CN109636029B - Power distribution network medium-short term voltage out-of-limit early warning method based on big data - Google Patents
Power distribution network medium-short term voltage out-of-limit early warning method based on big data Download PDFInfo
- Publication number
- CN109636029B CN109636029B CN201811500365.5A CN201811500365A CN109636029B CN 109636029 B CN109636029 B CN 109636029B CN 201811500365 A CN201811500365 A CN 201811500365A CN 109636029 B CN109636029 B CN 109636029B
- Authority
- CN
- China
- Prior art keywords
- limit
- voltage
- month
- data
- results
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000007637 random forest analysis Methods 0.000 claims abstract description 41
- 238000003066 decision tree Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 29
- 238000007477 logistic regression Methods 0.000 claims description 12
- 238000011160 research Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000005457 optimization Methods 0.000 abstract description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012360 testing method 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
- G06Q10/04—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
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A power distribution network medium and short term voltage out-of-limit early warning method based on big data is disclosed. Relates to the technical field of electric power system early warning. The large data-based power distribution network medium and short term voltage out-of-limit early warning method is clear in logic, orderly in steps and capable of well and effectively judging whether the medium and short term voltage is out-of-limit or not. According to the method, multi-dimensional related data are analyzed by fusing the voltage risk of the power distribution network, the out-of-limit prediction of the medium-short-term voltage of a single distribution transformer under the influence of multiple factors is realized by using a random forest technology, the out-of-limit risk type of the medium-short-term voltage of the single distribution transformer is identified, the out-of-limit risk probability is given, and the optimization and treatment capacity of the out-of-limit problem of the medium-short-term power distribution network voltage is. The method has the advantages of clear logic, ordered steps and capability of well and effectively judging whether the voltage in the middle-short period exceeds the limit or not.
Description
Technical Field
The invention relates to the technical field of electric power system early warning.
Background
At present, people research the voltage control technology of a power distribution network, and the development of the technology is mainly divided into three stages, namely a traditional reactive voltage control stage without considering the distributed power supply, a distributed autonomous control stage with considering the distributed power supply and a centralized cooperative control stage with considering the distributed power supply. Power grid companies and research institutions of various countries fully recognize the influence of factors such as insufficient reactive power configuration, voltage fluctuation caused by active load fluctuation and distributed power supply access on the voltage risk of the power distribution network, and make certain research on the basic theory of voltage risk judgment. However, the voltage risk analysis and prediction technology of the power distribution network driven by the data of the power distribution automation system still needs to be further researched. The voltage risk refers to the voltage fluctuation beyond the normal 198V-242V, and the voltage is out of limit, so that the operation of the power grid is influenced.
Disclosure of Invention
Aiming at the problems, the invention provides the power distribution network medium and short term voltage out-of-limit early warning method based on the big data, which has clear logic and orderly steps and can well and effectively judge whether the voltage in the medium and short term is out of limit.
The technical scheme of the invention is as follows: the method comprises the following steps:
1) analyzing and researching the influence factors of the voltage out-of-limit: through the guidance and research results of earlier scientific research documents, the method for drawing out relevant factors of voltage out-of-limit mainly comprises the following steps: the system comprises a distribution transformer capacity, a distribution transformer power, a distribution transformer voltage, a bus voltage, a feeder power, a feeder capacity, a power factor, a power distance, a bus power supply radius and a climate temperature;
2) analysis of major causes of out-of-limit: the method comprises the following steps of calling relevant data of a transformer area with overvoltage threshold crossing, and establishing a logistic regression model of the voltage threshold crossing phenomenon and each relevant factor by utilizing the relevant data:
wherein p is the current month out-of-limit probability, x is the relevant factor of the last month voltage out-of-limit, namely whether the distribution transformer is out-of-limit in a certain period of time is taken as an output variable, the fluctuation characteristics of all the influencing factors of the transformer in the previous period of time are taken as input variables and are sequentially input into a logistic regression model, the logistic regression model is adopted for training and learning, the correlation degree is counted, and the main cause of out-of-limit of each transformer area is analyzed;
the average value of the bus voltage, the average value of the distribution and transformation voltage, the maximum difference value of the distribution and transformation power and the maximum difference value of the feeder line power have the maximum relevance with the out-of-limit, and are the main cause of the out-of-limit;
3) extracting, preprocessing and fusing data of main causes related to the voltage risk of the distribution network to a distribution network voltage risk prejudgment analysis characteristic table;
4) predicting the first future month:
4.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the last month and data of the current month out-of-limit condition into a random forest model, and establishing a plurality of decision trees I;
4.2), predicting: inputting data of main causes of the voltage threshold crossing in the current month into all decision trees I for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage threshold crossing in the next month or not;
5) predicting the second future month:
5.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the second month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees II;
5.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees II for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the second month in the future;
6) predicting the third future month:
6.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the third month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees III;
6.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees III for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage is out-of-limit in the third month in the future;
7) predicting the fourth future month:
7.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fourth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees IV;
7.2), prediction: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees IV for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage out-of-limit in the fourth month exists in the future;
8) predicting the fifth future month:
8.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fifth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
8.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees V for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage in the fifth month is out-of-limit or not in the future;
9) predicting the future sixth month:
9.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the sixth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
9.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees VI for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the future sixth month or not;
10) and result integration: counting the prediction result of whether the voltage exceeds the limit for one to six months in the future; and (6) finishing.
And 2) establishing a logistic regression model for multiple times according to the relevant factors of voltage out-of-limit in the step 2), wherein x is used as one of distribution transformer capacity, distribution transformer power, distribution transformer voltage, bus voltage, feeder power, feeder capacity, power factors, power source distance, bus power supply radius and climate temperature every time.
According to the method, multi-dimensional related data are analyzed by fusing the voltage risk of the power distribution network, the out-of-limit prediction of the medium-short-term voltage of a single distribution transformer under the influence of multiple factors is realized by using a random forest technology, the out-of-limit risk type of the medium-short-term voltage of the single distribution transformer is identified, the out-of-limit risk probability is given, and the optimization and treatment capacity of the out-of-limit problem of the medium-short-term power distribution network voltage is. The method has the advantages of clear logic, ordered steps and capability of well and effectively judging whether the voltage in the middle-short period exceeds the limit or not.
Drawings
Fig. 1 is a schematic diagram of a prediction process in the present case.
Detailed Description
The method is operated according to the following steps:
1) analyzing and researching the influence factors of the voltage out-of-limit: through the guidance and research results of earlier scientific research documents, the method for drawing out relevant factors of voltage out-of-limit mainly comprises the following steps: the system comprises a distribution transformer capacity, a distribution transformer power, a distribution transformer voltage, a bus voltage, a feeder power, a feeder capacity, a power factor, a power distance, a bus power supply radius and a climate temperature;
2) analysis of major causes of out-of-limit: the method comprises the following steps of calling relevant data of a transformer area with overvoltage threshold crossing, and establishing a logistic regression model of the voltage threshold crossing phenomenon and each relevant factor by utilizing the relevant data:
wherein p is the current month out-of-limit probability, x is the relevant factor of the last month voltage out-of-limit, namely whether the distribution transformer is out-of-limit in a certain period of time is taken as an output variable, the fluctuation characteristics of all the influencing factors of the transformer in the previous period of time are taken as input variables and are sequentially input into a logistic regression model, the logistic regression model is adopted for training and learning, the correlation degree is counted, and the main cause of out-of-limit of each transformer area is analyzed;
the average value of the bus voltage, the average value of the distribution and transformation voltage, the maximum difference value of the distribution and transformation power and the maximum difference value of the feeder line power have the maximum relevance with the out-of-limit, and are the main cause of the out-of-limit;
3) extracting, preprocessing and fusing data of main causes related to the voltage risk of the distribution network to a distribution network voltage risk prejudgment analysis characteristic table; the quality of data acquired by the distribution transformer out-of-limit factor correlation has obvious influence on a future prediction model, and the quality of original data needs to be preprocessed and cleaned; setting data points with the distribution transformation voltage larger than 0 and smaller than 110 or larger than 330 as invalid data, and deleting the whole data to obtain invalid distribution transformation voltage data; and identifying and marking the out-of-range condition of the transformer in the collected data according to a low-voltage overvoltage standard, wherein the low-voltage mark is-1, the normal mark is 0, and the high-voltage mark is 1. Calculating power distribution network voltage risk pre-judgment and auxiliary analysis characteristic data based on data after data preprocessing and cleaning; importing the structured data into a database for storage according to the fluctuation characteristics of distribution and transformation voltage and power, bus voltage, feeder power and the like of each month and relevant fixed attribute relevant factors;
the prediction flow is shown in FIG. 1;
4) predicting the first future month:
4.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the last month and data of the current month out-of-limit condition into a random forest model, and establishing a plurality of decision trees I;
4.2), predicting: inputting data of main causes of the voltage threshold crossing in the current month into all decision trees I for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage threshold crossing in the next month or not; during random forest model training, a sample and variable double random extraction mode is adopted, a plurality of different short-term decision trees are constructed, whether distribution transformer voltages cross the boundary or not is judged in advance from different dimensions, finally, voltage threshold crossing judgment is realized according to a minority obedience majority principle, and threshold crossing probability is obtained;
5) predicting the second future month:
5.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the second month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees II;
5.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees II for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the second month in the future;
6) predicting the third future month:
6.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the third month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees III;
6.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees III for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage is out-of-limit in the third month in the future;
7) predicting the fourth future month:
7.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fourth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees IV;
7.2), prediction: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees IV for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage out-of-limit in the fourth month exists in the future;
8) predicting the fifth future month:
8.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fifth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
8.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees V for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage in the fifth month is out-of-limit or not in the future;
9) predicting the future sixth month:
9.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the sixth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
9.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees VI for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the future sixth month or not;
10) and result integration: counting the prediction result of whether the voltage exceeds the limit for one to six months in the future; and (6) finishing.
Training data needs to be preprocessed before model training, distribution transformer out-of-limit conditions are obviously unbalanced in distribution, data distribution needs to be corrected first, the asymmetry of the distribution of samples in a production environment is considered, the ratio of normal voltage to out-of-limit voltage in the training samples is set to be 10:1, and data are balanced by adopting a normal transformer under-sample method. Dimension difference exists in each dimension data of the characteristic data table, and the data are reduced to 0-1. Training and testing set: random drawing and monthly prediction are adopted.
And (3) realizing medium and short term voltage out-of-limit early warning, wherein the problem of machine learning time span needs to be considered, and the model pre-judging time span is consistent with the model training time span. The time span selection is reasonable in 1 month by combining with the actual production environment, namely the current month out-of-range condition and the previous 1 month (or 2-6 months) are used for training, the current month data is used for realizing the early warning of the out-of-range condition of the next 1 month (or 2-6 months), and the prediction results of each month are integrated, so that the early warning of the out-of-range voltage of the medium-short term is realized. In actual use, the closer the month in which the input main cause is located and the month in which the prediction result is located, the higher the prediction accuracy, and therefore, each month needs to predict the next half year.
And during random forest model training, a sample and variable double random extraction mode is adopted, a plurality of different decision trees are constructed, the different decision trees look like experts in different fields, whether the distribution transformer voltage is out of range or not is pre-judged from different dimensions, finally, voltage out-of-range pre-judgment is realized according to a minority obedience majority principle, and the out-of-range probability is obtained. The random forest model needs to be trained for 6 times, the pre-judgment result and the pre-judgment probability of the step length of 1-6 months are respectively obtained, and the boundary crossing condition and the probability of the future 6 months are comprehensively obtained according to the pre-judgment result and the pre-judgment probability.
And 2) establishing a logistic regression model for multiple times according to the relevant factors of voltage out-of-limit in the step 2), wherein x is used as one of distribution transformer capacity, distribution transformer power, distribution transformer voltage, bus voltage, feeder power, feeder capacity, power factors, power source distance, bus power supply radius and climate temperature every time.
Claims (1)
1. The power distribution network medium and short term voltage out-of-limit early warning method based on big data is characterized by comprising the following steps of:
1) analyzing and researching the influence factors of the voltage out-of-limit: through the guidance and research results of earlier scientific research documents, the method for drawing out relevant factors of voltage out-of-limit mainly comprises the following steps: the system comprises a distribution transformer capacity, a distribution transformer power, a distribution transformer voltage, a bus voltage, a feeder power, a feeder capacity, a power factor, a power distance, a bus power supply radius and a climate temperature;
2) analysis of major causes of out-of-limit: the method comprises the following steps of calling relevant data of a transformer area with overvoltage threshold crossing, and establishing a logistic regression model of the voltage threshold crossing phenomenon and each relevant factor by utilizing the relevant data:
wherein p is the current month out-of-limit probability, x is the relevant factor of the last month voltage out-of-limit, namely whether the distribution transformer is out-of-limit in a certain period of time is taken as an output variable, the fluctuation characteristics of all the influencing factors of the transformer in the previous period of time are taken as input variables and are sequentially input into a logistic regression model, the logistic regression model is adopted for training and learning, the correlation degree is counted, and the main cause of out-of-limit of each transformer area is analyzed;
the average value of the bus voltage, the average value of the distribution and transformation voltage, the maximum difference value of the distribution and transformation power and the maximum difference value of the feeder line power have the maximum relevance with the out-of-limit, and are the main cause of the out-of-limit;
establishing a logistic regression model for multiple times in the step 2) according to the relevant factors of voltage out-of-limit, wherein x is used as one of distribution transformer capacity, distribution transformer power, distribution transformer voltage, bus voltage, feeder power, feeder capacity, power factors, power source distance, bus power supply radius and climate temperature every time;
3) extracting, preprocessing and fusing data of main causes related to the voltage risk of the distribution network to a distribution network voltage risk prejudgment analysis characteristic table;
4) predicting the first future month:
4.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the last month and data of the current month out-of-limit condition into a random forest model, and establishing a plurality of decision trees I;
4.2), predicting: inputting data of main causes of the voltage threshold crossing in the current month into all decision trees I for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage threshold crossing in the next month or not;
5) predicting the second future month:
5.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the second month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees II;
5.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees II for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the second month in the future;
6) predicting the third future month:
6.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the third month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees III;
6.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees III for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage is out-of-limit in the third month in the future;
7) predicting the fourth future month:
7.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fourth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees IV;
7.2), prediction: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees IV for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage out-of-limit in the fourth month exists in the future;
8) predicting the fifth future month:
8.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the fifth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
8.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees V for prediction to obtain a plurality of results, and taking the result with the most occurrence in the results as a prediction result of whether the voltage in the fifth month is out-of-limit or not in the future;
9) predicting the future sixth month:
9.1), training a random forest model: inputting data of main causes of voltage out-of-limit in the sixth month before the current month and data of the voltage out-of-limit condition in the current month into a random forest model, and establishing a plurality of decision trees;
9.2), predicting: inputting data of main causes of the voltage out-of-limit in the current month into all decision trees VI for prediction to obtain a plurality of results, and taking the result with the most occurrence in the plurality of results as a prediction result of whether the voltage is out-of-limit in the future sixth month or not;
10) and result integration: counting the prediction result of whether the voltage exceeds the limit for one to six months in the future; and (6) finishing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811500365.5A CN109636029B (en) | 2018-12-10 | 2018-12-10 | Power distribution network medium-short term voltage out-of-limit early warning method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811500365.5A CN109636029B (en) | 2018-12-10 | 2018-12-10 | Power distribution network medium-short term voltage out-of-limit early warning method based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109636029A CN109636029A (en) | 2019-04-16 |
| CN109636029B true CN109636029B (en) | 2020-07-24 |
Family
ID=66072300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811500365.5A Active CN109636029B (en) | 2018-12-10 | 2018-12-10 | Power distribution network medium-short term voltage out-of-limit early warning method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109636029B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110266000B (en) * | 2019-06-17 | 2022-07-19 | 国网江苏省电力有限公司 | Power distribution network voltage out-of-limit reason analysis method, system and storage medium |
| CN110739698B (en) * | 2019-11-20 | 2023-07-28 | 国网山东省电力公司潍坊市寒亭区供电公司 | Method and system for allocating 10kV bus voltage and public distribution transformer outlet voltage |
| CN117614132B (en) * | 2023-11-27 | 2024-06-11 | 广州航海学院 | Distribution transformer voltage out-of-limit portrait method and device for power distribution network |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107666148B (en) * | 2017-08-10 | 2021-03-05 | 国网浙江宁波市鄞州区供电公司 | Line fault studying and judging method based on distribution transformer power failure signal |
| CN108022001B (en) * | 2017-09-20 | 2021-10-19 | 河海大学 | Short-term load probability density prediction method based on PCA (principal component analysis) and quantile regression forest |
| CN108123436B (en) * | 2017-12-01 | 2021-07-20 | 国网浙江省电力公司绍兴供电公司 | Voltage out-of-limit prediction model based on principal component analysis and multiple regression algorithm |
-
2018
- 2018-12-10 CN CN201811500365.5A patent/CN109636029B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109636029A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109636029B (en) | Power distribution network medium-short term voltage out-of-limit early warning method based on big data | |
| CN107169628B (en) | Power distribution network reliability assessment method based on big data mutual information attribute reduction | |
| CN106909933B (en) | A kind of stealing classification Forecasting Methodology of three stages various visual angles Fusion Features | |
| CN107967542B (en) | Long-short term memory network-based electricity sales amount prediction method | |
| CN109902954B (en) | Flexible job shop dynamic scheduling method based on industrial big data | |
| CN110689195A (en) | Power daily load prediction method | |
| CN111090643B (en) | Mass electricity consumption data mining method based on data analysis system | |
| CN113205203A (en) | CNN-LSTM-based building energy consumption prediction method and system | |
| CN110610121B (en) | Small-scale source load power abnormal data identification and restoration method based on curve clustering | |
| CN114048870A (en) | Power system abnormity monitoring method based on log characteristic intelligent mining | |
| CN105469219A (en) | Method for processing power load data based on decision tree | |
| CN110690701A (en) | Analysis method for influence factors of abnormal line loss | |
| CN106054858A (en) | Decision tree classification and fault code classification-based vehicle remote diagnosis and spare part retrieval method | |
| CN111815060A (en) | Short-term load prediction method and device for power utilization area | |
| CN106095785B (en) | Fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification | |
| CN111723085B (en) | Method and system for cleaning electric energy consumption data of public building air conditioner | |
| CN111160404A (en) | Method and device for analyzing reasonable value of line loss marking pole of power distribution network | |
| CN112308298A (en) | Multi-scenario performance index prediction method and system for semiconductor production line | |
| CN112200209A (en) | Poor user identification method based on day-to-day power consumption | |
| CN111339167A (en) | Method for analyzing influence factors of transformer area line loss rate based on K-means and principal component linear regression | |
| CN116826745B (en) | Layered and partitioned short-term load prediction method and system in power system background | |
| CN106487571A (en) | A kind of method and device of assessment network performance index variation tendency | |
| CN115879915B (en) | Cross-platform standardized overhaul method for power plant | |
| CN108123436B (en) | Voltage out-of-limit prediction model based on principal component analysis and multiple regression algorithm | |
| CN104318489A (en) | Transformer grouping method based on load characteristic analysis |
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 |