取得部と、
前記マップデータと、当該マップデータに対応する前記教師データとを組み合わせた学習用データに基づく機械学習により構築された前記プラントの状態の推定用モデルに前記マップデータを入力して前記プラントの状態の推定データを導出する推定部と、を備え、
前記マップデータ生成部は、前記マップデータの少なくとも一部の前記データにファジー集合のメンバーシップ関数を適用して前記マップデータを生成する、プラント管理装置。 For each cell specified by the combination of the time interval and the state item in the two-dimensional space of the time axis including a plurality of time intervals arranged in a time series and the item axis including a plurality of state items related to the operation state of the plant. A map data generator that generates map data including data,
A data acquisition unit that acquires teacher data indicating the state of the plant corresponding to the map data, and
The map data is input to the model for estimating the state of the plant constructed by machine learning based on the learning data in which the map data and the teacher data corresponding to the map data are combined, and the state of the plant is changed. Equipped with an estimation unit that derives estimation data ,
The map data generation unit is a plant management device that generates the map data by applying a fuzzy set membership function to at least a part of the map data .
前記マップデータに対応する前記プラントの状態を示す教師データを取得するデータ取得部と、
前記マップデータと、当該マップデータに対応する前記教師データとを組み合わせた学習用データに基づく機械学習により構築された前記プラントの状態の推定用モデルに前記マップデータを入力して前記プラントの状態の推定データを導出する推定部と、を備え、
前記項目軸は、いずれかの前記状態項目の微分値を示す状態項目を含む、プラント管理装置。 For each cell specified by the combination of the time interval and the state item in the two-dimensional space of the time axis including a plurality of time intervals arranged in a time series and the item axis including a plurality of state items related to the operation state of the plant. A map data generator that generates map data including data,
A data acquisition unit that acquires teacher data indicating the state of the plant corresponding to the map data, and
The map data is input to the model for estimating the state of the plant constructed by machine learning based on the learning data in which the map data and the teacher data corresponding to the map data are combined, and the state of the plant is changed. Equipped with an estimation unit that derives estimation data ,
The item axis is a plant management device including a state item indicating a derivative value of any of the state items .
前記マップデータに対応する前記プラントの状態を示す教師データを取得するデータ取得部と、
前記マップデータと、当該マップデータに対応する前記教師データとを組み合わせた学習用データに基づく機械学習により構築された前記プラントの状態の推定用モデルに前記マップデータを入力して前記プラントの状態の推定データを導出する推定部と、を備え、
前記マップデータ生成部は、前記セルのデータのフォーマットが、画素用のデータフォーマットである前記マップデータを生成する、プラント管理装置。 For each cell specified by the combination of the time interval and the state item in the two-dimensional space of the time axis including a plurality of time intervals arranged in a time series and the item axis including a plurality of state items related to the operation state of the plant. A map data generator that generates map data including data,
A data acquisition unit that acquires teacher data indicating the state of the plant corresponding to the map data, and
The map data is input to the model for estimating the state of the plant constructed by machine learning based on the learning data in which the map data and the teacher data corresponding to the map data are combined, and the state of the plant is changed. Equipped with an estimation unit that derives estimation data ,
The map data generation unit is a plant management device that generates the map data in which the data format of the cell is a data format for pixels .
前記マップデータに対応する前記プラントの状態を示す教師データを取得するデータ取得部と、
前記マップデータと、当該マップデータに対応する前記教師データとを組み合わせた学習用データに基づく機械学習により構築された前記プラントの状態の推定用モデルに前記マップデータを入力して前記プラントの状態の推定データを導出する推定部と、を備え、
前記プラントは、セメントキルンを含み、
前記項目軸は、前記セメントキルン内の温度に関する状態項目と、前記セメントキルン内のガスの濃度に関する状態項目と、を含み、
前記教師データは、前記セメントキルン内におけるフリーライムの量に関する評価値を含む、プラント管理装置。 For each cell specified by the combination of the time interval and the state item in the two-dimensional space of the time axis including a plurality of time intervals arranged in a time series and the item axis including a plurality of state items related to the operation state of the plant. A map data generator that generates map data including data,
A data acquisition unit that acquires teacher data indicating the state of the plant corresponding to the map data, and
The map data is input to the model for estimating the state of the plant constructed by machine learning based on the learning data in which the map data and the teacher data corresponding to the map data are combined, and the state of the plant is changed. Equipped with an estimation unit that derives estimation data ,
The plant contains a cement kiln and
The item axis includes a state item relating to the temperature in the cement kiln and a state item relating to the concentration of gas in the cement kiln.
The teacher data is a plant management device including an evaluation value regarding the amount of free lime in the cement kiln .
学習用データ管理部が、少なくとも二つの前記状態項目同士の間で、前記データの変動時刻のずれを縮小するように、当該二つの状態項目間で時間をずらして前記マップデータを補正することと、
モデル構築部が、学習用データ管理部により蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを生成することと、を含む、推定用モデルの生成方法。 The training data management unit combines the time interval and the state item in a two-dimensional space of a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant. Accumulation of training data that combines map data including data for each cell specified by the above and teacher data indicating the state of the plant corresponding to the map data.
The learning data management unit corrects the map data by shifting the time between the two state items so as to reduce the deviation of the fluctuation time of the data between at least two state items. ,
A method for generating an estimation model, which comprises generating a model for estimating the state of the plant by machine learning based on the learning data accumulated by the learning data management unit .
前記学習用データ管理部が、少なくとも二つの前記状態項目同士の間で、前記データの変動時刻のずれを縮小するように、当該二つの状態項目間で時間をずらして前記マップデータを補正することと、
モデル構築部が、学習用データ管理部により蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを生成することと、を含む、推定用モデルの生成方法。 The training data management unit combines the time interval and the state item in a two-dimensional space of a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant. Accumulation of training data that combines map data including data for each cell specified by the above and teacher data indicating the state of the plant corresponding to the map data.
The learning data management unit corrects the map data by shifting the time between the two state items so as to reduce the deviation of the fluctuation time of the data between at least two state items. When,
A method for generating an estimation model, which comprises generating a model for estimating the state of the plant by machine learning based on the learning data accumulated by the learning data management unit .
モデル構築部が、学習用データ管理部により蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを生成することと、を含み、
前記項目軸は、いずれかの前記状態項目の微分値を示す状態項目を含む、推定用モデルの生成方法。 The training data management unit combines the time interval and the state item in a two-dimensional space of a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant. Accumulation of training data that combines map data including data for each cell specified by the above and teacher data indicating the state of the plant corresponding to the map data.
The model building unit includes generating a model for estimating the state of the plant by machine learning based on the learning data accumulated by the learning data management unit .
The item axis is a method of generating an estimation model including a state item indicating a differential value of any of the state items .
モデル構築部が、学習用データ管理部により蓄積された前記学習用データに基づく機械学習により、前記プラントの状態の推定用モデルを生成することと、を含み、
前記プラントは、セメントキルンを含み、
前記項目軸は、前記セメントキルン内の温度に関する状態項目と、前記セメントキルン内のガスの濃度に関する状態項目と、を含み、
前記教師データは、前記セメントキルン内におけるフリーライムの量に関する評価値を含む、推定用モデルの生成方法。 The training data management unit combines the time interval and the state item in a two-dimensional space of a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant. Accumulation of training data that combines map data including data for each cell specified by the above and teacher data indicating the state of the plant corresponding to the map data.
The model building unit includes generating a model for estimating the state of the plant by machine learning based on the learning data accumulated by the learning data management unit .
The plant contains a cement kiln and
The item axis includes a state item relating to the temperature in the cement kiln and a state item relating to the concentration of gas in the cement kiln.
The teacher data is a method of generating an estimation model including an evaluation value regarding the amount of free lime in the cement kiln .
前記モデル構築部が、いずれかの前記学習用データを除外した場合に、残った前記学習用データに基づく機械学習により、前記推定用モデルを再構築することと、を更に含む、請求項13~16のいずれか一項に記載の推定用モデルの生成方法。 When the deviation between the estimated data derived by inputting the map data of the learning data into the estimation model and the teacher data of the learning data exceeds a predetermined range by the learning data selection unit . In addition, excluding the data for the learning
13 to claim 13 , further comprising reconstructing the estimation model by machine learning based on the remaining learning data when the model building unit excludes any of the learning data. 16. The method for generating an estimation model according to any one of 16 .
教師データ生成部が、前記マップデータに対応する前記プラントの状態に関する教師データを、機械学習の出力側データとして生成することと、を含み、
前記マップデータの少なくとも一部の前記データにファジー集合のメンバーシップ関数を適用して前記マップデータを生成する、プラント管理装置による学習用データの生成方法。 In a two-dimensional space where the map data generation unit includes a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant, the combination of the time interval and the state item causes the map data generation unit. Generating map data including data for each specified cell as input side data for machine learning,
The teacher data generation unit includes generating teacher data regarding the state of the plant corresponding to the map data as output side data of machine learning.
A method of generating training data by a plant management device, which applies a fuzzy set membership function to at least a part of the map data to generate the map data .
教師データ生成部が、前記マップデータに対応する前記プラントの状態に関する教師データを、機械学習の出力側データとして生成することと、を含み、
前記項目軸は、いずれかの前記状態項目の微分値を示す状態項目を含む、プラント管理装置による学習用データの生成方法。 In a two-dimensional space where the map data generation unit includes a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant, the combination of the time interval and the state item causes the map data generation unit. Generating map data including data for each specified cell as input side data for machine learning,
The teacher data generation unit includes generating teacher data regarding the state of the plant corresponding to the map data as output side data of machine learning.
The item axis is a method of generating learning data by a plant management device, including a state item indicating a differential value of any of the state items .
教師データ生成部が、前記マップデータに対応する前記プラントの状態に関する教師データを、機械学習の出力側データとして生成することと、を含み、
前記セルのデータのフォーマットが、画素用のデータフォーマットである前記マップデータを生成する、プラント管理装置による学習用データの生成方法。 In a two-dimensional space where the map data generation unit includes a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant, the combination of the time interval and the state item causes the map data generation unit. Generating map data including data for each specified cell as input side data for machine learning,
The teacher data generation unit includes generating teacher data regarding the state of the plant corresponding to the map data as output side data of machine learning.
A method of generating learning data by a plant management device, wherein the data format of the cell is the data format for pixels, and the map data is generated .
教師データ生成部が、前記マップデータに対応する前記プラントの状態に関する教師データを、機械学習の出力側データとして生成することと、を含み、
前記プラントは、セメントキルンを含み、
前記項目軸は、前記セメントキルン内の温度に関する状態項目と、前記セメントキルン内のガスの濃度に関する状態項目と、を含み、
前記教師データは、前記セメントキルン内におけるフリーライムの量に関する評価値を含む、プラント管理装置による学習用データの生成方法。 In a two-dimensional space where the map data generation unit includes a time axis including a plurality of time intervals arranged in a time series and an item axis including a plurality of state items related to the operating state of the plant, the combination of the time interval and the state item causes the map data generation unit. Generating map data including data for each specified cell as input side data for machine learning,
The teacher data generation unit includes generating teacher data regarding the state of the plant corresponding to the map data as output side data of machine learning.
The plant contains a cement kiln and
The item axis includes a state item relating to the temperature in the cement kiln and a state item relating to the concentration of gas in the cement kiln.
The teacher data is a method of generating learning data by a plant management device, including an evaluation value regarding the amount of free lime in the cement kiln .
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2022041688A JP7339382B2 (en) | 2017-12-26 | 2022-03-16 | Plant management system, plant management server, plant management device, estimation model generation method, and learning data generation method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2017248863A JP7043831B2 (en) | 2017-12-26 | 2017-12-26 | Plant management system, plant management server, plant management device, estimation model generation method, and learning data generation method |
JP2022041688A JP7339382B2 (en) | 2017-12-26 | 2022-03-16 | Plant management system, plant management server, plant management device, estimation model generation method, and learning data generation method |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2017248863A Division JP7043831B2 (en) | 2017-12-26 | 2017-12-26 | Plant management system, plant management server, plant management device, estimation model generation method, and learning data generation method |
Publications (3)
Publication Number | Publication Date |
---|---|
JP2022069608A JP2022069608A (en) | 2022-05-11 |
JP2022069608A5 true JP2022069608A5 (en) | 2022-05-20 |
JP7339382B2 JP7339382B2 (en) | 2023-09-05 |
Family
ID=87882220
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP2022041688A Active JP7339382B2 (en) | 2017-12-26 | 2022-03-16 | Plant management system, plant management server, plant management device, estimation model generation method, and learning data generation method |
Country Status (1)
Country | Link |
---|---|
JP (1) | JP7339382B2 (en) |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02186405A (en) * | 1988-09-30 | 1990-07-20 | Omron Tateisi Electron Co | Switching system for controller |
JP2002258942A (en) | 2001-03-01 | 2002-09-13 | Mitsubishi Electric Corp | Monitor data collecting and recording device |
JP6037954B2 (en) * | 2013-06-26 | 2016-12-07 | 株式会社日立製作所 | Boiler tube leak detection device, boiler tube leak detection method, data monitoring center using these, information providing service, and boiler plant. |
JP6216242B2 (en) | 2013-12-13 | 2017-10-18 | 株式会社日立ハイテクノロジーズ | Anomaly detection method and apparatus |
EP3200038A4 (en) | 2014-09-26 | 2018-06-13 | Nec Corporation | Model evaluation device, model evaluation method, and program recording medium |
-
2022
- 2022-03-16 JP JP2022041688A patent/JP7339382B2/en active Active
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Bahamonde et al. | Deceleration versus acceleration universe in different frames of F (R) gravity | |
Monserud et al. | Comparing global vegetation maps with the Kappa statistic | |
Rebelo et al. | Symmetry preserving numerical schemes for partial differential equations and their numerical tests | |
JP6992526B2 (en) | Demand forecasting program, demand forecasting method and demand forecasting device | |
DE112014006015T5 (en) | Display of activity-quantity-related information | |
WO2009058392A3 (en) | User interface framework for viewing large scale graphs | |
Neves et al. | Modellus: Interactive computational modelling to improve teaching of physics in the geosciences | |
CN101776487A (en) | Infrared focal plane non-uniformity fingerprint extraction and image correction method | |
WO2019098158A1 (en) | Plant management system and management device | |
du Bos et al. | Modeling stress-strain curves with neural networks: a scalable alternative to the return mapping algorithm | |
CN103970017A (en) | TS fuzzy control based synchronizing method for fractional order chaotic system and integer order chaotic system | |
CN114326492A (en) | Digital twin virtual-real linkage system of process industrial equipment | |
JP2022069608A5 (en) | ||
Saku et al. | Spatio-temporal prediction of soil deformation in bucket excavation using machine learning | |
CN108197424A (en) | E-learning behavior power-law distribution modelling method of probabilistic in large scale network education | |
Lott et al. | On the use of operations research for decision making with uncertainty for IoT devices in battlefield situations: simulations and outcomes | |
Garzón-Alvarado et al. | Comparative analysis of numerical integration schemes of density equation for a computational model of bone remodelling | |
JPWO2021020339A5 (en) | ||
Levinson | Information technologies as vehicles of evolution | |
CN103700011B (en) | A kind of Feature Extraction Method and device | |
Taiwo et al. | Robust global mantle flow trajectories and their validation via dynamic topography histories | |
Bayoumi et al. | The Fourth Industrial Revolution: Digital Transformation and Industry 4.0 Applied to Product Design, Manufacturing and Operation | |
Kumar et al. | A framework for human efficiency measurement in advanced manufacturing: HCI in INDUSTRY4. 0 | |
Kolesnikov et al. | Modeling and software implementation of fibrous waste disposal processes | |
Wieder | Final SF-425 DOE Award#: DE-SC0014374 |