CN108694479A - Consider the distribution network reliability prediction technique that weather influences time between overhaul - Google Patents
Consider the distribution network reliability prediction technique that weather influences time between overhaul Download PDFInfo
- Publication number
- CN108694479A CN108694479A CN201810758399.8A CN201810758399A CN108694479A CN 108694479 A CN108694479 A CN 108694479A CN 201810758399 A CN201810758399 A CN 201810758399A CN 108694479 A CN108694479 A CN 108694479A
- Authority
- CN
- China
- Prior art keywords
- overhaul
- weather
- time
- variable
- distribution network
- 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
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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
Abstract
The invention discloses a kind of distribution network reliability prediction techniques that consideration weather influences time between overhaul, include the following steps:(1) according to the historical data of multiple weather variables, the related coefficient of each weather variable and time between overhaul and the importance values to time between overhaul are calculated separately;(2) it is based on each corresponding related coefficient of weather variable and importance values chooses prevailing weather variable from the multiple weather variable;(3) historical data based on the prevailing weather variable is trained BP neural network model, and is predicted the following time between overhaul index according to the BP neural network model for completing training;(4) prediction result based on the following time between overhaul index predicts distribution network reliability.The present invention can be directed to influence of the different weather variable to distribution mesh element time between overhaul, predict distribution network reliability, to improve the accuracy of fail-safe analysis.
Description
Technical field
The present invention relates to a kind of distribution network reliability prediction technique, especially a kind of consideration weather is to time between overhaul shadow
Loud distribution network reliability prediction technique.
Background technology
Distribution system refers to the part that electricity generation system transmits electric power from transformation node to end user comprising it is various not
With the substation of voltage class, distribution transforming device, distribution line and other electrically set by various types of user connects
It is standby.The system of 35KV or more is often known as high tension distribution system, 6-20KV systems for intermediate distribution system, 380V/220V is low
It is press-fitted electric system.The differentiation of this Distribution level can not be defined only according to voltage class, it is also necessary to consider the work(of facility
Energy.
Since power distribution network stops for that can cause larger economic loss, power supply of the user to power distribution network to user and society
Reliability requirement is higher and higher.Currently, distribution network reliability analysis method be mainly analytic method and and simulation, wherein analytic method
Including fault mode method and fault effects analysis method etc., but with the expansion of distribution network, the calculation amount of such algorithm drastically on
It rises, therefore occurs many refined Hook Jeeves algorighms on the basis of analytic method, such as shortest path method, Fault traversal method, network reduction
Equivalent method etc..Such optimization algorithm can simplify Complicated Distribution Network system, however such algorithm do not count mostly and failure after
Load transfer, it is difficult to meet reliability requirement.It is proposed to this end that the method that piecemeal cuts load after failure;Consider voltage matter
The factors such as amount and protective value;Detailed load transfer case after failure is considered using capacity-constrained, and proposes and is adapted to
The reliability calculation method of extensive practical power distribution network.It should be pointed out that above-mentioned algorithm almost only considered fault outage feelings
Condition, and it is unaware that influence problem of the weather to power distribution network time between overhaul.
In fact, according to City of Guangdong Province power supply bureau distribution network reliability statistical data in 2017 it is found that city's power distribution network is used
Family mean failure rate power off time is 2.68h, and user is averaged pre-arranged power off time as 1.49h.And pre-arranged have a power failure not only with
Power supply plan and power supply management level are related, are also largely influenced by Changes in weather.Therefore, ignore weather to matching
The influence of electric network element time between overhaul would potentially result in the reliability for being unable to Accurate Prediction power distribution network.
Invention content
Goal of the invention:The present invention is intended to provide a kind of consideration weather matches power distribution network pre-arranged off interval time effects
Electric network reliability prediction technique, to further increase the accuracy of distribution network reliability analysis.
Technical solution:The present invention's considers distribution network reliability of the weather to power distribution network pre-arranged off interval time effects
Prediction technique includes the following steps:(1) according to the historical data of multiple weather variables, each weather variable and maintenance are calculated separately
The related coefficient of interval time and importance values to time between overhaul;(2) it is based on each corresponding phase relation of weather variable
Number and importance values choose prevailing weather variable from the multiple weather variable;(3) going through based on the prevailing weather variable
History data are trained BP neural network model, and according to complete training BP neural network model to the following maintenance interval when
Between index predicted that the time between overhaul index includes the fault correction time and failure rate of element;(4) it is based on
The prediction result of the following time between overhaul index predicts distribution network reliability.
Further, in step (1), the multiple weather variable includes:Wind variable, frost state quantities, temperature become
Amount, the relative humidity variable of air and lightning stroke variable;The wherein described frost state quantities are calculated by the following formula:
Wherein, LNDRSc be the current value of maximum number of days of sleet assessment, the current value that TPSc is rainfall and snowfall,
LNDCLTc is the current value of continuous low temperature number of days, and LNDRSav is that average value, the TPSav of the maximum number of days of sleet assessment are rainfall
Average value and LNDCLTav with snowfall are the average value of continuous low temperature number of days.
Further, in step (1), it is related to time between overhaul that each weather variable is calculated by the following formula
Coefficient Correlation:
Wherein, x is the probability of certain year a certain weather occurrences, and y is this year element health degree under the weather variable
Decrease speed,For the probability average of the weather occurrences of each year,For each year under the weather variable element health degree
The average value of decrease speed.
Further, in step (1), the importance values are calculated by P value methods.
Further, in step (1), by related coefficient more than 0.5 and when importance values are more than 0.8, corresponding weather becomes
Amount is selected as prevailing weather variable.
Further, step (3) further comprises:(3.1) using the historical data of the prevailing weather variable as BP god
Inputting to calculate the predicted value of time between overhaul through network model;(3.2) by the predicted value of the time between overhaul of calculating
It is compared to calculate error with actual value, and the BP neural network model is trained according to the error of feedback, it is described
Training includes the error based on feedback to update the threshold value and weights in BP neural network model;(3.3) step is repeated
(3.1) and (3.2) are until the square-error between the predicted value and known predicted value of the time between overhaul that training obtains reaches
Or the threshold value less than setting, to complete the training to BP neural network model;(3.4) based on the BP neural network mould for completing training
Type predicts the following time between overhaul index.
Further, in step (3.2), the threshold value and power in BP neural network model are updated using gradient descent method
Value.
Further, step (4) includes the following steps:(4.1) weather condition for assuming certain following a period of time, using pre-
The following distribution mesh element time between overhaul index of survey calculates in distribution mesh element 1 year:
Element total failare repair time TCOH:TCOH=∑si∈Rri;
System System average interruption duration index SAIDI:
System System average interruption frequency index SAIFI:
The reliability index ASAI of distribution mesh element:
Wherein, riFor the fault correction time of load bus i, niFor the number of users of load bus i, R is load bus
Number, TCO is element major overhaul number and TCO is that the failure rate based on element obtains;
(4.2) system System average interruption duration index SAIDI, system System average interruption frequency index SAIFI are based on and is matched
The reliability index ASAI of electric network element predicts the reliability of power distribution network.
Advantageous effect:Compared with the existing technology, the reality of the invention that can preferably adapt to various different grid structures is matched
The reliability assessment of power grid.Power supply unit can be according to reliability assessment as a result, instructing planning and the day-to-day operation repair of power distribution network
Work.The present invention can also be directed to influence of the different weather variable to distribution mesh element time between overhaul, reliable to power distribution network
Property is predicted.
Description of the drawings
Fig. 1 is the distribution network reliability prediction technique flow chart of the present invention;
Fig. 2 is BP neural network model schematic;
Fig. 3 is influence statistical chart of the weather variable to electric network element time between overhaul;
Fig. 4 is the statistical chart of the Nanjing Suburb state of weather in 2014 of the present invention;
Fig. 5 is the statistics of probability of malfunction and repair time of the Nanjing Suburb distribution mesh element under different weather state
Figure.
Specific implementation mode
Below in conjunction with attached drawing, further the present invention is described in detail.
As shown in Figure 1, the distribution network reliability prediction technique of the present invention includes the following steps:
Step 1:According to the historical data of multiple weather variables, each weather variable and time between overhaul are calculated separately
Related coefficient and importance values to time between overhaul.It specifically includes:
1) variable of gust velocity, synthesis demeanour and mean wind speed as research windage is selected;
2) anti-lightning strike electric current and peak value thunder-strike current is selected to indicate the variable of effects of lightning;
3) with maximum number of days (LNDRS), rainfall and the snowfall (TPS) and continuous low temperature number of days of sleet assessment
Any one of (LNDCLT) or it is several come to assess ice condition be one sided, therefore the present invention is added using equal weight
Come indicate frost situation influence variable, by sleet assessment maximum number of days, rainfall and snowfall and continuous low temperature number of days phase
Deng weight phase Calais indicate frost situation influence variable:
Wherein, LNDRScCurrent value, TPS for the maximum number of days of sleet assessmentcFor the current value of rainfall and snowfall,
LNDCLTcFor the current value of continuous low temperature number of days, LNDRSavFor average value, the TPS of the maximum number of days of sleet assessmentavFor rainfall and
The average value and LNDCLT of snowfallavFor the average value of continuous low temperature number of days.
4) correlation level of the analysis different weather variable to distribution mesh element time between overhaul;
5) it can be indicated with related coefficient:
Wherein, x is the probability of certain year a certain weather occurrences, and y is this year element health degree under the weather variable
Decrease speed,For the probability average of the weather occurrences of each year,For each year under the weather variable element health degree
Decrease speed average value;
6) value of related coefficient is higher, then shows that influence of the weather variable to element time between overhaul is bigger.
Step 2:It is selected from the multiple weather variable based on each corresponding related coefficient of weather variable and importance values
Take prevailing weather variable
About different weather variable to distribution mesh element time between overhaul, the P value methods pair in statistics can be passed through
The historical data of multiple weather variables is analyzed, to calculate separately out the p value for different weather variable:
P=2P (z > |zc|),
Wherein, z is that weather variable corresponds to time between overhaul statistic, zcIt is the weather variable pair obtained from sample data
Answer time between overhaul statistic.Each p value is ranked up after the completion of calculating.In general, p value is less than 0.05 and just illustrates to be somebody's turn to do
Influence of the weather variable to time between overhaul is notable, just more notable if it is less than 0.01.In other words, it can be indicated with 1-p pair
Answer importance values of the weather variable to time between overhaul.Model enforceability and model accuracy in order to balance can select
Related coefficient Correlation is more than 0.5 and weather variable of the importance values (i.e. 1-p) more than 0.8 is used as prevailing weather change
Amount.
Step 3:Historical data based on the prevailing weather variable is trained BP neural network model, and according to
The BP neural network model for completing training predicts the following time between overhaul.
Such as Fig. 2, BP neural network is roughly divided into input layer, hidden layer and output layer.Each layer is being changed using BP neural network
Threshold value and weights method, when establishing distribution mesh element time between overhaul prediction model, become with the prevailing weather of selection
The historical data y of the historical data x of amount and corresponding distribution mesh element time between overhaul is respectively input quantity and output quantity, right
BP neural network is trained, and determines the optimal value of all weight θ of prediction model and threshold alpha.BP neural network is that one kind is answered
With widest artificial neural network and foremost supervised learning technology, there is the good ability for capturing non-linear rule.
The structure of neural network determines its training effect and generalization ability.Therefore, the determination of neural network structure is for nerve net
The foundation of network model is of crucial importance.Theoretically, when the hidden neuron in neural network is enough, three layers
BP neural network can simulate the mapping that arbitrary K dimensions are input to I dimension outputs.
Input layer number is determined by input variable number.In order to avoid input layer input variable is excessive, using Pearson came
Correlation analysis method chooses the weather conditions being affected to power distribution network time between overhaul.Therefore, the present invention is established
Neural network model in the number N of input layer i.e. the number of index key influence factor;Hidden layer numerous studies
Show that the number of hidden neuron is more, has also more easily established more accurate mathematical model.However, the nerve in hidden layer
First quantity determines the time required to completing training process, bigger in the neuronal quantity of this layer of setting, completes training and is taken
Between just increase therewith, and more hidden neuron may also reduce the fault-tolerant ability of model to a certain extent.Therefore,
The number of hidden neuron is needed to combine practical experience or be obtained by multiple test result.The number of hidden neuron can be byIt is calculated, is widely used and obtains good result in many engineering problems, N in formulap
For for trained sample size;K, L, I are input layer, hidden layer and output layer neuron number output layer since the target of prediction is defeated
It is NCHI-F indexs to go out, therefore output layer neuron number is 1, i.e. I=1.
Essential idea according to BP training:According to the error of feedback, internetwork threshold value and weights are updated, until trained
To data and known sample output valve carry out obtained square-error after comparing calculation and reach that threshold value or no longer is determined in advance
Variation.In the present invention, with E come the error of the real output value of expression model and both desired outputs:
Wherein,In formula, dpIt is the original of p-th of sample as a result, ypIt is pth
The reality output result of a sample;Net is the input of output layer neuron;xiIt is the input of i-th of neuron;vijIt is into layer god
The weights of hidden neuron j are transferred to through first i;wijThe weights of output layer neuron are transferred to for hidden neuron j;ψ(x),The respectively transmission function of output layer and hidden layer;θj, α be respectively hidden layer neuron j and output layer neuron threshold value.
Each layer threshold value will determine final error E with weights it can be seen from above-mentioned formula, so the essence of BP neural network
Marrow is to change the threshold value and weights of each layer, to achieve the purpose that improve prediction error precision.It is using gradient descent algorithm
3 layers of BP neural network model can be solved, threshold value is with weighed value adjusting formula:
Δ α in formula,The changes of threshold amount of output layer and weights variable quantity in respectively each iterative process;The variable quantity of hidden layer threshold value and the variable quantity of weights in respectively each iterative process.
The training step of neural network model is as follows:1. completing model real output value and desired output by means of above formula
The calculating of the two error;2. using gradient descent method, Learning Algorithm threshold value and weights are adjusted.3. error in judgement E is
It is no to reach minimum or be less than desired value, if so, terminating;Otherwise step is gone to 2..
The step of providing the power distribution network random fault neural network prediction of meter and Correlative Influence Factors:
1) structure of neural network is determined.
2) data sample is collected, i.e.,:2005~2014 years each weather conditions sequences and element time between overhaul index sequence
Row.
3) sample data is pre-processed.
4) pretreated data being divided into two set of training and test, 2005~2013 annual datas are for training, and 2014
Annual data is for testing.
5) neural network is trained.
6) 2015 annual datas is used to examine the correctness of neural network parameter prediction.
7) it determines the numerical value of each key influence factor in 2015, and is inputted established Neural Network model predictive and obtains
To element time between overhaul index in 2015, which included the fault correction time and failure rate of element.
Table I is the corresponding Mishap Database of each state of weather and the probability of stability at a certain node.Two are actually considered in Table I
Mishap Database under the most important weather variable of kind and the probability of stability, the first weather variable include m kind state of weather section,
Second of weather variable includes s kind state of weather section.Although it should be noted that only account for two kinds of weather variables in Table I,
But the practical different weather state that more kinds of weather variables can be considered.
Table I
Failure rate of the power distribution network element fault probability under all possible state of weather be:
In formula, λiIt is the probability that the power line under i-th of state of weather needs to overhaul every year, PiIt is i-th kind of state of weather
The probability of stability of generation.
In fact, element maintenance number is a part for failure sum under i-th kind of state of weather, therefore have:
Wherein NiHave a power failure for element under i-th kind of state of weather total.
Fault correction time r under i-th kind of state of weatheriFor:
When distribution mesh element needs to overhaul under i-th kind of weather conditions declines situation to embody by the health degree of element, and
The decline situation of health degree can be indicated by failure rate:
It sums based on calculating above, then to various state of weather, so that it may be in various state of weather to obtain the node
In the case of total failare repair time and failure rate.
Step 4:Prediction result based on the following time between overhaul index predicts distribution network reliability.
By the prediction result of distribution mesh element time between overhaul index, the distribution mesh element section time can be obtained:
Element total failare repair time TCOH:TCOH=∑si∈Rri, riFor the total failare repair time of load bus i, R is
The number of load bus;
System System average interruption duration index SAIDI:niFor the number of users of load bus i;
System System average interruption frequency index SAIFI:In formula TCO representation elements major overhaul number and
TCO is obtained according to following method based on failure rate:When the failure rate of element under i-th kind of weather conditionsDrop to
When certain threshold value, representation element needs to overhaul, and element health degree is infinitely close to 1 after maintenance, and after a period of time, element is strong
Kang Du declines again, and when being re-lowered to the threshold value, element needs to overhaul, to the maintenance number under all prevailing weather conditions into
Row counts to get element major overhaul number TCO;
The reliability index of distribution mesh element
By analyzing above, to predict the reliability of power distribution network.
The present invention includes in the power distribution network by taking a certain localized power distribution net of 4.5 megawatts of Nanjing Suburb peak load as an example
5000 circuits have four class nodes in the power distribution network, wherein there is A, B, C, tetra- class nodes of D, A class nodes have 100 users, B
Class node has 150 users, C class nodes to have 100 users, D class nodes to have 220 users.In the power distribution network regional extent
Weather range of variables is as shown in the table, wherein WGS is gust velocity, RS is resultant wind velocity, AS is mean wind speed, NLAL is total
The natural logrithm of thunder-strike current, NLPL are the natural logrithm of peak value thunder-strike current, ICE is ice condition composite index, HT is highest
Temperature, LT are minimum temperature, AT is mean temperature, RH is air humidity.
Prevailing weather variable in the power distribution network regional extent unites to the influence that Nanjing Suburb power distribution network element fault has a power failure
Meter figure is as shown in Figure 3.
Based on the weather conditions in 2015 of Nanjing Suburb shown in Fig. 4, one-shot measurement is carried out per 12h, shares 730 kinds of days
Gaseity.Assuming that first section of WGS and first section of NLPL are state of weather 1, first section of WGS and NLPL
Second section be state of weather 2, five sections of five section NLPL of WGS combine one by one, and so on, the of WGS
Five sections and the 5th section of NLPL are state of weather 25.It can be calculated by prediction model and distribution network reliability
Go out time between overhaul and repair time of the Nanjing Suburb distribution mesh element under different weather state, the result of calculation of gained
As shown in Figure 5.
Using the present invention predict the 1 year distribution network reliability in Nanjing Suburb in 2014, and with electric company provide it is true
Data are compared, wherein SAIDI-F is that power failure duration, SAIDI-P caused by failure are when having a power failure caused by pre-arranged has a power failure
Long, the results are shown in table below:
It can according to a kind of power distribution network considering that weather influences the time between overhaul that power distribution network pre-arranged has a power failure of the present invention
It can be seen that by property prediction technique embodiment and be not much different using the method for the present invention and actual value.
Claims (8)
1. a kind of distribution network reliability prediction technique for considering weather and being influenced on time between overhaul, which is characterized in that including such as
Lower step:
(1) according to the historical data of multiple weather variables, the phase relation of each weather variable and time between overhaul is calculated separately
Importance values several and to time between overhaul;
(2) it is based on each corresponding related coefficient of weather variable and importance values chooses main day from the multiple weather variable
Gas variable;
(3) historical data based on the prevailing weather variable is trained BP neural network model, and is trained according to completion
BP neural network model predict that the following time between overhaul index, the time between overhaul index includes element
Fault correction time and failure rate;
(4) prediction result based on the following time between overhaul index predicts distribution network reliability.
2. distribution network reliability prediction technique according to claim 1, which is characterized in that the multiple in step (1)
Weather variable includes:Wind variable, frost state quantities, temperature variables, the relative humidity variable of air and lightning stroke variable;Wherein
The frost state quantities are calculated by the following formula:
Wherein, LNDRSc be the current value of maximum number of days of sleet assessment, the current value that TPSc is rainfall and snowfall,
LNDCLTc is the current value of continuous low temperature number of days, and LNDRSav is that average value, the TPSav of the maximum number of days of sleet assessment are rainfall
Average value and LNDCLTav with snowfall are the average value of continuous low temperature number of days.
3. distribution network reliability prediction technique according to claim 1, which is characterized in that in step (1), by following
Formula calculates the related coefficient Correlation of each weather variable and time between overhaul:
Wherein, x is the probability of certain year a certain weather occurrences, and y is the decline of this year element health degree under the weather variable
Speed,For the probability average of the weather occurrences of each year,For the decline of element health degree under the weather variable of each year
The average value of speed.
4. distribution network reliability prediction technique according to claim 1, which is characterized in that described important in step (1)
Property value is calculated by P value methods.
5. distribution network reliability prediction technique according to claim 1, which is characterized in that in step (1), by phase relation
More than 0.5 and when importance values are more than 0.8, corresponding weather variable is selected as prevailing weather variable to number.
6. distribution network reliability prediction technique according to claim 1, which is characterized in that step (3) further comprises:
(3.1) using the historical data of the prevailing weather variable as the input of BP neural network model come when calculating maintenance interval
Between predicted value;
(3.2) predicted value of the time between overhaul of calculating and actual value are compared to calculate error, and according to feedback
Error is trained the BP neural network model, and the training includes the error based on feedback to update BP neural network mould
Threshold value in type and weights;
(3.3) step (3.1) and (3.2) are repeated until training the predicted value of obtained time between overhaul and known prediction
Square-error between value reach or less than setting threshold value, to complete training to BP neural network model;
(3.4) the following time between overhaul index is predicted based on the BP neural network model for completing training.
7. distribution network reliability prediction technique according to claim 6, which is characterized in that in step (3.2), using ladder
Descent method is spent to update the threshold value and weights in BP neural network model.
8. distribution network reliability prediction technique according to claim 1, which is characterized in that step (4) includes the following steps:
(4.1) weather condition for assuming certain following a period of time, is referred to using the following distribution mesh element time between overhaul of prediction
In mark calculating distribution mesh element 1 year:
Element total failare repair time TCOH:TCOH=∑si∈Rri;
System System average interruption duration index SAIDI:
System System average interruption frequency index SAIFI:
The reliability index ASAI of distribution mesh element:
Wherein, riFor the fault correction time of load bus i, niFor the number of users of load bus i, R is of load bus
Number, TCO is element major overhaul number and TCO is that the failure rate based on element obtains;
(4.2) system System average interruption duration index SAIDI, system System average interruption frequency index SAIFI and power distribution network are based on
The reliability index ASAI of element predicts the reliability of power distribution network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810758399.8A CN108694479A (en) | 2018-07-11 | 2018-07-11 | Consider the distribution network reliability prediction technique that weather influences time between overhaul |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810758399.8A CN108694479A (en) | 2018-07-11 | 2018-07-11 | Consider the distribution network reliability prediction technique that weather influences time between overhaul |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108694479A true CN108694479A (en) | 2018-10-23 |
Family
ID=63850486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810758399.8A Pending CN108694479A (en) | 2018-07-11 | 2018-07-11 | Consider the distribution network reliability prediction technique that weather influences time between overhaul |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108694479A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110232510A (en) * | 2019-06-03 | 2019-09-13 | 国网能源研究院有限公司 | It is a kind of meter and bad weather condition electric system chance repair determining method |
| CN110414718A (en) * | 2019-07-04 | 2019-11-05 | 上海工程技术大学 | A kind of distribution network reliability index optimization method under deep learning |
| CN110443503A (en) * | 2019-08-07 | 2019-11-12 | 成都九鼎瑞信科技股份有限公司 | The training method and related system of water utilities system industrial gross output value analysis model |
| CN115271569A (en) * | 2022-10-08 | 2022-11-01 | 合肥优晟电力科技有限公司 | Cross-regional power grid safety operation and maintenance monitoring and analysis method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573340A (en) * | 2014-12-25 | 2015-04-29 | 国家电网公司 | Method for calculating time-variable reliability of power transmission line in freezing weather |
-
2018
- 2018-07-11 CN CN201810758399.8A patent/CN108694479A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104573340A (en) * | 2014-12-25 | 2015-04-29 | 国家电网公司 | Method for calculating time-variable reliability of power transmission line in freezing weather |
Non-Patent Citations (2)
| Title |
|---|
| YINGJUN WU等: "An Outage Risk Oriented Dynamic Distribution Network Reconfiguration Methodology Considering the Effects of Weather Conditions on Power Line Failure Rate", 《ELECTRIC POWER COMPONENTS AND SYSTEMS》 * |
| 田洪迅等: "基于BP神经网络的配电网可靠性关联因素灵敏度计算方法", 《电力系统保护与控制》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110232510A (en) * | 2019-06-03 | 2019-09-13 | 国网能源研究院有限公司 | It is a kind of meter and bad weather condition electric system chance repair determining method |
| CN110232510B (en) * | 2019-06-03 | 2021-06-15 | 国网能源研究院有限公司 | Power system opportunity maintenance decision method considering severe meteorological conditions |
| CN110414718A (en) * | 2019-07-04 | 2019-11-05 | 上海工程技术大学 | A kind of distribution network reliability index optimization method under deep learning |
| CN110443503A (en) * | 2019-08-07 | 2019-11-12 | 成都九鼎瑞信科技股份有限公司 | The training method and related system of water utilities system industrial gross output value analysis model |
| CN115271569A (en) * | 2022-10-08 | 2022-11-01 | 合肥优晟电力科技有限公司 | Cross-regional power grid safety operation and maintenance monitoring and analysis method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | Predicting electricity consumption in a building using an optimized back-propagation and Levenberg–Marquardt back-propagation neural network: Case study of a shopping mall in China | |
| CN108694479A (en) | Consider the distribution network reliability prediction technique that weather influences time between overhaul | |
| CN102722759B (en) | Method for predicting power supply reliability of power grid based on BP neural network | |
| Brusca et al. | A new design methodology to predict wind farm energy production by means of a spiking neural network–based system | |
| CN104484715A (en) | Neural network and particle swarm optimization algorithm-based building energy consumption predicting method | |
| CN104715292A (en) | City short-term water consumption prediction method based on least square support vector machine model | |
| CN110766212A (en) | Ultra-short-term photovoltaic power prediction method for historical data missing electric field | |
| CN102880905B (en) | Online soft sensor method done by a kind of normal top oil | |
| CN108596449A (en) | It is a kind of to consider distribution network reliability prediction technique of the weather to distribution network failure impact probability | |
| CN112101669B (en) | Photovoltaic power interval prediction method based on improved extreme learning machine and quantile regression | |
| CN109492792A (en) | A method of it is predicted based on particle group optimizing wavelet neural network powerline ice-covering | |
| CN108388962A (en) | A kind of wind power forecasting system and method | |
| CN106599417A (en) | Method for identifying urban power grid feeder load based on artificial neural network | |
| CN104036328A (en) | Self-adaptive wind power prediction system and prediction method | |
| CN108038568A (en) | A kind of changeable weight combination Short-Term Load Forecasting of Electric Power System based on particle cluster algorithm | |
| CN104820877A (en) | Photovoltaic system generation power prediction method based on cloud adaptive PSO-SNN | |
| CN110210670A (en) | A kind of prediction technique based on power-system short-term load | |
| CN110070223A (en) | A kind of short term power prediction technique applied to newly-built wind power plant | |
| CN113988655A (en) | Power transmission line running state evaluation method considering multiple meteorological factors | |
| CN114742283A (en) | Medium-voltage line loop closing current estimation and risk assessment method based on deep learning network | |
| CN113516271A (en) | Wind power cluster power day-ahead prediction method based on space-time neural network | |
| CN115392697A (en) | Data-model hybrid driven power system safety assessment method and system | |
| CN105205572A (en) | Photovoltaic power generation output forecasting method based on EMD (Empirical Mode Decomposition) and Elman algorithm | |
| CN108898245A (en) | A kind of needing forecasting method for Wind turbines components | |
| CN109948925A (en) | Consider the power communication system reliability estimation method that weather influences |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181023 |