CN105825298A - Electric network metering early-warning system and method based on load characteristic pre-estimation - Google Patents
Electric network metering early-warning system and method based on load characteristic pre-estimation Download PDFInfo
- Publication number
- CN105825298A CN105825298A CN201610143737.8A CN201610143737A CN105825298A CN 105825298 A CN105825298 A CN 105825298A CN 201610143737 A CN201610143737 A CN 201610143737A CN 105825298 A CN105825298 A CN 105825298A
- Authority
- CN
- China
- Prior art keywords
- load
- clustering
- peak
- time
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000013598 vector Substances 0.000 claims abstract description 33
- 230000002159 abnormal effect Effects 0.000 claims abstract description 11
- 238000005259 measurement Methods 0.000 claims abstract description 6
- 230000005611 electricity Effects 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004043 responsiveness Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses an electric network metering early-warning system and method based on load characteristic pre-estimation. Historical data of power load is collected, a load curve is drafted, a sample set is constructed, and a K-MEDOIDS algorithm is used to determine a typical load curve of a power user; sample data is segmented in a unified manner according to time interval, multiple targets are used as characteristic vectors, and a characteristic index is selected from a typical daily load curve; and a clustering result is combined with a Markov chain to establish a prediction chain, prediction is carried out, a prediction result is compared with actual measurement data detected by a metering device to determine whether invalid power utilization occurs and whether the metering device works normally, and it is determined that invalid power utilization occurs or the metering device is abnormal, invalid power utilization or the metering device is marked, and early warning is carried out.
Description
Technical Field
The invention relates to a power grid metering early warning system and method based on load characteristic estimation.
Background
Along with the development of economy, the demand of electricity is more and more large in recent years. During many peak hours, a problem of tight supply and demand of electricity occurs. Whether the user is an individual user or a large-scale enterprise user, the electricity consumption and the electricity charge need to be measured and calculated, the accurate measurement and calculation are very important, and the method and the device can also provide great help for the safe and efficient operation of a power grid.
Therefore, the method can accurately count the power consumption of the user during metering, and can judge whether the value is in a normal range, and has no electric leakage, electric larceny or electric larceny behaviors, so that the method is very important.
However, the current power utilization monitoring mainly depends on manual work or a metering box (or meter), the problems of unclear monitoring strategy, inadequate monitoring implementation and non-visual execution effect display exist in both manual work and the metering box (or meter), and the lack of a systematic power grid power utilization metering monitoring method results in stronger monitoring work randomness, so that the further analysis of improving the power grid operation efficiency and the power quality, such as the improvement of ordered power utilization and load characteristic analysis, cannot have correct data support. Therefore, the power consumption of the electricity consumer needs to be predicted to determine whether the actually measured power consumption is correct.
At present, the method for predicting the power load, in particular the power consumption load, comprises the following steps: kalman filtering, grey prediction, exponential smoothing, BP neural network, etc. The BP neural network has a low convergence speed, local minimum values are easy to appear, and the prediction accuracy is not high due to the limitations. The grey prediction method requires that the historical power usage be smooth discrete data and is therefore only suitable for short term predictions. Kalman filtering can predict the predicted value at the next moment through a state space model, and has the defect that the filtering efficiency is too low or even fails under the condition that the noise statistical characteristics are completely unknown or partially known. The exponential smoothing method is simple and easy to implement, the fluctuating data of the historical power load is eliminated by adopting a weighted moving average mode, the development trend of the power load can be found, and the method is suitable for preprocessing data and has the defect that a hysteresis effect is generated to seriously influence the prediction precision.
Meanwhile, when the optimization characteristic index of the load is established in the data preprocessing process, the existing prediction method is too single at present, the multi-target load characteristic analysis cannot be realized, the comprehensive consideration on uncertainty and multi-target attributes is lacked, the influence of uncertain factors cannot be eliminated by the clustered result under the multi-target framework, and the actual operation requirement cannot be met.
Disclosure of Invention
The invention aims to solve the problems and provides a power grid metering early warning system and method based on load characteristic estimation.
In order to achieve the purpose, the invention adopts the following technical scheme:
a power grid metering early warning method based on load characteristic estimation comprises the following steps:
(1) collecting historical data of the power load, drawing a load curve, constructing a sample set, and determining a typical load curve of a power consumer by using a K-MEDOIDS algorithm;
(2) uniformly segmenting sample data according to time intervals, constructing a clustering feature vector by taking the fluctuation intensity of each minimum time length in a time interval sequence, the digital features of the time interval sequence, security loads, a limitable load, a peak hour difference rate, a peak valley difference rate and a monthly unbalance coefficient as feature indexes, and selecting the feature indexes from a typical daily load curve;
(3) giving sample data bias parameters, clustering by taking the characteristic vector as an index based on an AP clustering algorithm, determining the clustering number and the clustering center, and outputting a clustering result if the clustering center of the sample point meets a judgment condition that the characteristic vector of a day period is taken as the clustering index;
(4) and constructing a prediction model according to the clustering result and a Markov chain, predicting, comparing the prediction result with the measured data detected by the metering device, confirming whether the illegal power utilization condition exists, judging whether the metering device normally works, and marking for early warning if the illegal power utilization condition exists or the metering device is abnormal.
In the step (1), historical load data of the power users within a period of time is collected, then a plurality of daily load curves are drawn, all the daily load curves of the users are respectively clustered by using a K-MEDOIDS algorithm, the electricity utilization condition of each user is divided into a plurality of classes, and the central load curve corresponding to the class with the largest classified days is selected as a typical daily load curve of the user.
In the step (2), the load sample data is uniformly segmented according to clustering time intervals, each clustering time interval is composed of a plurality of set minimum time lengths, and a change fluctuation sequence of the sample in each minimum time length in a clustering time interval time period is used for representing a data fluctuation trend and mean values of maximum active power, minimum active power and active power in a clustering time interval of actual active power are used for sequentially forming a characteristic vector of fluctuation intensity of each minimum time length in a time period sequence.
In the step (2), the amplitude of the fluctuation intensity sequence fully reflects the amplitude of the power fluctuation of the generalized load node, and the positive and negative of the amplitude reflects the flow direction of the power of the generalized load node.
In the step (2), the fluctuation rate of the load refers to the fluctuation size of the user power load, and reflects the fluctuation amplitude of the load, and the size is the ratio of the standard deviation of the load active power to the arithmetic mean value of the load active power.
In the step (2), the security load is calculated by selecting actual load data within a period of time and calculating an arithmetic average of a plurality of load points. The calculation formula is as follows:
wherein, PbFor security load, PminiN is a constant at the ith minimum load point, and preferably, N is 300 or more.
In the step (2), the difference between the peak typical load and the security load of the user with the limited load as the peak value is multiplied by the coincidence rate, and the peak typical load is the average value of the user load at the corresponding moment when the maximum power supply load occurs in the power grid power supply peak period.
In the step (2), the peak time difference rate refers to the time difference degree between the power grid load peak and the user load peak, and the size of the peak time difference rate is the product of the time difference sum of the power grid load peak and the user load peak divided by the set time length and M, wherein M is the smaller value of the number of the power grid load peaks and the number of the user load peaks.
In the step (2), the peak-to-valley difference refers to a difference between a maximum load and a minimum load in a certain time period of the power system, and usually takes a day as a unit, the peak-to-valley difference can reflect fluctuation of a user load, and the peak-to-valley difference rate is a ratio of a difference between a daily maximum load and a daily minimum load to a daily maximum load.
In the step (4), the specific steps include:
(4-1) setting a prediction time interval, determining the data concentration probability of each daily load value in the sample set according to the clustering result, carrying out normalization operation, determining each discrete type preliminary state probability vector, and constructing an initial Markov model;
(4-2) performing probability transfer by using a Chepman-Kolmogorov equation, converting the sample data into N layers of daily load sample data and 1 layer of daily load sample data, and integrating the data to form a final Markov model;
(4-3) dividing time periods, mapping all daily load values into the time periods according to the timestamps of the daily load values, calculating the probability that each object falls into each time period, and predicting the daily load value after any time period.
In the step (4), the prediction result is compared with the measured data, and the specific steps include:
(4-a) adding the prediction results of a plurality of time periods, determining the size of the actually measured data in the time period, performing difference processing on the two, and subtracting the data with small value from the data with large value;
and (4-b) dividing the data with small numerical value by the difference, and if the ratio exceeds the preset ratio threshold upper limit or lower limit, determining that illegal electricity utilization exists or the metering device is in an abnormal working state, and marking the user.
A power grid metering early warning system based on load characteristic estimation comprises:
the data preprocessing module is used for acquiring historical data of the power load, drawing a load curve and constructing a sample set;
the clustering determination module is used for determining a typical load curve of the electricity user by utilizing a K-MEDOIDS algorithm;
the time module uniformly segments the sample data according to time intervals;
the computing module is used for computing the fluctuation intensity of each minimum time length in the time period sequence recorded by the time module, the digital characteristics of the time period sequence, the security load, the limitable load, the peak time difference rate, the peak valley difference rate and the monthly unbalance coefficient;
the characteristic vector construction module is used for constructing a clustering characteristic vector by taking all calculation results in the calculation module as characteristic indexes, and selecting the characteristic indexes from a typical daily load curve;
the characteristic vector clustering module gives sample data bias parameters, carries out clustering by taking the characteristic vector as an index based on an AP clustering algorithm, and determines the clustering number and the clustering center;
the judging module is used for judging whether the clustering center of the sample point meets the judging condition which takes the day-time period characteristic vector as the clustering index or not, and if so, outputting a clustering result;
the prediction module is used for constructing a prediction model according to the clustering result and in combination with a Markov chain algorithm to predict;
and the comparison module compares the prediction result in the prediction module with the actual measurement data detected by the metering device of the electricity consumer, confirms whether the illegal electricity utilization condition exists or not and judges whether the metering device works normally or not.
The invention has the beneficial effects that:
(1) the method can correctly judge whether the electricity utilization value of various electricity utilization users is in a normal range or not, and whether the behaviors of electricity leakage, electricity stealing or electricity stealing exist or not, and is beneficial to improving the orderliness of the power grid market;
(2) a clustering method of multi-target feature vectors is provided, and the day-time characteristic rule of sample data can be reflected;
(3) the invention can realize the load prediction of a certain industry and a certain client, and provides data support for the reason analysis of the load characteristic change;
(4) the Markov chain is used for predicting the daily load value, so that the accuracy of the non-stable load data is effectively improved, and the prediction accuracy is greatly improved by taking time factors into account.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a graph illustrating a typical daily load curve for a region of the grid 2015 year 7 month according to the present invention;
FIG. 3 is a comparison of predicted results of the present invention.
The specific implementation mode is as follows:
the invention is further described with reference to the following figures and examples.
As shown in fig. 1, a power grid metering early warning method based on load characteristic estimation includes the following steps:
(1) collecting historical data of the power load, drawing a load curve, constructing a sample set, and determining a typical load curve of a power consumer by using a K-MEDOIDS algorithm;
(2) uniformly segmenting sample data according to time intervals, constructing a clustering feature vector by taking the fluctuation intensity of each minimum time length in a time interval sequence, the digital features of the time interval sequence, security loads, a limitable load, a peak hour difference rate, a peak valley difference rate and a monthly unbalance coefficient as feature indexes, and selecting the feature indexes from a typical daily load curve;
(3) giving sample data bias parameters, clustering by taking the characteristic vector as an index based on an AP clustering algorithm, determining the clustering number and the clustering center, and outputting a clustering result if the clustering center of the sample point meets a judgment condition that the characteristic vector of a day period is taken as the clustering index;
(4) and constructing a prediction model according to the clustering result and a Markov chain, predicting, comparing the prediction result with the measured data detected by the metering device, confirming whether the illegal power utilization condition exists, judging whether the metering device normally works, and marking for early warning if the illegal power utilization condition exists or the metering device is abnormal.
In the step (1), historical load data of the power users within a period of time is collected, then a plurality of daily load curves are drawn, all the daily load curves of the users are respectively clustered by using a K-MEDOIDS algorithm, the electricity utilization conditions of the users are divided into a plurality of classes, and the central load curve corresponding to the class with the largest classified days is selected as a typical daily load curve of the users.
Of course, the K-MEDOIDS algorithm may be replaced with a K-Means clustering algorithm or a CLARANS algorithm, as is well known to those skilled in the art.
The load characteristic of the power grid is mainly reflected on a load curve, and the change rule of the power grid load can be searched by analyzing the load curve of the power grid. Through the annual load characteristics, the power supply condition of each month of the power grid can be seen, and the month with the maximum power supply pressure is found out; through the monthly load characteristics, the monthly load change rule of the power grid can be seen, and whether the power grid load is influenced by saturday and sunday is mastered; through the daily load characteristics, the peak, flat and valley time periods of the power grid in one day can be obtained, and the time period of the power grid in one day is grasped as the power utilization peak.
As shown in fig. 2, a daily load curve of the power grid describes a change of a power supply load of the power grid for 24 hours a day with time, and a plurality of daily load curves can be obtained according to the data measurement system, so that a typical daily load curve of the power grid needs to be found when performing daily load characteristic analysis. When analyzing the typical daily load curve of the power grid, the key is to obtain the peak, flat and valley time periods of the power grid in one day, that is, the peak and valley time periods of the typical daily load curve of the power grid need to be divided.
The typical morning, mid-evening and evening peak hours are 8: 00-12:00, 12:00-17:00, 17:00-22:00.
In the step (2), load sample data is clustered according to a clustering time interval TJAnd uniformly segmenting, wherein a clustering time interval consists of a plurality of set minimum time lengths T, and a characteristic vector of the fluctuation intensity of each minimum time length in a time interval sequence is sequentially formed by utilizing the change fluctuation sequence of the sample in each minimum time length in the time interval of the clustering time interval to represent the data fluctuation trend and the average value of the maximum active power, the minimum active power and the active power in the clustering time interval of the actual active power.
Namely, it is
In the formula, p is a clustering time interval serial number; b is TJThe sequence number containing the minimum time length T; wmaxpAnd WminpRespectively the maximum active power and the minimum active power in the p-th clustering time interval;is the mean value of the active power in the p-th clustering time intervalp1，φp2，…，φpbAre respectively TJThe inner b minimum time length fluctuation intensity sequences.
In the step (2), the amplitude of the fluctuation intensity sequence fully reflects the amplitude of the power fluctuation of the generalized load nodes, and the positive and negative of the amplitude reflects the flow direction of the power of the generalized load nodes.
In the step (2), the fluctuation rate of the load refers to the fluctuation size of the user power load, reflects the fluctuation amplitude of the load, and is the ratio of the standard deviation of the load active power to the arithmetic mean value of the load active power.
The calculation formula of the load fluctuation rate is as follows:
wherein x isi(i ═ 1,2, …, N) represents a set of load real power values; n is the number of load points measured in one day.
In the step (2), the security load is calculated by selecting actual load data within a period of time and calculating the arithmetic mean of a plurality of load points. The calculation formula is as follows:
wherein, PbFor security load, PminiN is a constant at the ith minimum load point, and preferably, N is 300 or more.
In the step (2), the difference value between the peak typical load and the security load of the user with the limited load as the peak value is multiplied by the coincidence rate, and the peak typical load is the average value of the user load at the corresponding moment when the maximum power supply load occurs in the power grid power supply peak period.
In the step (2), the peak time difference rate refers to the time difference degree between the power grid load peak and the user load peak, and the size of the peak time difference rate is the product of the time difference sum of the power grid load peak and the user load peak divided by the set time length and M, wherein M is the smaller value of the number of the power grid load peaks and the number of the user load peaks.
In the step (2), the peak-to-valley difference refers to a difference between a maximum load and a minimum load in a certain time period of the power system, and usually takes a day as a unit, the peak-to-valley difference can reflect fluctuation conditions of user loads, and the peak-to-valley difference rate is a ratio of a difference between a daily maximum load and a daily minimum load to a daily maximum load.
The peak-to-valley difference refers to the difference between the maximum load and the minimum load of the power system over a certain period of time, usually in units of days. The size of the peak-to-valley difference directly reflects the peak regulation capacity required by the power grid.
Since the peak-to-valley difference reflects the absolute value of the daily load fluctuation, it does not reflect the relative magnitude of the fluctuation very well. But the peak-to-valley difference rate can better reflect the relative condition of the daily fluctuation.
In the step (3), the AP clustering algorithm does not need to specify the clustering number in advance, but instead, all data points are used as potential clustering centers, which are called exemplar. Taking the value S (k, k) on the diagonal of the S matrix as the criterion for judging whether k points can become the cluster center, this means that the larger the value is, the more likely the point becomes the cluster center is, and this value is also called the reference p (reference). The number of clusters is influenced by the reference p, which should take the same value if each data point is considered to be likely to be the cluster center. If the mean value of the similarity of the inputs is taken as the value of p, the number of clusters obtained is medium. If the minimum value is taken, the cluster with less class number is obtained.
Two types of messages are passed in the AP clustering algorithm, (responsiveness) and (availability). r (i, k) represents a numeric message sent from point i to candidate cluster center k, reflecting whether point k is suitable as the cluster center for point i. a (i, k) is a numerical message sent from candidate cluster center k to i, reflecting whether point i selects k as its cluster center. The stronger r (i, k) and a (i, k), the greater the likelihood that k points are the clustering center and the greater the likelihood that i points belong to a cluster having k points as the clustering center. The AP algorithm continuously updates the attractiveness and attribution values of each point through an iterative process until m high quality exemplars are generated, while assigning the remaining data points to the corresponding clusters.
In the step (4), the concrete steps include:
(4-1) setting a prediction time interval, determining the data concentration probability of each daily load value in the sample set according to the clustering result, carrying out normalization operation, determining each discrete type preliminary state probability vector, and constructing an initial Markov model;
(4-2) performing probability transfer by using a Chepman-Kolmogorov equation, converting the sample data into N layers of daily load sample data and 1 layer of daily load sample data, and integrating the data to form a final Markov model;
(4-3) dividing time periods, mapping all daily load values into the time periods according to the timestamps of the daily load values, calculating the probability that each object falls into each time period, and predicting the daily load value after any time period.
Setting the system to have N mutually incompatible states, the initial state probability vector of the systemWherein,initial probabilities in states 1,2, …, n, respectively, from the chepman-kolmogorov equation, the markov chain prediction model at the k +1 state can be found as: s(k＋1)＝S(k+1)*pij (k+1)Wherein p isij (k+1)The transition probabilities are k steps.
In the step (4), the prediction result is compared with the actually measured data, and the specific steps comprise:
(4-a) adding the prediction results of a plurality of time periods, determining the size of the measured data in the time period, and performing difference processing on the two;
and (4-b) dividing the difference value by the predicted value, and if the ratio exceeds the preset ratio threshold upper limit or lower limit, determining that illegal electricity utilization exists or the metering device is in an abnormal working state, and marking the user.
For example, the ratio threshold may be set to [ -50%, + 80% ]. If the predicted value is 100kwh and the measured value is 150kwh, the predicted value is subtracted from the measured value to be 50kwh, and 50/100 is equal to 50%, so that the potential power stealing problem or the damage of the metering device does not exist within the ratio threshold.
If the predicted value is 100kwh and the measured value is 50kwh, the predicted value minus the measured value is 50kwh, and 50/50 is 100%, which exceeds the threshold range of the ratio, resulting in potential power theft or damage to the metering device. And labeling the user. The power worker is reminded to verify and review.
As shown in FIG. 3, the prediction result of the present invention shows that the accuracy is high.
In addition, a failure or a leak of the conventional metering device itself is one of the main causes of the occurrence of an abnormal condition. For an electric power metering device, an electric energy meter, a metering loop, a mutual inductor and the like are important components, and when any one component fails, the component is abnormal in use, so that the component generates a large error for measuring power consumption data, and the error mainly comprises three aspects of an electric energy meter error, a metering loop voltage drop error and a mutual inductance device error; meanwhile, the unreasonable production process and the unqualified quality of the electric power metering device can cause the braking magnetic flux of the metering device to change abnormally, so that the normal operation cannot be realized.
The method can be used for detecting the voltage by adopting an electric power metering voltage detection method, and monitoring the installation position of the circuit breaker, the phase voltage variable, the three-phase unbalanced voltage variable and the like. When an abnormal phenomenon occurs in the power metering device, the breaker is usually accompanied by the situations that relevant maintenance information does not appear, the three-phase unbalanced voltage variation is not in an actually allowable range, the phase voltage is less than 70% of the rated voltage, and the normal state is not recovered for a long time. Of course, the electric power metering current monitoring method may be used, and when the electric power metering device is in an abnormal condition, if the breaker does not open or the phase current variable does not return to a normal state within a certain period of time, the worker needs to perform necessary maintenance work on the electric power metering device. Generally, the method for monitoring the electric power metering current mainly includes monitoring the installation position of an interrupter, the phase current, the three-phase unbalanced current and the like.
A power grid metering early warning system based on load characteristic estimation comprises:
the data preprocessing module is used for acquiring historical data of the power load, drawing a load curve and constructing a sample set;
the clustering determination module is used for determining a typical load curve of the electricity user by utilizing a K-MEDOIDS algorithm;
the time module uniformly segments the sample data according to time intervals;
the computing module is used for computing the fluctuation intensity of each minimum time length in the time period sequence recorded by the time module, the digital characteristics of the time period sequence, the security load, the limitable load, the peak time difference rate, the peak valley difference rate and the monthly unbalance coefficient;
the characteristic vector construction module is used for constructing a clustering characteristic vector by taking all calculation results in the calculation module as characteristic indexes, and selecting the characteristic indexes from a typical daily load curve;
the characteristic vector clustering module gives sample data bias parameters, carries out clustering by taking the characteristic vector as an index based on an AP clustering algorithm, and determines the clustering number and the clustering center;
the judging module is used for judging whether the clustering center of the sample point meets the judging condition which takes the day-time period characteristic vector as the clustering index or not, and if so, outputting a clustering result;
the prediction module is used for constructing a prediction model according to the clustering result and in combination with a Markov chain algorithm to predict;
and the comparison module compares the prediction result in the prediction module with the actual measurement data detected by the metering device of the electricity consumer, confirms whether the illegal electricity utilization condition exists or not and judges whether the metering device works normally or not.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A power grid metering early warning method based on load characteristic estimation is characterized by comprising the following steps: the method comprises the following steps:
(1) collecting historical data of the power load, drawing a load curve, constructing a sample set, and determining a typical load curve of a power consumer by using a K-MEDOIDS algorithm;
(2) uniformly segmenting sample data according to time intervals, constructing a clustering feature vector by taking the fluctuation intensity of each minimum time length in a time interval sequence, the digital features of the time interval sequence, security loads, a limitable load, a peak hour difference rate, a peak valley difference rate and a monthly unbalance coefficient as feature indexes, and selecting the feature indexes from a typical daily load curve;
(3) giving sample data bias parameters, clustering by taking the characteristic vector as an index based on an AP clustering algorithm, determining the clustering number and the clustering center, and outputting a clustering result if the clustering center of the sample point meets a judgment condition that the characteristic vector of a day period is taken as the clustering index;
(4) and constructing a prediction model according to the clustering result and a Markov chain, predicting, comparing the prediction result with the measured data detected by the metering device, confirming whether the illegal power utilization condition exists, judging whether the metering device normally works, and marking for early warning if the illegal power utilization condition exists or the metering device is abnormal.
2. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (1), historical load data of the power users within a period of time is collected, then a plurality of daily load curves are drawn, all the daily load curves of the users are respectively clustered by using a K-MEDOIDS algorithm, the electricity utilization condition of each user is divided into a plurality of classes, and the central load curve corresponding to the class with the largest classified days is selected as a typical daily load curve of the user.
3. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (2), the load sample data is uniformly segmented according to clustering time intervals, each clustering time interval is composed of a plurality of set minimum time lengths, and a change fluctuation sequence of the sample in each minimum time length in a clustering time interval time period is used for representing a data fluctuation trend and mean values of maximum active power, minimum active power and active power in a clustering time interval of actual active power are used for sequentially forming a characteristic vector of fluctuation intensity of each minimum time length in a time period sequence.
4. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (2), the security load is calculated by selecting actual load data within a period of time and calculating an arithmetic average of a plurality of load points, wherein the calculation formula is as follows:
wherein, PbFor security load, Pmin iN is a constant for the ith minimum load point.
5. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (2), the difference between the peak typical load and the security load of the user with the limited load as the peak value is multiplied by the coincidence rate, and the peak typical load is the average value of the user load at the corresponding moment when the maximum power supply load occurs in the power grid power supply peak period.
6. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (2), the peak time difference rate refers to the time difference degree between the power grid load peak and the user load peak, and the size of the peak time difference rate is the product of the time difference sum of the power grid load peak and the user load peak divided by the set time length and M, wherein M is the smaller value of the number of the power grid load peaks and the number of the user load peaks.
7. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (2), the peak-to-valley difference refers to a difference between a maximum load and a minimum load in a certain time period of the power system, and usually takes a day as a unit, the peak-to-valley difference can reflect fluctuation of a user load, and the peak-to-valley difference rate is a ratio of a difference between a daily maximum load and a daily minimum load to a daily maximum load.
8. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (4), the specific steps include:
(4-1) setting a prediction time interval, determining the data concentration probability of each daily load value in the sample set according to the clustering result, carrying out normalization operation, determining each discrete type preliminary state probability vector, and constructing an initial Markov model;
(4-2) performing probability transfer by using a Chepman-Kolmogorov equation, converting the sample data into N layers of daily load sample data and 1 layer of daily load sample data, and integrating the data to form a final Markov model;
(4-3) dividing time periods, mapping all daily load values into the time periods according to the timestamps of the daily load values, calculating the probability that each object falls into each time period, and predicting the daily load value after any time period.
9. The power grid metering early warning method based on load characteristic estimation as claimed in claim 1, wherein: in the step (4), the prediction result is compared with the measured data, and the specific steps include:
(4-a) adding the prediction results of a plurality of time periods, determining the size of the actually measured data in the time period, performing difference processing on the two, and subtracting the data with small value from the data with large value;
and (4-b) dividing the data with small numerical value by the difference, and if the ratio exceeds the preset ratio threshold upper limit or lower limit, determining that illegal electricity utilization exists or the metering device is in an abnormal working state, and marking the user.
10. A power grid metering early warning system based on load characteristic estimation is characterized in that: the method comprises the following steps:
the data preprocessing module is used for acquiring historical data of the power load, drawing a load curve and constructing a sample set;
the clustering determination module is used for determining a typical load curve of the electricity user by utilizing a K-MEDOIDS algorithm;
the time module uniformly segments the sample data according to time intervals;
the computing module is used for computing the fluctuation intensity of each minimum time length in the time period sequence recorded by the time module, the digital characteristics of the time period sequence, the security load, the limitable load, the peak time difference rate, the peak valley difference rate and the monthly unbalance coefficient;
the characteristic vector construction module is used for constructing a clustering characteristic vector by taking all calculation results in the calculation module as characteristic indexes, and selecting the characteristic indexes from a typical daily load curve;
the characteristic vector clustering module gives sample data bias parameters, carries out clustering by taking the characteristic vector as an index based on an AP clustering algorithm, and determines the clustering number and the clustering center;
the judging module is used for judging whether the clustering center of the sample point meets the judging condition which takes the day-time period characteristic vector as the clustering index or not, and if so, outputting a clustering result;
the prediction module is used for constructing a prediction model according to the clustering result and in combination with a Markov chain algorithm to predict;
and the comparison module compares the prediction result in the prediction module with the actual measurement data detected by the metering device of the electricity consumer, confirms whether the illegal electricity utilization condition exists or not and judges whether the metering device works normally or not.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610143737.8A CN105825298B (en) | 2016-03-14 | 2016-03-14 | Power grid metering early warning system and method based on load characteristic estimation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610143737.8A CN105825298B (en) | 2016-03-14 | 2016-03-14 | Power grid metering early warning system and method based on load characteristic estimation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105825298A true CN105825298A (en) | 2016-08-03 |
| CN105825298B CN105825298B (en) | 2020-05-01 |
Family
ID=56988034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610143737.8A Expired - Fee Related CN105825298B (en) | 2016-03-14 | 2016-03-14 | Power grid metering early warning system and method based on load characteristic estimation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105825298B (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106447123A (en) * | 2016-10-14 | 2017-02-22 | 国网天津市电力公司 | Typical daily load characteristics based substation locating method and substation locating device |
| CN106484983A (en) * | 2016-09-29 | 2017-03-08 | 中国电力科学研究院 | A kind of appraisal procedure of power system relay protection device running status and device |
| CN106779186A (en) * | 2016-11-30 | 2017-05-31 | 新奥泛能网络科技股份有限公司 | A kind of energy supply scale determination method and device with energy main body based on different industry situations |
| CN108133281A (en) * | 2017-12-05 | 2018-06-08 | 国网内蒙古东部电力有限公司电力科学研究院 | An Optimization Method for Location-Selection is paid in the electricity charge based on improved nearest neighbor classifier propagation algorithm |
| CN108256738A (en) * | 2017-12-22 | 2018-07-06 | 同济大学 | Track switch action reference curve choosing method and its application |
| CN108280479A (en) * | 2018-01-25 | 2018-07-13 | 重庆大学 | A kind of power grid user sorting technique based on Load characteristics index weighted cluster algorithm |
| CN109030897A (en) * | 2018-09-20 | 2018-12-18 | 国网湖南省电力有限公司 | Stealing detection method based on Fourier transformation |
| CN109238358A (en) * | 2018-09-12 | 2019-01-18 | 国网福建省电力有限公司 | A kind of distribution type electric energy batch meter multi-sensor data collection and anti-interference processing method |
| CN109409764A (en) * | 2018-11-08 | 2019-03-01 | 国网河北省电力有限公司电力科学研究院 | Production monitoring method and terminal device |
| CN110188221A (en) * | 2019-04-08 | 2019-08-30 | 国网浙江省电力有限公司舟山供电公司 | A kind of load curve hierarchy clustering method based on shape distance |
| CN110889019A (en) * | 2019-11-04 | 2020-03-17 | 华北电力大学 | Novel graphical description method for power load |
| CN110910173A (en) * | 2019-11-25 | 2020-03-24 | 深圳市深电能售电有限公司 | Power price analysis method for power spot market node |
| CN110991477A (en) * | 2019-10-29 | 2020-04-10 | 中国电力科学研究院有限公司 | Method and system for identifying users in abnormal industry and abnormal electricity utilization behaviors of power system |
| CN111105098A (en) * | 2019-12-25 | 2020-05-05 | 北京华电天仁电力控制技术有限公司 | Single user algorithm self-matching load prediction method and system |
| CN111291987A (en) * | 2020-01-23 | 2020-06-16 | 广东电网有限责任公司计量中心 | Method and device for predicting service life of mutual inductor |
| CN111539845A (en) * | 2020-04-21 | 2020-08-14 | 国网四川省电力公司电力科学研究院 | Enterprise environment-friendly management and control response studying and judging method based on power consumption mode membership grade |
| CN111600304A (en) * | 2020-06-17 | 2020-08-28 | 广东工业大学 | Building power scheduling method, device and equipment |
| CN111598330A (en) * | 2020-05-13 | 2020-08-28 | 国网河北省电力有限公司电力科学研究院 | Residential electricity load prediction method based on load identification technology |
| CN111817294A (en) * | 2020-07-07 | 2020-10-23 | 郑州工程技术学院 | Power grid power supply control system, method and equipment based on mathematical model |
| CN112381436A (en) * | 2020-11-23 | 2021-02-19 | 上海电气分布式能源科技有限公司 | Time-by-time electrical load generation method, device, electronic equipment and storage medium |
| CN112446416A (en) * | 2020-10-09 | 2021-03-05 | 国网山东省电力公司日照供电公司 | Non-invasive load monitoring and identifying method and system |
| CN113282639A (en) * | 2021-04-27 | 2021-08-20 | 深圳市中燃科技有限公司 | Gas leakage data monitoring method and system, intelligent terminal and storage medium |
| CN113705710A (en) * | 2021-09-02 | 2021-11-26 | 国网安徽省电力有限公司营销服务中心 | Method and device for identifying default electricity consumption of electric vehicle charging facility |
| CN113869601A (en) * | 2021-10-18 | 2021-12-31 | 深圳供电局有限公司 | Power consumer load prediction method, device and equipment |
| CN114511040A (en) * | 2022-03-31 | 2022-05-17 | 北京邮电大学 | Network traffic classification method and device based on AP clustering |
| CN114580907A (en) * | 2022-03-03 | 2022-06-03 | 北京能云盈和科技有限公司 | Intelligent energy efficiency monitoring method and system |
| CN115436699A (en) * | 2022-11-07 | 2022-12-06 | 北京志翔科技股份有限公司 | Method and device for detecting electricity utilization abnormity and electronic equipment |
| CN115577836A (en) * | 2022-09-29 | 2023-01-06 | 深圳市三正电子有限公司 | Method and device for information acquisition based on MCU |
| CN116166940A (en) * | 2023-04-24 | 2023-05-26 | 施维智能计量系统服务(长沙)有限公司 | User power load time characteristic classification and identification method based on thermodynamic diagram |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426674A (en) * | 2011-10-28 | 2012-04-25 | 山东电力集团公司青岛供电公司 | Power system load prediction method based on Markov chain |
| CN102509173A (en) * | 2011-10-28 | 2012-06-20 | 山东电力集团公司青岛供电公司 | Markov chain based method for accurately forecasting power system loads |
| CN102999791A (en) * | 2012-11-23 | 2013-03-27 | 广东电网公司电力科学研究院 | Power load forecasting method based on customer segmentation in power industry |
| CN103325071A (en) * | 2013-06-09 | 2013-09-25 | 浙江大学 | User typical load curve construction method based on density clustering |
| CN103439572A (en) * | 2013-08-15 | 2013-12-11 | 国家电网公司 | Electricity larceny prevention monitoring method based on photovoltaic power generation power prediction |
| CN103577881A (en) * | 2013-11-12 | 2014-02-12 | 国家电网公司 | Client peak alternating and avoiding power limiting effect evaluation method based on orderly power utilization |
| CN103886518A (en) * | 2014-03-21 | 2014-06-25 | 国家电网公司 | Early warning method for voltage sag based on electric energy quality data mining at monitoring point |
| CN104239963A (en) * | 2014-08-14 | 2014-12-24 | 国家电网公司 | Method for finding abnormal electric energy meter based on gray GM (1, 1) model |
| CN104361233A (en) * | 2014-11-11 | 2015-02-18 | 国家电网公司 | Anti-electric larceny management method under condition of access of distributed generation |
| CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
| CN105005708A (en) * | 2015-08-13 | 2015-10-28 | 山东大学 | Generalized load characteristic clustering method based on AP clustering algorithm |
| CN105139080A (en) * | 2015-08-04 | 2015-12-09 | 国家电网公司 | Improved photovoltaic power sequence prediction method based on Markov chain |
| CN105574606A (en) * | 2015-12-10 | 2016-05-11 | 济南大学 | Power consumption peak avoidance method based on load characteristic index system |
-
2016
- 2016-03-14 CN CN201610143737.8A patent/CN105825298B/en not_active Expired - Fee Related
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426674A (en) * | 2011-10-28 | 2012-04-25 | 山东电力集团公司青岛供电公司 | Power system load prediction method based on Markov chain |
| CN102509173A (en) * | 2011-10-28 | 2012-06-20 | 山东电力集团公司青岛供电公司 | Markov chain based method for accurately forecasting power system loads |
| CN102999791A (en) * | 2012-11-23 | 2013-03-27 | 广东电网公司电力科学研究院 | Power load forecasting method based on customer segmentation in power industry |
| CN103325071A (en) * | 2013-06-09 | 2013-09-25 | 浙江大学 | User typical load curve construction method based on density clustering |
| CN103439572A (en) * | 2013-08-15 | 2013-12-11 | 国家电网公司 | Electricity larceny prevention monitoring method based on photovoltaic power generation power prediction |
| CN103577881A (en) * | 2013-11-12 | 2014-02-12 | 国家电网公司 | Client peak alternating and avoiding power limiting effect evaluation method based on orderly power utilization |
| CN103886518A (en) * | 2014-03-21 | 2014-06-25 | 国家电网公司 | Early warning method for voltage sag based on electric energy quality data mining at monitoring point |
| CN104239963A (en) * | 2014-08-14 | 2014-12-24 | 国家电网公司 | Method for finding abnormal electric energy meter based on gray GM (1, 1) model |
| CN104361233A (en) * | 2014-11-11 | 2015-02-18 | 国家电网公司 | Anti-electric larceny management method under condition of access of distributed generation |
| CN104881706A (en) * | 2014-12-31 | 2015-09-02 | 天津弘源慧能科技有限公司 | Electrical power system short-term load forecasting method based on big data technology |
| CN105139080A (en) * | 2015-08-04 | 2015-12-09 | 国家电网公司 | Improved photovoltaic power sequence prediction method based on Markov chain |
| CN105005708A (en) * | 2015-08-13 | 2015-10-28 | 山东大学 | Generalized load characteristic clustering method based on AP clustering algorithm |
| CN105574606A (en) * | 2015-12-10 | 2016-05-11 | 济南大学 | Power consumption peak avoidance method based on load characteristic index system |
Non-Patent Citations (2)
| Title |
|---|
| 李光珍: "电网短期母线负荷数据预处理及预测模型研究", 《中国优秀硕士学位论文全文数据库（电子期刊）工程科技II辑》 * |
| 陈墨等: "基于统计理论的负荷特性分析及其预测", 《东北电力大学学报》 * |
Cited By (40)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484983A (en) * | 2016-09-29 | 2017-03-08 | 中国电力科学研究院 | A kind of appraisal procedure of power system relay protection device running status and device |
| CN106447123A (en) * | 2016-10-14 | 2017-02-22 | 国网天津市电力公司 | Typical daily load characteristics based substation locating method and substation locating device |
| CN106779186A (en) * | 2016-11-30 | 2017-05-31 | 新奥泛能网络科技股份有限公司 | A kind of energy supply scale determination method and device with energy main body based on different industry situations |
| CN106779186B (en) * | 2016-11-30 | 2021-04-06 | 新奥泛能网络科技股份有限公司 | Energy supply scale determination method and device based on energy consumption main bodies in different business states |
| CN108133281A (en) * | 2017-12-05 | 2018-06-08 | 国网内蒙古东部电力有限公司电力科学研究院 | An Optimization Method for Location-Selection is paid in the electricity charge based on improved nearest neighbor classifier propagation algorithm |
| CN108256738B (en) * | 2017-12-22 | 2021-12-14 | 同济大学 | Turnout action reference curve selection method and application thereof |
| CN108256738A (en) * | 2017-12-22 | 2018-07-06 | 同济大学 | Track switch action reference curve choosing method and its application |
| CN108280479A (en) * | 2018-01-25 | 2018-07-13 | 重庆大学 | A kind of power grid user sorting technique based on Load characteristics index weighted cluster algorithm |
| CN108280479B (en) * | 2018-01-25 | 2020-06-05 | 重庆大学 | Power grid user classification method based on load characteristic index weighted clustering algorithm |
| CN109238358A (en) * | 2018-09-12 | 2019-01-18 | 国网福建省电力有限公司 | A kind of distribution type electric energy batch meter multi-sensor data collection and anti-interference processing method |
| CN109030897A (en) * | 2018-09-20 | 2018-12-18 | 国网湖南省电力有限公司 | Stealing detection method based on Fourier transformation |
| CN109030897B (en) * | 2018-09-20 | 2020-12-04 | 国网湖南省电力有限公司 | Fourier transform-based electricity stealing detection method |
| CN109409764A (en) * | 2018-11-08 | 2019-03-01 | 国网河北省电力有限公司电力科学研究院 | Production monitoring method and terminal device |
| CN110188221B (en) * | 2019-04-08 | 2023-07-11 | 国网浙江省电力有限公司舟山供电公司 | Shape distance-based load curve hierarchical clustering method |
| CN110188221A (en) * | 2019-04-08 | 2019-08-30 | 国网浙江省电力有限公司舟山供电公司 | A kind of load curve hierarchy clustering method based on shape distance |
| CN110991477A (en) * | 2019-10-29 | 2020-04-10 | 中国电力科学研究院有限公司 | Method and system for identifying users in abnormal industry and abnormal electricity utilization behaviors of power system |
| CN110889019A (en) * | 2019-11-04 | 2020-03-17 | 华北电力大学 | Novel graphical description method for power load |
| CN110910173A (en) * | 2019-11-25 | 2020-03-24 | 深圳市深电能售电有限公司 | Power price analysis method for power spot market node |
| CN111105098B (en) * | 2019-12-25 | 2023-11-03 | 国能信控互联技术有限公司 | Load prediction method and system for self-matching of single user algorithm |
| CN111105098A (en) * | 2019-12-25 | 2020-05-05 | 北京华电天仁电力控制技术有限公司 | Single user algorithm self-matching load prediction method and system |
| CN111291987A (en) * | 2020-01-23 | 2020-06-16 | 广东电网有限责任公司计量中心 | Method and device for predicting service life of mutual inductor |
| CN111539845A (en) * | 2020-04-21 | 2020-08-14 | 国网四川省电力公司电力科学研究院 | Enterprise environment-friendly management and control response studying and judging method based on power consumption mode membership grade |
| CN111539845B (en) * | 2020-04-21 | 2020-12-29 | 国网四川省电力公司电力科学研究院 | Enterprise environment-friendly management and control response studying and judging method based on power consumption mode membership grade |
| CN111598330A (en) * | 2020-05-13 | 2020-08-28 | 国网河北省电力有限公司电力科学研究院 | Residential electricity load prediction method based on load identification technology |
| CN111600304A (en) * | 2020-06-17 | 2020-08-28 | 广东工业大学 | Building power scheduling method, device and equipment |
| CN111600304B (en) * | 2020-06-17 | 2021-11-26 | 广东工业大学 | Building power scheduling method, device and equipment |
| CN111817294B (en) * | 2020-07-07 | 2022-02-11 | 郑州工程技术学院 | Power grid power supply control system, method and equipment based on mathematical model |
| CN111817294A (en) * | 2020-07-07 | 2020-10-23 | 郑州工程技术学院 | Power grid power supply control system, method and equipment based on mathematical model |
| CN112446416A (en) * | 2020-10-09 | 2021-03-05 | 国网山东省电力公司日照供电公司 | Non-invasive load monitoring and identifying method and system |
| CN112381436A (en) * | 2020-11-23 | 2021-02-19 | 上海电气分布式能源科技有限公司 | Time-by-time electrical load generation method, device, electronic equipment and storage medium |
| CN112381436B (en) * | 2020-11-23 | 2024-06-04 | 上海电气分布式能源科技有限公司 | Time-by-time electric load generation method and device, electronic equipment and storage medium |
| CN113282639A (en) * | 2021-04-27 | 2021-08-20 | 深圳市中燃科技有限公司 | Gas leakage data monitoring method and system, intelligent terminal and storage medium |
| CN113705710A (en) * | 2021-09-02 | 2021-11-26 | 国网安徽省电力有限公司营销服务中心 | Method and device for identifying default electricity consumption of electric vehicle charging facility |
| CN113869601A (en) * | 2021-10-18 | 2021-12-31 | 深圳供电局有限公司 | Power consumer load prediction method, device and equipment |
| CN114580907A (en) * | 2022-03-03 | 2022-06-03 | 北京能云盈和科技有限公司 | Intelligent energy efficiency monitoring method and system |
| CN114511040A (en) * | 2022-03-31 | 2022-05-17 | 北京邮电大学 | Network traffic classification method and device based on AP clustering |
| CN115577836A (en) * | 2022-09-29 | 2023-01-06 | 深圳市三正电子有限公司 | Method and device for information acquisition based on MCU |
| CN115577836B (en) * | 2022-09-29 | 2023-06-30 | 深圳市三正电子有限公司 | Method and device for information acquisition based on MCU |
| CN115436699A (en) * | 2022-11-07 | 2022-12-06 | 北京志翔科技股份有限公司 | Method and device for detecting electricity utilization abnormity and electronic equipment |
| CN116166940A (en) * | 2023-04-24 | 2023-05-26 | 施维智能计量系统服务(长沙)有限公司 | User power load time characteristic classification and identification method based on thermodynamic diagram |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105825298B (en) | 2020-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105825298B (en) | Power grid metering early warning system and method based on load characteristic estimation | |
| Park et al. | Sliding window-based LightGBM model for electric load forecasting using anomaly repair | |
| EP2514068B1 (en) | Method for non-intrusive load monitoring using a hybrid systems state estimation approach | |
| Jota et al. | Building load management using cluster and statistical analyses | |
| Chicco et al. | Emergent electricity customer classification | |
| CN103245881A (en) | Power distribution network fault analyzing method and device based on tidal current distribution characteristics | |
| CN109818349A (en) | A kind of power grid robust state prediction technique based on multi-dimensional state matrix shiding matching | |
| CN105701554A (en) | Electrical equipment energy efficiency chaos analysis method based on massive measurement data | |
| Liu et al. | Remote malfunctional smart meter detection in edge computing environment | |
| CN112149873A (en) | Low-voltage transformer area line loss reasonable interval prediction method based on deep learning | |
| CN111598165A (en) | Density clustering outlier detection method based on extreme learning machine | |
| CN114878934A (en) | Electric energy consumption data abnormity early warning method | |
| Hartmann et al. | Suspicious electric consumption detection based on multi-profiling using live machine learning | |
| Velasco et al. | Day-ahead base, intermediate, and peak load forecasting using k-means and artificial neural networks | |
| CN112101673A (en) | Power grid development trend prediction method and system based on hidden Markov model | |
| CN118100446B (en) | Intelligent detection method for node running state of power system | |
| Olivares-Rojas et al. | Machine learning model for the detection of electric energy fraud using an edge-fog computing architecture | |
| Wang et al. | Abnormal Electricity Detection of Users Based on Improved Canopy-Kmeans and Isolation Forest Algorithms | |
| CN110288383B (en) | Group behavior power distribution network electricity utilization abnormality detection method based on user attribute tags | |
| CN118152836A (en) | Stability evaluation method for operation process of electric energy meter | |
| CN115146715B (en) | Method, device, equipment and storage medium for diagnosing potential safety hazard of electricity | |
| Liu et al. | An abnormal detection of positive active total power based on local outlier factor | |
| CN116842464A (en) | Battery system SOC estimation method | |
| CN115908082A (en) | Enterprise pollution discharge monitoring method and device based on electricity utilization characteristic indexes | |
| Li et al. | VRLA battery fault prediction for data center based on random forest model and feature enhancement method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200501 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |