CN116090821A - Power distribution network line security risk assessment method considering extreme weather - Google Patents
Power distribution network line security risk assessment method considering extreme weather Download PDFInfo
- Publication number
- CN116090821A CN116090821A CN202310049388.3A CN202310049388A CN116090821A CN 116090821 A CN116090821 A CN 116090821A CN 202310049388 A CN202310049388 A CN 202310049388A CN 116090821 A CN116090821 A CN 116090821A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- power distribution
- fault
- power failure
- power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 101
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012502 risk assessment Methods 0.000 title claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 33
- 230000008859 change Effects 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 12
- 238000001556 precipitation Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000011084 recovery Methods 0.000 claims description 3
- 230000008447 perception Effects 0.000 abstract description 3
- 241000287127 Passeridae Species 0.000 description 26
- 230000000875 corresponding effect Effects 0.000 description 9
- 230000002431 foraging effect Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000010845 search algorithm Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 4
- 244000062645 predators Species 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000006403 short-term memory Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 241000271566 Aves Species 0.000 description 1
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012546 transfer 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
- G06Q10/06—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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a power distribution network line safety risk assessment method considering extreme weather, which is applied to a power distribution network safety early warning system, and is characterized in that firstly, a power distribution network line fault prediction model based on SSA-LSTM is established, the change of external meteorological conditions is reflected through the input of meteorological parameters, the frequency of a regional power distribution network fault line is output, then, a line fault outage result evaluation index is established, the four indexes of outage capacity, outage amount, outage user hours and outage user importance level reflect the difference of each line fault, finally, the safety early warning level of the power distribution network is refined by combining with a risk theory, the accurate perception and timely warning of various meteorological potential hazards can be realized, and the safety of an electric line is converted from 'post-disaster warning' to 'pre-disaster early warning'.
Description
Technical Field
The invention belongs to the technical field of power distribution network line safety, and particularly relates to a power distribution network line safety risk assessment method considering extreme weather.
Background
Under extreme weather conditions, faults of the power distribution network are easier to occur, so that the extreme weather factors become important factors affecting the faults of the power distribution network increasingly, but the faults of the power distribution network caused by the weather factors are not easy to monitor effectively due to the characteristics of irregularity, burst, diversity and the like of weather changes.
And the early hidden danger of the electric safety of the distribution network has the characteristics of weak signals, slow change, large data volume, small or far smaller value than alarm threshold value and the like, and various real-time data of the early hidden danger contain little hidden danger information, and most of the early hidden danger data are repeatable garbage data, so that the communication bandwidth cost and the server storage cost are greatly increased.
In order to solve the above problems, it is necessary to develop a power distribution network line security risk assessment method considering extreme weather.
Disclosure of Invention
Aiming at the problem of frequent faults of the power distribution network under the extreme weather condition, the invention provides a power distribution network line safety risk assessment method considering the extreme weather based on an SSA-LSTM model (Sparrow Search Algorithm sparrow search algorithm/Long Short-Term Memory network), which can realize accurate perception and timely alarm of various weather safety hidden dangers and ensure that the safety of an electric line is changed from 'post-disaster alarm' to 'pre-disaster early warning'.
The purpose of the invention is realized in the following way: a power distribution network line security risk assessment method considering extreme weather comprises the following steps:
step a, acquiring historical fault data and contemporaneous external meteorological data of a power distribution network;
b, training an SSA-LSTM model by using the meteorological data and the number of fault lines obtained in the step a, establishing a power distribution network line fault probability model under the consideration of extreme weather factors, and outputting a probability value of line faults under the current meteorological conditions based on the power distribution network line fault probability model;
c, establishing a power distribution network fault result evaluation index by using the fault data obtained in the step a;
and d, integrating the fault probability value obtained in the step b with the power distribution network fault result evaluation index column-written risk matrix obtained in the step c, and evaluating the security risk level of the power distribution network under the current meteorological conditions.
Preferably, the meteorological data comprises precipitation, wind speed, thunderstorm and high temperature, and the fault data comprises fault line number, power outage capacity, power shortage capacity, power outage user hours and importance level of power outage users.
Preferably, in the step d, the security risk level of the power distribution network is determined by a risk value of the power distribution network line, the risk value of the power distribution network line is determined by the possibility of a fault and the failure consequence degree, specifically, the risk value of the power distribution network line is determined by multiplying the frequency of the occurrence of the fault by the failure effect degree, and the calculation formula is as follows:
wherein n is the number of power distribution network line faults in the area, i is the line number, i= … n, risk is the Risk value of equipment failure and power failure, P (i) is the frequency of the faults of the power distribution network line in the area, and SEV (i) is the degree of the consequences of the equipment failure and power failure.
Preferably, the fault frequency P (i) of the power distribution network line is output by the built SSA-LSTM model, wherein grading data of precipitation, wind speed, thunderstorm and high temperature form SSA-LSTM input, and the change of external meteorological conditions is reflected through the input of meteorological parameters.
Preferably, in the step c, the evaluation of the power distribution network line fault outage influence includes the establishment of a fault outage load set S and the establishment of a fault outage result evaluation index.
Preferably, the evaluation index of the fault outage result is represented by comprehensive load loss, wherein the comprehensive load loss comprises outage capacity, outage amount, outage user hours and outage user importance level;
after any line of the power distribution network fails, the power failure capacity L in the power failure load set S L The annual average load of the distribution transformer is represented by the following calculation formula:
in which A i The power is obtained by copying the power consumption of the distribution transformer corresponding to the load i in the power failure load set S A The annual average running time of the distribution transformer in the power distribution network;
power shortage E L The calculation formula of (2) is as follows:
t in R For the failure recovery time, assume T R Continuous power failure in the power failure load set S in time;
hours of power failure user pou The calculation formula of (2) is as follows:
w in the formula Lui The number of users in the power failure load set S;
importance level W of power failure user pou The calculation formula of (2) is as follows:
in which a is m The method comprises the steps that the level of an mth user in a corresponding i-number load in a power failure load set S is valued, urban users take Ka1, urban users take Ka2, and rural users take 1;
determining the weight of each evaluation index by adopting an entropy weight method, calculating the information entropy of each index, calculating the information utility value, and normalizing to obtain the entropy weight of each index;
first, the information entropy e of four evaluation indexes is calculated j The calculation formula is as follows:
in e j The larger the information entropy of the j index is, and the smaller the corresponding information quantity is;
then calculate the information utility value d j The calculation formula is as follows:
then normalizing the information utility value to obtain entropy weight omega of four evaluation indexes j The calculation formula is as follows:
and finally, calculating the comprehensive coincidence loss SEV, wherein the calculation formula is as follows:
omega in 1 、ω 2 、ω 3 、ω 4 The entropy weights of the four evaluation indexes are obtained.
Preferably, in the step d, the failure frequency of the line under the current meteorological condition output by the SSA-LSTM model is divided into five grades according to the frequency, namely rare [0,3 respectively]Unlikely [4,10]Possibly [11,15]Most likely [16, 25]Almost certainly [26 ],
]the method comprises the steps of carrying out a first treatment on the surface of the The results of the analysis of the consequences of a fault outage are divided into five classes according to the size of the hazard to the user, which are respectively negligible [0,0.15 ]]Smaller [0.16,0.35 ]]Intermediate [0.36,0.50 ]]Larger [0.51, 0.70]Catastrophic [0.71,1 ]]；
Thus, the risk matrix shown in the following table is written to obtain a security risk assessment grade:
remapping 25 risk values in the matrix into I-V levels according to the range standard of risk result division: the interval [20,25] is V grade; the intervals [15,20] are IV levels; the interval [10,15] is class III; the interval [5,10] is grade II; the interval [1,5] is level I.
Due to the adoption of the technical scheme, the invention has the beneficial effects that:
(1) According to the method, the SSA-LSTM model is trained by using historical operation data, fault data and simultaneous meteorological data of the power distribution network, a power distribution network line fault prediction model considering weather factors is generated, the change of external meteorological conditions is reflected through the input of meteorological parameters, the fault frequency of the power distribution network line in 24 hours under the current operation conditions is obtained, then a fault outage result evaluation index is established, the personalized difference of line faults is reflected through four indexes of outage capacity, outage amount, outage user hours and importance level of the outage user, then the fault frequency and the fault outage result evaluation index are synthesized, risk evaluation is realized on the power distribution network by using a risk matrix, and the safety risk level of the power distribution network line is accurately evaluated;
(2) According to the invention, deconstructing analysis is carried out on data through autonomous learning training by utilizing a deep learning technology, useless and similar information is deleted, implicit information is extracted, risk trend prediction is carried out, unknown abnormal information is identified, the defect that the early hidden danger cannot be found in the traditional threshold alarming or tripping mode is overcome on the basis of the unknown abnormal information, accurate perception and timely alarming of various potential safety hazards are realized, corresponding control measures are adopted on the power grid side and the information side in advance, the workload of operation and maintenance personnel is reduced, and meanwhile, the safety risk is isolated and blocked before causing great influence, so that the safety of an electric circuit is changed from 'post-disaster alarming' to 'pre-disaster early warning'.
Drawings
Fig. 1 is a flowchart of a power distribution network line security risk assessment method of the present invention.
FIG. 2 is a flow chart of the SSA-LSTM fault probability model of the present invention.
FIG. 3 is a diagram of a security risk assessment indicator according to the present invention.
FIG. 4 is a graph of SSA-LSTM model convergence at the time of specific verification of the present invention.
FIG. 5 is a graph comparing the predicted and actual results of a test set for a specific verification of the present invention.
Fig. 6 is a graph of the combined load loss of a distribution line fault at the time of specific verification of the present invention.
Detailed Description
The technical scheme of the invention is further specifically described below through examples and with reference to the accompanying drawings.
As shown in fig. 1, the invention provides a power distribution network line security risk assessment method considering extreme weather, which comprises the following steps:
and a step a, acquiring historical fault data of the power distribution network and contemporaneous external meteorological data.
Wherein the data sources include: the system comprises a power distribution network supervision and management system, a power distribution network geographic information system and a meteorological information system.
Wherein, meteorological data includes: precipitation, wind speed, thunderstorm, high temperature.
Wherein the fault data comprises: the number of fault cables, the power failure capacity, the power failure quantity, the power failure user hours and the importance level of the power failure user; the fault data are used for calculating the power failure result evaluation index in the subsequent step c so as to calculate the comprehensive load loss.
And b, training an SSA-LSTM model by using the historical meteorological data obtained in the step a and the number of fault lines of the regional distribution network, and establishing a distribution network line fault probability model under the consideration of extreme weather factors.
Step b1, sparrow search algorithm:
sparrows are a sort of social birds that each individual has a well-defined division and collaboration in their colony foraging process. According to the foraging and anti-predation processes of sparrow populations, a novel intelligent optimization algorithm, namely a sparrow search algorithm (Sparrow Search Algorithm, SSA), is provided. The sparrow population is classified into discoverers, enrollees and scouts according to the different branches in the sparrow predation process and the different locations in the population. Wherein the discoverers are the earliest discovered individuals in the sparrow population, and they lead the predation direction of the population, and act as food seekers in the population. The joining person is predated by the finder after the finder. The scout is located at the periphery of the population in the predation process, and when the predator is found, a signal is sent immediately to remind the found person and the added person to go to the safe area for foraging. During the foraging process of the sparrow population, the role played by each individual is not constant, but the proportion of discoverers and participants in the population is constant, i.e., when one participant is converted to a discoverer, one discoverer also becomes a participant.
First, a sparrow population with n individuals can be represented as follows:
where d represents the dimension of the problem to be optimized and n is the number of individuals in the sparrow population.
The fitness f of each individual in the sparrow population is represented by the following formula:
in the sparrow search algorithm, the discoverer firstly discovers food and predates, and guides the foraging direction of the subsequent joiners, so that compared with other individuals, the discoverer has a larger foraging search range and has higher fitness in the search process. The size of the individual's energy reserve is positively correlated with its corresponding fitness level. According to the relation between fitness and energy reserve and the individual role change rule, in the subsequent iteration process, the position description formula of the discoverer is as follows:
where t represents the number of iterations, j=1, 2,3 … m,
is a fixed constant for the maximum number of iterations. L is a matrix of 1*d, each element in L being 1./>
Is a random number subject to a standard normal distribution, +.>
Is the position information of the ith individual in the jth dimension,/->
Is [0,1 ]]Random number between->
Representing the early warning value->
Representing a security value.
When (when)
When the scout in the sparrow population is alerted to the danger outside, the scout calls to remind other individuals, and the sparrow population rapidly escapes to the safe area to find food.
When (when)
When the sparrow population is in a safe state, predators are not found in the foraging area, and the discoverers can further search.
For the participants, they have a monitoring and competing relationship with the discoverers, the participants monitor the discoverers ' dynamics from time to time, if the discoverers search for better food sources, the participants change their own locations to compete with the discoverers for food, if the competition is successful, the participants will replace the discoverers ' locations and obtain the discoverers ' food. The location information update formula of the enrollee is as follows:
wherein,,
is the optimal position for the current state finder, < >>
Is the worst position in the total population. A is an element 1 or-1 with a dimension of +.>
And satisfy->
. When->
When the i-th participant is too low to get enough food, it will update the location to predate to get enough energy reserve.
For a scout, a call is made when a predator approaches, and discoverers and joiners in the population discard the current food and immediately transfer to a safe area. The scout location update formula is as follows:
wherein,,
is the current global optimum. />
Is a normal distribution random number with the obeying mean value of 0 and the variance of 1, and is used as a step control parameter. />
Is a random number, < >>
Is the adaptability of the current sparrow individuals. />
For the current global best fitness +.>
Is the current global worst fitness. />
To avoid a constant with zero denominator.
When (when)
Indicating that sparrows are at the edge of the population, and are extremely vulnerable to attack by predators. />
Representing the optimal safest location in the population. />
When the individuals in the sparrow population are aware of the danger surrounding, the sparrow population quickly moves to other individuals, and the predation risk is reduced.
Step b2, a distribution network line fault probability prediction model based on SSA-LSTM:
LSTM (Long Short-Term Memory) is a Long-Term and Short-Term Memory network, is a time recurrent neural network, and is trained by acquiring a certain amount of sample data, and the weight value and the bias Term of the neural network are always updated so that an error function is reduced along the negative gradient direction until the requirement is met, but the problem of low local optimal solution and slow convergence speed is easily caused when nonlinear fitting is performed.
Therefore, the sparrow search algorithm is utilized to continuously correct the hidden layer node number during LSTM neural network training, and the optimal weight value and the bias term are obtained, so that the error between the output value and the expected value of the network is reduced, and the convergence rate of the LSTM neural network is further improved.
Step b3, a distribution network line fault probability model based on SSA-LSTM:
the sparrow search optimization algorithm is used for continuously optimizing the node number of the hidden layer of the LSTM neural network model by iteratively searching the optimal position in the population individuals, so that the weight value and the bias term of the LSTM model are obtained, the convergence speed of the model is improved, and the error between the true value and the predicted value is reduced.
The algorithm flow chart of the SSA optimized LSTM neural network model is shown in fig. 2, and the specific steps are as follows: (1) acquiring meteorological data and line fault data, preprocessing the data, removing irrelevant data, dividing the line fault data into two groups, wherein one group is used as a training sample, and the rest is used as a test sample; (2) determining an activation function of the LSTM neural network; (3) resetting network structure parameters of the LSTM neural network hidden layer; (4) calculating fitness and optimal individual positions of the sparrow population; (5) updating the individual through foraging and anti-predation behaviors, and calculating the positions of the updated discoverers, the enrollees and the scouts; (6) judging whether the model meets the convergence accuracy requirement, executing the next step if the model meets the convergence accuracy requirement, otherwise, returning to the step (4), and continuously searching for the individual optimal position; (7) and giving the obtained optimal position to the weight value and the bias term of the LSTM to obtain a trained model, and inputting a test sample to carry out solving and model accuracy verification.
In order to evaluate the model prediction accuracy, the root mean square error RMSE, the mean absolute error MAE, the mean absolute percentage error MAPE and the decision coefficient R2 are selected, and the performance of the SSA-LSTM model in the aspect of fault probability prediction is verified.
Wherein,,
。
and c, verifying the accuracy and the effectiveness of the SSA-LSTM model by the line fault number of the certain urban distribution network and the external meteorological data obtained in the step a.
Firstly, preprocessing data, screening out data which are obviously irrelevant to weather, such as equipment factors, planned power regulation, construction influence and the like, and counting the number of fault lines in the area by taking a day as a unit. According to the analysis of the local climate characteristics, precipitation, wind speed, thunderstorm and high temperature are dominant meteorological factors of the line faults of the power distribution network. The change of external weather conditions is reflected by the input of weather data. Dividing the data into two groups, wherein one group is used as a training sample, and the total data is failure data of 240 days; the other group is taken as a test sample, and has 120 days of fault data, and is used for testing the precision of the SSA-LSTM model.
Initial parameter setting is carried out on the SSA optimization algorithm, the sparrow population number is set to be 50, the proportion of the adding person is 0.7, and the proportion of the alerter is 0.2.
Fig. 4 shows the convergence curve of SSA-LSTM model, and it can be seen that SSA has good optimizing capability, can find the optimal value quickly, and has quick convergence as a result. At iteration number 5, SSA finds the current optimum and converges rapidly. At this time, the LSTM model has 3 hidden layers, the initial learning rate is 0.002, and the optimal regularization coefficient is 100.
FIG. 5 is a graph comparing predicted values to actual values of a test set. The test set predicts the evaluation index and,
=0.96906, mae=0.80982, rmse= 1.1026. Wherein the decision coefficient->
The value of (1, 0) is given, and the closer the value is to 1, the higher the model fitting degree is. FIG. 5 illustrates that the weather-factor-considered fault probability model fits the historical fault samples well, the prediction model has high accuracy, and the fault probability model reflects the influence of external weather factors on the distribution line well when the external weather conditions are severe, such asStorm, lightning strike, strong wind, etc., the number of faults of the distribution line in the area is steeply increased due to the impact of meteorological factors on the line.
Step c, evaluating indexes of fault and power failure results:
the fault outage result evaluation comprises the establishment of a fault outage load set S and the establishment of a fault outage influence index. FIG. 3 is a diagram showing a security risk assessment index structure according to the present invention.
The evaluation index of the failure and power failure results is expressed by comprehensive load loss, and the comprehensive load loss comprises power failure capacity, power failure amount, power failure user hours and importance level of the power failure user.
Aiming at the power shortage L in a power failure load set S after any line of a power distribution network fails L Expressed in terms of annual average load of the distribution transformer:
wherein A is i The power is obtained by copying the power consumption of the distribution transformer corresponding to the load i in the power failure load set S A Is the annual average running time of distribution transformer in the distribution network.
E for lack of supply L The representation is:
wherein T is R For the failure recovery time, assume T R And continuously cutting off power in the power-off load set S in the time.
Hours of power failure user pou The representation is:
wherein W is Lui Is the number of users in the power outage load set S.
W for importance level of power failure user pou The representation is:
wherein a is m For the level value of the mth user in the corresponding i-number load in the power failure load set S, the urban user takes Ka1, the urban user takes Ka2, and the rural user takes 1.
And determining the weight of each evaluation index by adopting an entropy weight method, calculating the information entropy of each index, calculating the information utility value, and normalizing to obtain the entropy weight of each index.
Calculating the information entropy e of four evaluation indexes j The calculation formula is as follows:
wherein e j The larger the information entropy of the j-th index is, and the smaller the corresponding information amount is.
Calculating an information utility value d j The calculation formula is as follows:
normalizing the information utility value to obtain entropy weight omega of four evaluation indexes j The calculation formula is as follows:
the comprehensive coincidence loss SEV is calculated as follows:
wherein omega 1 、ω 2 、ω 3 、ω 4 The entropy weights of the four evaluation indexes are obtained.
The users in the example distribution network structure mainly comprise urban users, town users and rural users. And (5) counting the power outage capacity, the power outage quantity, the power outage user hours and the importance level of the power outage user of each user, and determining the weight of each evaluation index through an entropy weight method. Then, the daily integrated load loss of each sample was calculated, and the calculation result is shown in fig. 6.
And d, combining the fault probability prediction result in the step b and the comprehensive coincidence loss in the step c, listing and writing a risk matrix, and evaluating the security risk level of the regional power distribution network.
The line risk assessment results and the early warning grades of part of the sample days are shown in the following table:
the power distribution network fault outage risk is determined by the probability of fault occurrence and the severity of fault outage results. The risk value of the power transmission line of the power distribution network is determined by multiplying the probability of occurrence of faults by the severity of the fault outage result.
The Risk value of equipment failure and power failure is Risk, the frequency of faults of the regional power distribution network line is P (i), and the consequence degree of equipment failure and power failure is SEV (i).
The probability of failure P (i) output by the SSA-LSTM model is divided into five levels according to the probability size, namely rare, unlikely, probable, likely and almost positive. The results of the analysis of the consequences of the fault SEV (i) are classified into five classes, negligible, small, medium, large, catastrophic, respectively, according to the size of the hazard to the user. And (3) writing a risk matrix shown in the following table, and remapping 25 risk values in the matrix into I-V levels according to the range standard of risk result division: the interval [20,25] is V grade; the intervals [15,20] are IV levels; the interval [10,15] is class III; the interval [5,10] is grade II; the interval [1,5] is level I.
The fault condition of the distribution network line is closely related to external meteorological factors such as precipitation, thunder and lightning, wind speed, temperature and the like. If the 21 st sample day risk evaluation level is V, storm and thunderstorm are accompanied in the power distribution network area on the same day, the weather condition is extremely bad, and the power distribution network line is extremely easy to be impacted by external weather factors to fail, so that the safety evaluation method can fully reflect the influence of the external weather conditions on the power distribution network line, and accurately predict the safety level of the power distribution network line under the current weather condition.
And (3) evaluating the possibility of risk occurrence and the consequences of risk occurrence from two aspects of quantification (grade score) and qualitative (grade description) for risks faced by the power distribution network faults, and finally obtaining the safety grade of the power distribution network line under the current meteorological conditions. And obtaining fault probability and fault outage results according to the power distribution network simulation model of a certain region, calculating a risk value, and referring to the risk grade obtained by the risk matrix, so that the accuracy of risk grade assessment is improved.
According to the risk assessment result, operation and maintenance personnel can take corresponding control measures on the power grid side and the information side in advance, so that the workload of the operation and maintenance personnel is reduced, the safety risk is isolated and blocked before great influence is caused, and the method has very important significance for the safety of an electric circuit from 'post-disaster alarming' to 'pre-disaster early warning'.
Finally, it should be noted that the above-mentioned embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same, and although the present invention has been described in detail with reference to the above-mentioned embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the specific embodiments of the present invention without departing from the spirit and scope of the present invention, and any modifications and equivalents are intended to be encompassed in the scope of the claims of the present invention.
Claims (7)
1. The power distribution network line security risk assessment method considering extreme weather is characterized by comprising the following steps of:
step a, acquiring historical fault data and contemporaneous external meteorological data of a power distribution network;
b, training an SSA-LSTM model by using the meteorological data and the number of fault lines obtained in the step a, establishing a power distribution network line fault probability model under the consideration of extreme weather factors, and outputting a probability value of line faults under the current meteorological conditions based on the power distribution network line fault probability model;
c, establishing a power distribution network fault result evaluation index by using the fault data obtained in the step a;
and d, integrating the fault probability value obtained in the step b with the power distribution network fault result evaluation index column-written risk matrix obtained in the step c, and evaluating the security risk level of the power distribution network under the current meteorological conditions.
2. The method for evaluating the security risk of a power distribution network line taking extreme weather into consideration according to claim 1, wherein: the meteorological data comprise precipitation, wind speed, thunderstorm and high temperature, and the fault data comprise fault line number, power failure capacity, power failure quantity, power failure user hours and importance level of the power failure user.
3. The method for evaluating the security risk of a power distribution network line taking extreme weather into consideration according to claim 2, wherein: in the step d, the security risk level of the power distribution network is determined by a risk value of the power distribution network line, the risk value of the power distribution network line is determined by the possibility of a fault and the failure consequence degree, specifically, the risk value of the power distribution network line is determined by multiplying the frequency of the occurrence of the fault of the line and the failure influence degree, and the calculation formula is as follows:
wherein n is the number of power distribution network line faults in the area, i is the line number, i= … n, risk is the Risk value of equipment failure and power failure, P (i) is the frequency of the faults of the power distribution network line in the area, and SEV (i) is the degree of the consequences of the equipment failure and power failure.
4. A method for evaluating the security risk of a power distribution network line taking extreme weather into account according to claim 3, wherein: the fault frequency P (i) of the power distribution network line is output by the built SSA-LSTM model, wherein grading data of precipitation, wind speed, thunderstorm and high temperature form SSA-LSTM input, and the change of external weather conditions is reflected through the input of weather parameters.
5. The method for evaluating the security risk of a power distribution network line taking extreme weather into consideration according to claim 2, wherein: in the step c, the evaluation of the power distribution network line fault power failure influence comprises the establishment of a fault power failure load set S and the establishment of a fault power failure result evaluation index.
6. The method for evaluating the security risk of the power distribution network line considering extreme weather according to claim 5, wherein: the fault power failure result evaluation index is expressed by comprehensive load loss, wherein the comprehensive load loss comprises power failure capacity, power failure quantity, power failure user hours and importance level of the power failure user;
after any line of the power distribution network fails, the power failure capacity L in the power failure load set S L The annual average load of the distribution transformer is represented by the following calculation formula:
in which A i The power is obtained by copying the power consumption of the distribution transformer corresponding to the load i in the power failure load set S A The annual average running time of the distribution transformer in the power distribution network;
power shortage E L The calculation formula of (2) is as follows:
t in R For the failure recovery time, assume T R Continuous power failure in the power failure load set S in time;
hours of power failure user pou The calculation formula of (2) is as follows:
w in the formula Lui The number of users in the power failure load set S;
importance level W of power failure user pou The calculation formula of (2) is as follows:
in which a is m The method comprises the steps that the level of an mth user in a corresponding i-number load in a power failure load set S is valued, urban users take Ka1, urban users take Ka2, and rural users take 1;
determining the weight of each evaluation index by adopting an entropy weight method, calculating the information entropy of each index, calculating the information utility value, and normalizing to obtain the entropy weight of each index;
first, the information entropy e of four evaluation indexes is calculated j The calculation formula is as follows:
in e j The larger the information entropy of the j index is, and the smaller the corresponding information quantity is;
then calculate the information utility value d j The calculation formula is as follows:
then normalizing the information utility value to obtain entropy weight omega of four evaluation indexes j The calculation formula is as follows:
and finally, calculating the comprehensive coincidence loss SEV, wherein the calculation formula is as follows:
omega in 1 、ω 2 、ω 3 、ω 4 The entropy weights of the four evaluation indexes are obtained.
7. The method for evaluating the security risk of a power distribution network line taking extreme weather into consideration according to claim 2, wherein the method comprises the following steps ofThe method comprises the following steps: in the step d, the fault frequency of the line under the current meteorological conditions output by the SSA-LSTM model is divided into five grades according to the frequency, namely rare [0,3 ]]Unlikely [4,10]Possibly [11,15]Most likely [16, 25]Almost certainly [26 ],
]the method comprises the steps of carrying out a first treatment on the surface of the The results of the analysis of the consequences of a fault outage are divided into five classes according to the size of the hazard to the user, which are respectively negligible [0,0.15 ]]Smaller [0.16,0.35 ]]Intermediate [0.36,0.50 ]]Larger [0.51, 0.70]Catastrophic [0.71,1 ]]；
Thus, the risk matrix shown in the following table is written to obtain a security risk assessment grade:
remapping 25 risk values in the matrix into I-V levels according to the range standard of risk result division: the interval [20,25] is V grade; the intervals [15,20] are IV levels; the interval [10,15] is class III; the interval [5,10] is grade II; the interval [1,5] is level I.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310049388.3A CN116090821A (en) | 2023-02-01 | 2023-02-01 | Power distribution network line security risk assessment method considering extreme weather |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310049388.3A CN116090821A (en) | 2023-02-01 | 2023-02-01 | Power distribution network line security risk assessment method considering extreme weather |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116090821A true CN116090821A (en) | 2023-05-09 |
Family
ID=86202139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310049388.3A Pending CN116090821A (en) | 2023-02-01 | 2023-02-01 | Power distribution network line security risk assessment method considering extreme weather |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116090821A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116805210A (en) * | 2023-08-21 | 2023-09-26 | 国网安徽省电力有限公司合肥供电公司 | Intelligent power grid risk identification management and control method based on big data |
| CN117314175A (en) * | 2023-11-28 | 2023-12-29 | 合肥优尔电子科技有限公司 | Power grid system geological risk monitoring and evaluating method based on severe weather |
| CN117349692A (en) * | 2023-12-04 | 2024-01-05 | 国网江西省电力有限公司南昌供电分公司 | Distribution line lightning early warning method integrating multiple lightning early warning factors |
| CN118297397A (en) * | 2024-04-16 | 2024-07-05 | 粤港澳大湾区气象监测预警预报中心(深圳气象创新研究院) | Object-oriented urban extreme weather pressure testing method |
-
2023
- 2023-02-01 CN CN202310049388.3A patent/CN116090821A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116805210A (en) * | 2023-08-21 | 2023-09-26 | 国网安徽省电力有限公司合肥供电公司 | Intelligent power grid risk identification management and control method based on big data |
| CN116805210B (en) * | 2023-08-21 | 2024-01-12 | 国网安徽省电力有限公司合肥供电公司 | Intelligent power grid risk identification management and control method based on big data |
| CN117314175A (en) * | 2023-11-28 | 2023-12-29 | 合肥优尔电子科技有限公司 | Power grid system geological risk monitoring and evaluating method based on severe weather |
| CN117314175B (en) * | 2023-11-28 | 2024-02-23 | 合肥优尔电子科技有限公司 | Power grid system geological risk monitoring and evaluating method based on severe weather |
| CN117349692A (en) * | 2023-12-04 | 2024-01-05 | 国网江西省电力有限公司南昌供电分公司 | Distribution line lightning early warning method integrating multiple lightning early warning factors |
| CN117349692B (en) * | 2023-12-04 | 2024-05-07 | 国网江西省电力有限公司南昌供电分公司 | Distribution line lightning early warning method integrating multiple lightning early warning factors |
| CN118297397A (en) * | 2024-04-16 | 2024-07-05 | 粤港澳大湾区气象监测预警预报中心(深圳气象创新研究院) | Object-oriented urban extreme weather pressure testing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116090821A (en) | Power distribution network line security risk assessment method considering extreme weather | |
| WO2022135265A1 (en) | Failure warning and analysis method for reservoir dispatching rules under effects of climate change | |
| CN105117602B (en) | A kind of metering device running status method for early warning | |
| CN110766212B (en) | Ultra-short-term photovoltaic power prediction method for historical data missing electric field | |
| CN104200288B (en) | A kind of equipment fault Forecasting Methodology based on dependency relation identification between factor and event | |
| CN106779129A (en) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor | |
| CN109657966A (en) | Transmission line of electricity risk composite valuations method based on fuzzy mearue evaluation | |
| CN115220133B (en) | Rainfall prediction method, device and equipment for multiple meteorological elements and storage medium | |
| CN109491339B (en) | Big data-based substation equipment running state early warning system | |
| CN115689114A (en) | Submarine cable running state prediction method based on combined neural network | |
| CN103440410A (en) | Main variable individual defect probability forecasting method | |
| CN110794308A (en) | Method and device for predicting train battery capacity | |
| CN107527121A (en) | A kind of method of the information system running status diagnosis prediction of power network | |
| CN114897204A (en) | Method and device for predicting short-term wind speed of offshore wind farm | |
| CN113988655A (en) | Power transmission line running state evaluation method considering multiple meteorological factors | |
| CN114357670A (en) | Power distribution network power consumption data abnormity early warning method based on BLS and self-encoder | |
| CN108764550A (en) | Lightning Warning method and system based on transmission line information data | |
| AU2019100631A4 (en) | Self-correcting multi-model numerical rainfall ensemble forecasting method | |
| CN112100904A (en) | ICOA-BPNN-based distributed photovoltaic power station active power virtual acquisition method | |
| CN113689053B (en) | Strong convection weather overhead line power failure prediction method based on random forest | |
| CN114564513A (en) | Sea fog prediction method, device, equipment and storage medium based on neural network | |
| CN117434450A (en) | Battery health state prediction method and system | |
| CN108446202A (en) | A kind of judgment method of the safe condition of calculator room equipment | |
| CN112418662A (en) | Power distribution network operation reliability analysis method using artificial neural network | |
| CN112036682A (en) | Early warning method and device for frequent power failure |
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 |