CN114118479A - Bidirectional RNN-based power project abnormity early warning processing method - Google Patents
Bidirectional RNN-based power project abnormity early warning processing method Download PDFInfo
- Publication number
- CN114118479A CN114118479A CN202111450533.6A CN202111450533A CN114118479A CN 114118479 A CN114118479 A CN 114118479A CN 202111450533 A CN202111450533 A CN 202111450533A CN 114118479 A CN114118479 A CN 114118479A
- Authority
- CN
- China
- Prior art keywords
- project
- abnormal
- early warning
- electric power
- 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.)
- Pending
Links
- 230000002457 bidirectional effect Effects 0.000 title claims abstract description 20
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 230000002159 abnormal effect Effects 0.000 claims abstract description 63
- 238000013135 deep learning Methods 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 22
- 230000008447 perception Effects 0.000 claims abstract description 4
- 238000013480 data collection Methods 0.000 claims abstract 3
- 238000004458 analytical method Methods 0.000 claims description 14
- 230000005856 abnormality Effects 0.000 claims description 9
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000004088 simulation Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 206010063385 Intellectualisation Diseases 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The electric power project abnormity early warning processing method based on the bidirectional RNN comprises the steps of conducting historical data collection on an electric power project, judging whether the electric power project is normal or abnormal according to expert experience, extracting features of a normal project and an abnormal project through a deep learning method, classifying the extracted features, establishing an online early warning model based on deep learning according to the features and the classification, conducting abnormity perception on current electric power project data, and conducting early warning if the electric power project data are abnormal. The method disclosed by the invention integrates deep learning and manual evaluation, overcomes the problem of dynamic fluctuation of normal and abnormal projects, is beneficial to helping managers to make decisions on electric power projects effectively and reasonably in real time, and optimizes and deepens the decision flow of real-time monitoring of the projects.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, relates to exception handling of data, and discloses a bidirectional RNN-based power project exception early warning handling method.
Background
With the continuous promotion of the industrialization process, the power demand of China is increasing day by day, and the power project is coming, so that the serious management problem is brought, if personnel and time cannot be well distributed and the real-time progress is mastered, the project efficiency can be greatly reduced, and a large amount of resources can be wasted along with the accumulation of abnormal projects. In order to help a manager to find abnormal projects in time and automatically analyze mass data, so that personnel and time are arranged more reasonably and effectively and project progress is mastered, a method capable of performing abnormity early warning processing on project data is needed.
In reality, projects are mainly managed in a manual judgment mode, personnel and time thresholds are set according to manual experience, and whether the projects are within the threshold range is judged according to the actual conditions of the projects, so that abnormal projects are identified.
Deep learning is a nontrivial process that reveals implicit, previously unknown, and potentially valuable information from the vast amount of data in a database. Deep learning is a business process for detecting a large amount of data to find meaningful patterns (patterns) and rules (rule), and is mainly based on artificial intelligence, machine learning, pattern recognition, statistics, databases, visualization technologies and the like, and the deep learning is used for analyzing data of enterprises in a highly automated manner, making inductive reasoning, mining potential patterns from the data, helping decision makers to adjust market strategies, reducing risks and making correct decisions.
Disclosure of Invention
The invention aims to solve the problem that a large amount of power project data are difficult to analyze only by means of expert experience, and provides a data abnormity early warning processing method combining the expert experience and artificial intelligence.
The technical scheme of the invention is as follows: the electric power project abnormity early warning processing method based on the bidirectional RNN systematically collects historical data of an electric power project, wherein the historical data comprises the number requirement, time arrangement and progress condition of the project, the electric power project is judged to be normal or abnormal according to expert experience, the characteristics of a normal project and an abnormal project are extracted through a deep learning method, the extracted characteristics are classified, an online early warning model based on deep learning is established according to the characteristics and the classification and is used for carrying out abnormity perception on the current electric power project data, if the electric power project is abnormal, early warning is carried out,
the deep learning adopts a bidirectional RNN neural network, after historical project data are learned, simulation analysis of different conditions is carried out on the projects by changing project parameters, if the parameters are abnormal, an alarm is given, and after the abnormality is adjusted by a manager, the simulation is carried out until the abnormality does not occur any more, so that dynamic learning of characteristics of normal projects and abnormal projects is completed.
Preferably, the method comprises the following steps:
step one, data acquisition and condition analysis: collecting historical data of the power project, including time, personnel and progress of the project, analyzing and judging the running state of the project by combining with expert experience, and determining whether the project is normal or abnormal;
step two, characteristic analysis and problem definition: analyzing the data characteristics of the power project by adopting a bidirectional RNN neural network, extracting the characteristics of a normal project and an abnormal project, and classifying the abnormal data of the project according to the abnormal characteristics;
step three, model establishment and exception reporting: and after learning is finished, the bidirectional RNN neural network obtains an online early warning model, senses abnormal items in real time by combining the characteristics and classification of the current items, reports the abnormal items if abnormal, and performs early warning processing.
The innovation points of the invention are as follows:
(1) deep learning combines artificial experience with rules.
And processing the abnormal data types of the abnormally sensed items through item features and corresponding classification rules according to the item data, and constructing an abnormal item sensing model based on deep learning by combining expert experience to define the range of normal item data and abnormal item data.
(2) And (4) bidirectional RNN early warning decision analysis.
The method updates the project definition range, dynamically identifies the abnormal data of the project, thereby improving the reliability and scientificity of the abnormal identification of the project, and finally judges the early warning content and the possible abnormal reason through early warning decision analysis.
The traditional RNN is a unidirectional neural network, the current output is determined by the current input and the output at the last moment, and the importance degree of each position is not distinguished, so that the modeling capability of the traditional RNN on sequence data is limited. Learning data such as time, number of people, progress and the like by using the bidirectional RNN, then performing simulation analysis on the data by inputting parameters, and if 1 parameter is abnormal, sending a primary alarm; if 2 parameters are abnormal, a secondary alarm is sent out; if 3 or more than 3 parameters are abnormal, a three-level alarm is sent. And the manager continues to simulate after processing abnormal data, so that secondary abnormality is avoided, and the manager is helped to accurately master the condition of the project.
The invention provides a novel and effective model for solving the early warning treatment of project abnormity in the power environment, the model integrates deep learning and manual judgment, overcomes the problem of dynamic fluctuation of normal and abnormal projects, is beneficial to helping managers to make decisions on power projects in real time, effectively and reasonably, optimizes and deepens the decision flow of real-time project monitoring, and provides a new idea for classification identification of abnormal project data and solving the multi-classification problem.
The method can be applied to an electric power project monitoring system, assist in monitoring the state and progress of a project in real time, definitely identify abnormal data, guide the normal operation of the project according to the learned normal project characteristics, improve the execution efficiency, realize the optimal control of the project and promote the intellectualization and refinement of management. The method of the invention changes project data from 'detection' to 'mining', and changes project management from 'target management' to 'process management', thereby providing a constructive suggestion for the continuation of deep learning in the field of electric power science and technology and promoting the establishment of a comprehensive management platform.
Drawings
Fig. 1 is a flowchart of an abnormality warning processing method according to the present invention.
FIG. 2 is a schematic diagram of an on-line early warning model established by the bidirectional RNN neural network of the present invention.
Detailed Description
The method combines a deep learning-based method with manual evaluation, helps managers to effectively monitor project progress in real time, definitely identify abnormal data, guide normal operation of projects, improve execution efficiency, realize optimal control of projects, and promote intellectualization and refinement of management.
The invention provides a bidirectional RNN-based electric power project abnormity early warning processing method, which systematically acquires data of an electric power project, wherein the data comprises the number requirement, time arrangement and progress condition of the project, deeply knows the characteristics of a normal project and an abnormal project by means of a deep learning method, analyzes the complexity of the project abnormity early warning processing, provides a project abnormity data classification method, establishes an online early warning model based on deep learning, designs an abnormal project real-time perception method by combining the characteristics and classification of the abnormal project, can dynamically capture abnormal project data, timely reports the abnormal condition, helps a manager to timely find the abnormal project, reasonably distributes personnel and time, and grasps the project progress.
The bidirectional RNN-based power project abnormity early warning processing method realizes the following steps:
step one, data acquisition and condition analysis: relevant data of the power project, including time, personnel and progress of the project, are collected through literature reference, field investigation and other modes, subjective analysis and judgment are carried out on the operation state of the project, and whether the project is normal or abnormal is confirmed.
Step two, characteristic analysis and problem definition: the data characteristics of the electric power project are analyzed in a deep learning mode, the normal project and the abnormal project are included, the complexity of project abnormity early warning processing is analyzed, and a project abnormity data classification method is provided.
Step three, model establishment and exception reporting: on the basis of theoretical research and problem analysis, an online early warning model based on deep learning is established, an abnormal project real-time sensing method is designed by combining the characteristics and classification of projects, the state of the projects can be mastered in real time, and if the abnormal projects exist, the abnormal projects are reported to perform early warning processing.
The process of the invention is shown in fig. 1, the problem of electric power project abnormity early warning is to help managers to better optimize projects, firstly, the electric power project abnormity early warning is simulated and analyzed by inputting parameters according to project monitoring states or real-time information of each project, such as personnel, time, progress and the like, and if 1 parameter is abnormal, a first-level alarm is sent; if 2 parameters are abnormal, a secondary alarm is sent out; if 3 or more than 3 parameters are abnormal, a three-level alarm is sent. After sensing the abnormality, performing project decision analysis on the abnormality by combining historical project data and real project elements, then performing project abnormality diagnosis with a manager, judging whether an abnormal project exists, finally performing project optimization on the whole scientific and technological project, and providing related data for later data acquisition.
Specifically, as shown in fig. 2, an electric power project abnormity online early warning model based on the bidirectional RNN is characterized in that real-time project data and characteristic attributes are input, normal data and abnormal data of a project are identified by utilizing a deep learning technology and rules respectively, artificial experience is fused to participate in alarm decision, the type of abnormity of the project, the number of people, time or progress is judged, abnormal conditions are reported in time, a manager is helped to find the abnormal project in time, and resources and time are distributed reasonably.
Claims (3)
1. A bidirectional RNN-based electric power project abnormity early warning processing method is characterized in that historical data collection is systematically carried out on an electric power project, the historical data collection comprises the number requirement, time arrangement and progress condition of the project, whether the project is normal or abnormal is judged according to expert experience, the characteristics of a normal project and an abnormal project are extracted through a deep learning method, the extracted characteristics are classified, an online early warning model based on deep learning is established according to the characteristics and the classification and is used for carrying out abnormity perception on the current electric power project data, early warning is carried out if the project is abnormal,
the deep learning adopts a bidirectional RNN neural network, after historical project data are learned, simulation analysis of different conditions is carried out on the projects by changing project parameters, if the parameters are abnormal, an alarm is given, and after the abnormality is adjusted by a manager, the simulation is carried out until the abnormality does not occur any more, so that dynamic learning of characteristics of normal projects and abnormal projects is completed.
2. The bidirectional RNN-based power project abnormity early warning processing method according to claim 1, comprising the following steps:
step one, data acquisition and condition analysis: collecting historical data of the power project, including time, personnel and progress of the project, analyzing and judging the running state of the project by combining with expert experience, and determining whether the project is normal or abnormal;
step two, characteristic analysis and problem definition: analyzing the data characteristics of the power project by adopting a bidirectional RNN neural network, extracting the characteristics of a normal project and an abnormal project, and classifying the abnormal data of the project according to the abnormal characteristics;
step three, model establishment and exception reporting: and after learning is finished, the bidirectional RNN neural network obtains an online early warning model, senses abnormal items in real time by combining the characteristics and classification of the current items, reports the abnormal items if abnormal, and performs early warning processing.
3. The bidirectional RNN-based electric power project abnormity early warning processing method as claimed in claim 1, wherein the online early warning model monitors the electric power project in real time, if abnormity early warning exists, a manager adjusts project parameters according to abnormity characteristics and monitors the project until the project is normal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111450533.6A CN114118479A (en) | 2021-11-29 | 2021-11-29 | Bidirectional RNN-based power project abnormity early warning processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111450533.6A CN114118479A (en) | 2021-11-29 | 2021-11-29 | Bidirectional RNN-based power project abnormity early warning processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114118479A true CN114118479A (en) | 2022-03-01 |
Family
ID=80368947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111450533.6A Pending CN114118479A (en) | 2021-11-29 | 2021-11-29 | Bidirectional RNN-based power project abnormity early warning processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114118479A (en) |
-
2021
- 2021-11-29 CN CN202111450533.6A patent/CN114118479A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112926749A (en) | Intelligent power grid information equipment monitoring system and method | |
| WO2023142424A1 (en) | Power financial service risk control method and system based on gru-lstm neural network | |
| WO2021007871A1 (en) | Alumina production operation optimization system and method employing cloud-side collaboration | |
| CN112085261A (en) | Enterprise production status diagnosis method based on cloud fusion and digital twin technology | |
| CN108958215A (en) | A kind of engineering truck failure prediction system and its prediction technique based on data mining | |
| CN113627032B (en) | Intelligent decision-making method for equipment design/maintenance scheme based on digital twinning | |
| CN116028887B (en) | Analysis method of continuous industrial production data | |
| CN115511367B (en) | Intelligent quality management system of production line | |
| CN117348503B (en) | Propylene oxide production data monitoring system and method | |
| CN114689965A (en) | Power equipment online monitoring method and system based on embedded intelligent sensor | |
| CN116991130A (en) | Intelligent automatic control system and method for petrochemical production | |
| CN114418177B (en) | New product material distribution prediction method based on digital twin workshops for generating countermeasure network | |
| CN115760024A (en) | Intelligent building management and control platform based on BIM | |
| CN117234785B (en) | Centralized control platform error analysis system based on artificial intelligence self-query | |
| CN113065798A (en) | Safety production risk early warning method and system | |
| CN117390944A (en) | Substation operation condition simulation system | |
| CN117172751A (en) | Construction method of intelligent operation and maintenance information analysis model | |
| CN117113135A (en) | Carbon emission anomaly monitoring and analyzing system capable of sorting and classifying anomaly data | |
| CN117240887A (en) | Wisdom thing networking energy management platform system | |
| CN114118479A (en) | Bidirectional RNN-based power project abnormity early warning processing method | |
| CN116880402A (en) | Intelligent factory cooperative control system and method thereof | |
| CN116704729A (en) | Industrial kiln early warning system and method based on big data analysis | |
| CN115660288A (en) | Analysis management system based on internet big data | |
| CN115987692A (en) | Safety protection system and method based on flow backtracking analysis | |
| CN109978396A (en) | A kind of early screening system and method for risk case |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |