CN112183805A - Method for predicting state of online inspection result of power transmission line - Google Patents
Method for predicting state of online inspection result of power transmission line Download PDFInfo
- Publication number
- CN112183805A CN112183805A CN201911337935.8A CN201911337935A CN112183805A CN 112183805 A CN112183805 A CN 112183805A CN 201911337935 A CN201911337935 A CN 201911337935A CN 112183805 A CN112183805 A CN 112183805A
- Authority
- CN
- China
- Prior art keywords
- model
- data
- auc
- adopting
- power transmission
- 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
- 238000007689 inspection Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000005540 biological transmission Effects 0.000 title claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000011156 evaluation Methods 0.000 claims abstract description 20
- 238000013145 classification model Methods 0.000 claims abstract description 4
- 238000011478 gradient descent method Methods 0.000 claims abstract description 4
- 238000007477 logistic regression Methods 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims description 11
- 230000005855 radiation Effects 0.000 claims description 5
- 239000011248 coating agent Substances 0.000 claims description 3
- 238000000576 coating method Methods 0.000 claims description 3
- 239000012212 insulator Substances 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000010801 machine learning Methods 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000010276 construction Methods 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 239000004753 textile Substances 0.000 description 1
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of power transmission line inspection, and particularly relates to a method for predicting the state of an online inspection result of a power transmission line. The method mainly comprises the following steps: the method comprises the steps of collecting data regularly, generating training sample data, splitting a sample data set, predicting by adopting a logistic regression binary classification model, establishing a model, solving by adopting a gradient descent method, training the model, evaluating the model by adopting AUC (AUC) evaluation indexes, updating the model, comparing evaluation indexes obtained by training data collected in the previous period with evaluation indexes obtained by currently collecting data, and judging whether the model in the previous period is replaced by the current model according to the size. The invention has the beneficial effects that: the invention adopts the machine learning technology to intelligently evaluate and predict the state of the on-line inspection operation result of the power transmission line so as to support intelligent operation scheduling and plan generation.
Description
Technical Field
The invention belongs to the technical field of power transmission line inspection, and particularly relates to a method for predicting the state of an online inspection result of a power transmission line.
Background
The safety of the power transmission channel is the root of influencing the power supply safety, but the power transmission channel is usually located in complex environments such as original forests and unmanned areas, and therefore, the inspection of the power transmission channel is one of the major and difficult works of relevant departments. Under the traditional mode, patrol to the transmission of electricity passageway mainly goes on with modes such as artifical patrol, regular helicopter, unmanned aerial vehicle patrol, has the degree of difficulty big, with high costs, inefficiency scheduling problem. With the rapid development of the internet of things technology, the network communication technology and the big data technology and the proposal of the national power grid about the ubiquitous power internet of things, the remote inspection mode of the power transmission line is widely applied through the sensors and the online monitoring equipment on the power transmission line.
Because the core of the online patrol operation is to complete the patrol operation by utilizing patrol equipment distributed on a transmission tower base, the online patrol is faced with the problems of large operation quantity, occupied network channels and the like, so that the operation scheduling and plan generation are more difficult. The online patrol operation of the power transmission line is automatically and intelligently realized on the basis of big data and artificial intelligence technology, and the online patrol operation is a trend for solving the problems of low efficiency, insufficient standardization and low intelligence degree in the manual driving operation process.
The on-line patrol operation is to use a plurality of patrol devices distributed on a power transmission channel to complete the periodic or temporary patrol of the channel environment, the tower footing and the components, and the patrol completion standard is as follows: and finishing the shooting of live pictures or video recording according to the standard and intelligently evaluating the health state of the inspection object. However, due to the influence of various factors such as the health state of the equipment, meteorological conditions, operation types, network signal interference and the like, the issued inspection task cannot be successful every time. In order to ensure the intellectualization and the integrity of the data-driven online patrol operation process, it is necessary to predict and evaluate whether the task execution result can be successful before the patrol task is issued.
Machine learning is a technology that learns data, learns knowledge and converts the data into system intelligence, and is widely applied to various industries along with the rise and development of concepts such as data mining, big data, artificial intelligence and the like in recent years. Such as boiler state assessment, textile machine state prediction, internet of things equipment health diagnosis, and the like. These studies and applications show the value of machine learning techniques in the operation of equipment. Meanwhile, in the field of job result prediction, methods such as ensemble learning and decision tree are researched and applied to job scheduling in cloud computing. The research and application do not solve the problems related to feature extraction and model construction in the online patrol operation scene, and lack the discussion of machine learning cold start and model optimization learning mechanism in the operation result state prediction.
Disclosure of Invention
The invention relates to a modeling and predicting method for evaluating result states by machine learning, which aims at solving the problem that the result states (whether the result states are successful) of inspection work are predicted before an inspection work plan is generated by surrounding the characteristics of the field of online inspection of power transmission lines.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for predicting the state of an online patrol result of a power transmission line comprises the following steps:
s1, collecting data regularly, including weather forecast data of a working area, historical online inspection operation execution records of equipment and operation result state data;
s2, generating training sample data, extracting characteristic data from the data collected in the step S1 to form input characteristics, specifically extracting whether rainfall occurs in the daytime, whether the daytime is fine, whether the daytime is cloudy, the highest temperature of the day, the lowest temperature of the day, the highest wind speed of the day, the highest wind direction of the day, the total rainfall amount of the last n days, whether the last n days is rainfall, the average light radiation of the last n days, the task failure proportion of the last 1 month, the minimum electric quantity average value of the last 1 month day, whether ice coating operation is performed, whether mountain fire inspection operation is performed, whether insulator inspection operation is performed, whether inspection operation is performed and the number of tasks form input characteristics x 'from the weather forecast data of the operation area and the execution records of the equipment historical inspection operation on line'iAcquiring job completion result flag data y from job result status dataiThe subscript i refers to the ith job, and the generated training sample data set is data { (x'1,y1),(x′2,y2),…,(x′n,yn) In which yiE {0, 1}, i ═ 1, 2, …, n, n is the total number of samples;
s3, splitting the sample data set, specifically:
s31, randomly scrambling the position index I ═ {1, 2, …, n } of the data set data to obtain a random index I' ═ { I ═ I }1,i2,…,in};
S32, sequentially pair I'r={i|yiR is randomly split into 5 subsets, r ∈ {0, 1}, I ∈ I ', labeled I'r,c,c∈{0,1,…,5};
S33, generating a data subset index I'c=I′0,c∪I′1,c;
S34, generating training set TrainlAnd Test set Testl:
Wherein l, c ∈ {1, 2, …, 5} and l ═ c;
s4, adopting a logistic regression binary classification model to predict, and constructing the model asθTRepresenting the weight vector, the cost function of the model is:
wherein l (θ) represents a log likelihood function, x'iRepresenting the input features of the ith sample, yiA target value representing the ith sample;
s5, setting a target to obtain a parameter theta which enables the cost function J (theta) to be minimum, and solving by adopting a gradient descent method, wherein a parameter updating formula is as follows:
wherein j represents the weight of the jth parameter, alpha is a constant, and Train is setlAnd l is belonged to {1, 2, …, 5}, and parameter solution is carried out to obtain a modell;
S6, evaluating the model by adopting an AUC evaluation index: model will belApplication to Testl(l epsilon {1, 2, …, 5}) data to obtain a prediction result predlAccording to the principleResults of measurement and TestlThe true value of (A) is calculated to obtain an evaluation index auc of the modellAnd obtaining a final evaluation index:
s7, performing parameter solution on all data sets data by adopting the step S5 to obtain a model;
s8, updating the model: in step S1, in order to periodically collect data, the model and the evaluation index obtained by collecting data in the previous period and performing the above steps are defined as a modelold、aucoldDefining the model and the evaluation index obtained by collecting data in the current period and through the steps as a modelnew、aucnewJudgment aucnew>aucoldIf true, use modelnewReplacement modeloldBy modelnewMake a prediction, otherwise use modeloldAnd (6) performing prediction.
The invention has the beneficial effects that: the invention adopts the machine learning technology to intelligently evaluate and predict the state of the on-line inspection operation result of the power transmission line so as to support intelligent operation scheduling and plan generation.
Detailed Description
The present invention is described in further detail below.
The invention solves the problems of data acquisition, characteristic engineering, target variable processing and the like aiming at the service scene and model requirements of the on-line inspection operation result state prediction.
1. Data acquisition
Around the requirement of the prediction modeling of the work result state, the data acquisition is realized by acquiring factor data influencing the work completion and data of the work completion result state during the execution of the historical inspection task.
1) Factor data affecting job completion mainly include:
A. weather forecast data is acquired by a weather department about a working area for a period of time in the future (7-10 days), including: temperature, day weather type, night weather type, humidity, wind speed, wind direction, light radiation data.
B. The state perception and stability of the operation equipment are mainly measured by performing record on historical online inspection operation of the equipment and monitoring heartbeat data of the equipment.
C. The three basic attributes of the job are job type, task amount and tour object, and the data mainly comes from the job itself.
D. And the communication network factor data comprises networking topological relation data among the patrol equipment.
2) The job result status data is a target value for recording the completion of the job task performed in history, performing model training and learning, and is denoted as yi。
And forming training data for model training by collecting the data in 1) and 2) during the historical online patrol operation. And when the data can not meet the model training, generating expert experience estimation of the operation completion condition by the service expert according to the data in the step 1), and finishing the subsequent inspection task plan formulation.
2. Feature engineering
In the invention, the characteristic engineering mainly carries out characteristic design extraction and characteristic data preprocessing around the requirement of the work completion result state prediction.
1) Feature design and extraction
Designing and extracting features based on collected factor data influencing job completion, wherein the feature is marked as xi,k(ith job, kth feature).
A. Weather forecast type characteristics: day weather type, day maximum/minimum temperature, day maximum wind speed, day main wind direction, day total rainfall n days before, whether it rains n days before, average light radiation n days before.
B. The stability characteristic of the operation equipment is as follows: task failure rate of last 1 month, daily minimum charge average of last 1 month.
C. And the job attribute characteristics comprise job types and task numbers.
2) Feature data preprocessing
xi,kThe types of (i, k ∈ N) include character type, numerical type, and Boolean type. The invention carries out corresponding preprocessing aiming at different types, and the converted characteristic data is recorded as x'i,k。
A. Character-type features, such as day-day weather type, contain "sunny", "rainy", "cloudy" attribute values. The invention adopts two steps of sparse attribute value combination and one-hot coding to process the preprocessing of the characteristic indexes, thereby ensuring the consistency of processing results in model training and prediction application.
B. Logarithmic characteristics, such as a maximum temperature of 37 ℃ for the current day and a total rainfall of 0 ml for the previous n days. The invention adopts a min-max normalization method to perform dimensionless processing on the characteristics.
C. Boolean-type characteristics, such as whether it was raining the previous n days. In the present invention, the treatment is not performed separately.
Final formation of input feature x'iThe method comprises the following steps: whether the vehicle is rained in the daytime, whether the vehicle is sunny or not in the daytime, whether the vehicle is cloudy or not in the daytime, the highest temperature in the day, the lowest temperature in the day, the highest wind speed in the day, the highest wind direction in the day, the total rainfall in the previous n days, whether the vehicle is rained in the previous n days, the average light radiation in the previous n days, the task failure proportion in the last 1 month, the lowest electric quantity average value in the last 1 month, whether the vehicle is subjected to ice coating operation, whether mountain fire inspection operation is performed, whether insulator inspection operation is performed, whether inspection operation is.
3) Training sample data set generation
Converting feature data x'iAnd job completion result flag data yiPerforming correlation matching, and generating a training sample data set data { (x {'1,y1),(x′2,y2),…,(x′n,yn)}. Wherein, yiE {0, 1} (i ═ 1, 2, …, n), n is the total number of samples.
3. Model construction and evaluation
According to the method, model construction is carried out on the data by adopting a supervised learning technology around the target of on-line patrol operation result state prediction.
1) Data splitting
The invention adopts a 5-fold cross test method to evaluate the model. Because the proportion of negative samples (the operation state is incomplete) in the training data is very small, the invention adopts a layered random splitting mode to split the data set. The specific process is as follows:
A. the position index I of the data set data is randomly scrambled {1, 2, …, n }. Obtaining a random index of I' ═ I1,i2,…,in}(ij∈I)。
B. In turn to I'r={i|yiR (r ∈ {0, 1}, I ∈ I ') is randomly split into 5 subsets, labeled I'r,c(c∈{0,1,…,5})。
C. Generating data subset index I'c=I′0,c∪I′1,c(c∈{1,…,5})。
2) Model construction
The invention adopts a logistic regression binary classification model to predict the result state and construct a model(θTRepresenting weight vector)
Cost function of the model:
wherein h isθ(x ') represents the established model, theta is a weight parameter of the model, l (theta) represents a log-likelihood function, x'iRepresenting the input features of the ith sample, yiRepresenting the target value for the ith sample.
3) Model training
And (3) generating a training set and a testing set by using the result of data splitting:
where l, c ∈ {1, 2, …, 5} and l ═ c.
A. Parameter solving
In order to obtain a parameter theta which enables the cost function J (theta) to be minimum, the method adopts a gradient descent method to solve, and a parameter updating formula is as follows:
where j represents the weight of the jth parameter, n represents the total number of samples, and α is a constant.
In the training set TrainlAnd l is belonged to {1, 2, …, 5}, and parameter solution is carried out to obtain a modell。
B. Model evaluation
Due to the unbalance of the proportion of the positive sample and the negative sample, the method adopts AUC evaluation indexes to evaluate the model. The specific process is as follows: model will belApplication to Testl(l epsilon {1, 2, …, 5}) data to obtain a prediction result predlBased on the prediction result and TestlThe true value of (A) is calculated to obtain an evaluation index auc of the modell。
C. Final model
And performing parameter solution on all the data sets data by using the parameter solution step A to obtain the model.
4. Model management and publishing mechanism
1) Model iterative training
Corresponding sample data can be gradually increased along with the increase of the on-line patrol task, and the model is trained and iterated by regularly utilizing the process. Evaluation indexes of the generated model and the 5-fold cross validation are respectively model and auc.
2) Model online deployment mechanism
If the original current application model and the evaluation index are respectively modelonline、auconline. Then:
when auc is more than auconlineIn a model replace modelonline。
Claims (1)
1. A method for predicting the state of an online patrol result of a power transmission line is characterized by comprising the following steps:
s1, collecting data regularly, including weather forecast data of a working area, historical online inspection operation execution records of equipment and operation result state data;
s2, generating training sample data, extracting characteristic data from the data collected in the step S1 to form input characteristics, specifically extracting whether rainfall occurs in the daytime, whether the daytime is fine, whether the daytime is cloudy, the highest temperature of the day, the lowest temperature of the day, the highest wind speed of the day, the highest wind direction of the day, the total rainfall amount of the last n days, whether the last n days is rainfall, the average light radiation of the last n days, the task failure proportion of the last 1 month, the minimum electric quantity average value of the last 1 month day, whether ice coating operation is performed, whether mountain fire inspection operation is performed, whether insulator inspection operation is performed, whether inspection operation is performed and the number of tasks form input characteristics x 'from the weather forecast data of the operation area and the execution records of the equipment historical inspection operation on line'iAcquiring job completion result flag data y from job result status dataiThe subscript i refers to the ith job, and the generated training sample data set is data { (x'1,y1),(x′2,y2),…,(x′n,yn) In which yiE {0, 1}, i ═ 1, 2, …, n, n is the total number of samples;
s3, splitting the sample data set, specifically:
s31, randomly scrambling the position index I ═ {1, 2, …, n } of the data set data to obtain a random index I' ═ { I ═ I }1,i2,…,in};
S32, sequentially pair I'r={i|yiR is randomly split into 5 subsets, r ∈ {0, 1}, I ∈ I ', labeled I'r,c,c∈{0,1,…,5};
S33, generating a data subset index I'c=I′0,c∪I′1,c;
S34, generating training set TrainlAnd Test set Testl:
Wherein l, c ∈ {1, 2, …, 5} and l ═ c;
s4, adopting a logistic regression binary classification model to predict, and constructing the model asθTRepresenting the weight vector, the cost function of the model is:
wherein l (θ) represents a log likelihood function, x'iRepresenting the input features of the ith sample, yiA target value representing the ith sample;
s5, setting a target to obtain a parameter theta which enables the cost function J (theta) to be minimum, and solving by adopting a gradient descent method, wherein a parameter updating formula is as follows:
wherein j represents the weight of the jth parameter, alpha is a constant, and Train is setlAnd l is belonged to {1, 2, …, 5}, and parameter solution is carried out to obtain a modell;
S6, evaluating the model by adopting an AUC evaluation index: model will belApplication to Testl(l epsilon {1, 2, …, 5}) data to obtain a prediction result predlBased on the prediction result and TestlThe true value of (A) is calculated to obtain an evaluation index auc of the modellAnd obtaining a final evaluation index:
s7, performing parameter solution on all data sets data by adopting the step S5 to obtain a model;
s8, updating the model: in step S1, in order to periodically collect data, the model and the evaluation index obtained by the data collected in the previous period and obtained through the above steps are defined as a modelold、aucoldDefining the model and the evaluation index obtained by the data collected in the current period and the steps as the modelnew、aucnewJudgment aucnew>aucoldIf true, use modelnewReplacement modeloldBy modelnewMake a prediction, otherwise use modeloldAnd (6) performing prediction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911337935.8A CN112183805B (en) | 2019-12-23 | 2019-12-23 | Prediction method for online inspection result state of power transmission line |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911337935.8A CN112183805B (en) | 2019-12-23 | 2019-12-23 | Prediction method for online inspection result state of power transmission line |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112183805A true CN112183805A (en) | 2021-01-05 |
CN112183805B CN112183805B (en) | 2023-10-24 |
Family
ID=73918780
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911337935.8A Active CN112183805B (en) | 2019-12-23 | 2019-12-23 | Prediction method for online inspection result state of power transmission line |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112183805B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113902314A (en) * | 2021-10-14 | 2022-01-07 | 成都卡普数据服务有限责任公司 | Expert knowledge fused power transmission line environment risk assessment method |
CN118245723A (en) * | 2024-05-24 | 2024-06-25 | 上海海能信息科技股份有限公司 | Real-time thermal equivalent wind speed calculation method for power transmission line |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103455708A (en) * | 2013-07-24 | 2013-12-18 | 安徽省电力科学研究院 | Power transmission line disaster monitoring and risk assessment platform based on satellite and weather information |
CN105989542A (en) * | 2015-01-30 | 2016-10-05 | 国家电网公司 | Relay protection state online evaluation method based on fuzzy support vector machine |
CN106786560A (en) * | 2017-02-14 | 2017-05-31 | 中国电力科学研究院 | A kind of power system stability characteristic automatic extraction method and device |
CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O&M intelligent monitoring system and method |
CN109003254A (en) * | 2018-05-28 | 2018-12-14 | 南方电网科学研究院有限责任公司 | Method, device, equipment, system and medium for detecting icing based on logistic regression |
US20190370684A1 (en) * | 2018-06-01 | 2019-12-05 | Sas Institute Inc. | System for automatic, simultaneous feature selection and hyperparameter tuning for a machine learning model |
US20190377625A1 (en) * | 2018-06-08 | 2019-12-12 | Microsoft Technology Licensing, Llc | Computing node failure and health prediction for cloud-based data center |
-
2019
- 2019-12-23 CN CN201911337935.8A patent/CN112183805B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103455708A (en) * | 2013-07-24 | 2013-12-18 | 安徽省电力科学研究院 | Power transmission line disaster monitoring and risk assessment platform based on satellite and weather information |
CN105989542A (en) * | 2015-01-30 | 2016-10-05 | 国家电网公司 | Relay protection state online evaluation method based on fuzzy support vector machine |
CN106786560A (en) * | 2017-02-14 | 2017-05-31 | 中国电力科学研究院 | A kind of power system stability characteristic automatic extraction method and device |
CN108174165A (en) * | 2018-01-17 | 2018-06-15 | 重庆览辉信息技术有限公司 | Electric power safety operation and O&M intelligent monitoring system and method |
CN109003254A (en) * | 2018-05-28 | 2018-12-14 | 南方电网科学研究院有限责任公司 | Method, device, equipment, system and medium for detecting icing based on logistic regression |
US20190370684A1 (en) * | 2018-06-01 | 2019-12-05 | Sas Institute Inc. | System for automatic, simultaneous feature selection and hyperparameter tuning for a machine learning model |
US20190377625A1 (en) * | 2018-06-08 | 2019-12-12 | Microsoft Technology Licensing, Llc | Computing node failure and health prediction for cloud-based data center |
Non-Patent Citations (1)
Title |
---|
杨航: "城市电网输变电设备状态检修在线决策系统研究与应用" * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113902314A (en) * | 2021-10-14 | 2022-01-07 | 成都卡普数据服务有限责任公司 | Expert knowledge fused power transmission line environment risk assessment method |
CN118245723A (en) * | 2024-05-24 | 2024-06-25 | 上海海能信息科技股份有限公司 | Real-time thermal equivalent wind speed calculation method for power transmission line |
CN118245723B (en) * | 2024-05-24 | 2024-08-16 | 上海海能信息科技股份有限公司 | Real-time thermal equivalent wind speed calculation method for power transmission line |
Also Published As
Publication number | Publication date |
---|---|
CN112183805B (en) | 2023-10-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107341349B (en) | Method and system for evaluating health of fan, memory and controller | |
CN108320043B (en) | Power distribution network equipment state diagnosis and prediction method based on electric power big data | |
CN112749904B (en) | Power distribution network fault risk early warning method and system based on deep learning | |
CN110533331B (en) | Fault early warning method and system based on transmission line data mining | |
CN110689171A (en) | Turbine health state prediction method based on E-LSTM | |
CN102982393B (en) | A kind of on-line prediction method of electric transmission line dynamic capacity | |
CN113962988A (en) | Power inspection image anomaly detection method and system based on federal learning | |
CN111160626B (en) | Power load time sequence control method based on decomposition fusion | |
CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
CN112183805B (en) | Prediction method for online inspection result state of power transmission line | |
CN116742799A (en) | Auxiliary power distribution monitoring and early warning system based on Internet of things technology | |
CN115796434A (en) | Management and control method and device for power distribution network, electronic equipment and storage medium | |
CN110766313A (en) | Cable tunnel comprehensive state evaluation method based on operation and maintenance system | |
CN112100904B (en) | ICOA-BPNN-based distributed photovoltaic power station active power virtual acquisition method | |
CN116599151A (en) | Source network storage safety management method based on multi-source data | |
CN112819189A (en) | Wind power output prediction method based on historical predicted value | |
CN116796194A (en) | IDBO-KELM-BiGRU neural network-based active power virtual collection method for distributed photovoltaic power station | |
CN109165770A (en) | Novel photovoltaic power forecasting method based on AFSA-Elman | |
CN112215410A (en) | Power load prediction method based on improved deep learning | |
CN116245418A (en) | Intelligent construction site safety risk assessment method based on data mining | |
CN102175202A (en) | Ice cover thickness predicting method based on fuzzy logic | |
CN113762591B (en) | Short-term electric quantity prediction method and system based on GRU and multi-core SVM countermeasure learning | |
CN118071205A (en) | Self-adaptive control system for raw material like coke for lithium battery cathode material | |
CN104298706A (en) | Transmission tower material actual strength computing method based on data mining | |
CN117474156A (en) | Data-driven-based electric vehicle charging demand prediction method, device and medium |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |