CN109993361B - PMU-based power distribution network operation trend prediction method - Google Patents
PMU-based power distribution network operation trend prediction method Download PDFInfo
- Publication number
- CN109993361B CN109993361B CN201910251522.1A CN201910251522A CN109993361B CN 109993361 B CN109993361 B CN 109993361B CN 201910251522 A CN201910251522 A CN 201910251522A CN 109993361 B CN109993361 B CN 109993361B
- Authority
- CN
- China
- Prior art keywords
- measurement data
- sequence
- prediction
- distribution network
- pmu
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000005259 measurement Methods 0.000 claims abstract description 92
- 238000013528 artificial neural network Methods 0.000 claims abstract description 48
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 230000004927 fusion Effects 0.000 claims abstract description 7
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000012937 correction Methods 0.000 claims description 13
- 238000009499 grossing Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 8
- 238000002347 injection Methods 0.000 claims description 7
- 239000007924 injection Substances 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 230000007547 defect Effects 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000011161 development Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 101100391182 Dictyostelium discoideum forI gene Proteins 0.000 claims description 3
- 230000009191 jumping Effects 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 230000002087 whitening effect Effects 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims 2
- 238000010586 diagram Methods 0.000 description 8
- 238000007689 inspection Methods 0.000 description 4
- 238000006467 substitution reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000016507 interphase Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
-
- 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
- Y04S10/22—Flexible AC transmission systems [FACTS] or power factor or reactive power compensating or correcting units
-
- 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)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Power Engineering (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to a PMU-based power distribution network operation trend prediction method, which comprises the following steps: s1: acquiring measurement data of a power distribution network including PMU measurement data, SCADA data and marketing data, establishing a power distribution network operation measurement database, and preprocessing bad measurement data; s2: establishing a multidimensional gray-neural network hybrid coordination prediction model, taking measurement data of a power distribution network as input data, and predicting the output condition of each node of a system at the next moment; s3: establishing a forward push back power flow algorithm of the branch current of the fusion PMU, and calculating the voltage at each prediction moment for the prediction result of the step S2 by using the algorithm to obtain a power flow result; s4: and obtaining a power flow result at each prediction moment, and realizing PMU-based power distribution network operation trend prediction. Compared with the prior art, the method has the advantages of more accurate reflection of the operation trend of the power distribution network, high prediction precision, high convergence speed and the like.
Description
Technical Field
The invention relates to a power distribution network operation trend prediction method, in particular to a PMU-based power distribution network operation trend prediction method.
Background
With the continuous increase of power demand in China and the increasingly serious problems of energy shortage, environmental pollution and the like in social and economic development, large-scale distributed power supply, electric automobile access and multi-user supply and demand interaction become typical characteristics of a power distribution network. However, the randomness, intermittence, load diversification and interactivity of the distributed power supply can cause the running state of the power distribution network to be complex and changeable, and the accurate depiction of the running state and trend change of the power distribution network is an important work for ensuring the safe and stable running of the power distribution network. In view of the fact that the application of the synchronous phasor measurement unit (Phasor Measurement Unit, PMU) can provide real-time operation data for describing the operation state of the power distribution network, popularization and application of the synchronous phasor measurement unit in the power distribution network are possible. The method is based on the fact that compared with a power transmission network, the power distribution network has the problems that the structure is complex, data acquisition is relatively difficult and the like, and therefore the high-density real-time operation trend of the power distribution network cannot be timely described by using a traditional measuring technology. In this regard, it is highly desirable to provide a suitable power distribution network operation trend prediction method, accurately analyze future operation trends of the power distribution network, and further improve safe and reliable operation levels of the power distribution network. Most of the prior art does not consider that the prediction precision is affected by various factors, and the operation trend of the power distribution network cannot be accurately reflected. Therefore, constructing a power distribution network operation trend prediction method under complex measurement conditions becomes a problem to be considered.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a PMU-based power distribution network operation trend prediction method.
The aim of the invention can be achieved by the following technical scheme:
the invention discloses a PMU-based power distribution network operation trend prediction method, which is used for acquiring accurate power distribution network operation trends to realize effective maintenance and construction of a power distribution network, and comprises the following steps:
step one, PMU measurement data, SCADA data and marketing data of the power distribution network are obtained, a power distribution network operation measurement database is built, and bad measurement data is preprocessed.
The PMU measurement data are actually measured data, and are directly applied to the second step and the third step without considering measurement errors, and load data in the measurement data often have errors such as defects, jumps and the like, so that the prediction accuracy is reduced, and a predicted load curve is deformed. The invention preprocesses the historical data from the following two aspects to improve the prediction precision.
First assume that 24-point load on a day is:
L＝{l(1),l(2),…l(24)}
1) The following determination is made for the measured data defect:
l (h) -l (h-1) | < λ (h=2, 3, …,24, λ is a constant)
If the difference between any two adjacent load measurement data in the day is smaller than lambda, judging the load measurement data in the day as abnormal data, and deleting the corresponding measurement data. If the load measurement data for any two hours and more on the day is less than lambda, the process can be performed as follows:
for holidays: the load measurement data has the characteristics of periodicity and annual similarity. According to the characteristics, the load measurement data of the day of the last year is found to be used as reference load L (y-1, d), the closest load measurement data is found before the correction date, the difference delta between the two is used as the annual increment of the load measurement data, and the expression of the load measurement data correction value L (y, d) corresponding to the day to be corrected is as follows:
L(y,d)＝L(y-1,d)+Δ
for the common day: according to the load day similarity principle, seeking normal operation data before an abnormal day, and processing according to the following formula:
in the formula ,PR (h, t) is the h normal measurement value, P R (f, t) is the f-th normal measurement value,
is the average value of f-1 normal measurement values, < >>
The average value of K normal points is the Nth normal measurement value.
2) The load measurement data jump is processed as follows:
if the load measurement data between any two points on the day satisfies the following equation:
the load measurement data may be determined to be a skip data change. And correcting the load measurement data by using a right-neighbor ratio generating method for the head-end load and a left-neighbor ratio generating method for the tail-end load, and taking the average value of the load measurement data by using the middle load. The expression is as follows:
l(1)＝[l(2)] 2 /l(3)
l(h)＝0.5×[l(h-1)+l(h+1)]h＝(2,3,…23)
l(24)＝[l(23)] 2 /l(22)
and step two, taking the measurement data of the power distribution network provided in the step one as input data, establishing a multi-dimensional gray-neural network hybrid coordination prediction model, and predicting the output condition of each node of the system at the next moment.
The multidimensional gray-neural network hybrid coordination prediction model is a multidimensional gray prediction model established by utilizing uncertainty and gray among factors. The specific process of reading the data of the operation measurement database of the power distribution network and inputting the multi-dimensional gray-neural network hybrid coordination prediction model for prediction is as follows:
the system is provided with the following characteristic sequences:
wherein ,
as a feature element, the relevant multifactor sequence is:
……
wherein ,
is->
G=1, 2, …, N, then +.>
Is->
The immediately adjacent mean sequence of (2) is: />
The GM (1, n) model is obtained as follows:
wherein ,
for gray derivative->
Is background value, -a is development coefficient, +.>
For driving terms, the multidimensional gray GM (1, n) model specifically calculates the following steps:
11): for a pair of
Generates +.>
wherein ：
12): for a pair of
And (5) performing quasi-smoothness inspection and exponential law inspection. The accuracy of gray predictions relates to the smoothness of the original sequence, the better the smoothness, the higher the accuracy of the predictions:
wherein ρ (k) is the sequence
Is a smooth ratio of (c). If->
The sequence is as follows:
ρ(k)∈[0,0.5],k＝2,3,…,n；
then call for
For a quasi-smooth sequence, gray modeling can be directly performed, otherwise, the original sequence needs to be preprocessed.
If k >3 and satisfies the following equation:
then it is determined that
Meets the rule of index.
13): from the following components
Generating sequence next to the mean->
And the gray scale parameters are calculated.
wherein ：
is provided with
Is a parameter column, and:
14): and determining a model, solving, and obtaining a predicted value through reducing and reducing.
solving for
Is a simulation of the values of:
when (when)
When the amplitude change is small, the patient is->
The approximate corresponding time for the ash constant GM (1, n) is:
wherein ,
get->
Reduction determination
Is a predicted value of (1):
the improved multidimensional gray model provided by the invention introduces a smoothing control parameter c into GM (1, N), ensures that an original sequence meets exponential smoothing, and increases prediction accuracy positively. The smoothing conditions of GM (1, n) may be limited during prediction:
where k=4, 5, …, n, adding the original signature sequence addition factor c to GM (1, n) such that the new sequence satisfies the following equation:
from the above formula, it can be found that
The appropriate c must be found to satisfy the above formula, expressed in detail as follows: />
Further, the following steps are obtained:
from the above formula, the appropriate c-addition sequence can be selected. In addition, for sequences to which the appropriate influencing factor c has been added
The ash differential equation can be established as:
estimating parameters by using a least square method to obtain:
[b 1 ,b2,…b n ,μ] T ＝(B T B) -1 B T Y H
wherein ：
under the initial conditions
The prediction of the accumulated sequence can be obtained:
pair sequence
The predicted reduction values of (2) are:
then finally pair the sequences
The predictive representation of each item of (2) is:>
and the residual prediction model based on the BP neural network is corrected by utilizing the residual of the BP neural network system, so that the prediction precision is improved. Taking the residual error of the multivariable gray prediction as the input quantity of the BP neural network, predicting the residual error, and making a difference between the input quantity of the BP neural network and the predicted value to obtain the residual value of the residual error, wherein the process is circulated twice and comprises the following steps:
21 The power sequence to be predicted is represented as S, the preprocessed load measurement data is taken as a characteristic sequence, voltage and current are taken as influencing factors, smoothness of the characteristic sequence is considered, a smoothing factor is introduced, a GM (1, N) model is determined, a predicted value is obtained through accumulation and subtraction, and a predicted result is represented as S GM . Calculate the sequences S and S GM The difference is the residual epsilon, where epsilon=s-S GM . Predicting epsilon sequence by BP neural network, and the predicted result sequence is expressed as epsilon BP 。
22 Calculating the predicted result sequence epsilon of the residual sequence epsilon and the BP neural network pair epsilon BP Is epsilon in the difference sequence 2 Having epsilon 2 ＝ε-ε BP For epsilon 2 BP neural network prediction is performed, and the predicted result is expressed as epsilon 2BP 。
23 Adding the predicted results of step 21) and step 22) to obtain a final predicted result sequence S GM1 ＝S GM +ε BP +ε 2BP 。
The multi-dimensional gray-neural network hybrid coordination prediction model is a combination form of a multi-dimensional gray prediction model and a residual prediction model based on a BP neural network, and can improve prediction precision.
And thirdly, establishing a forward push back power flow algorithm for merging the branch currents of the PMU, carrying out power flow calculation on the prediction result of the second step by using the algorithm to obtain the voltage at each prediction moment, and then realizing the prediction of the operation trend of the power distribution network based on the PMU. The forward push back power flow algorithm of the branch current utilizes the branch current as a forward push variable, and branch power loss does not need to be considered, so that calculation is simpler and more convenient.
First, network topology relation is to be determined: and determining the relation of the search nodes and analyzing the topological structure. In order to match the algorithm and avoid complex network numbering, the following original data input structure is adopted, and the node relation can be automatically searched to determine the network structure without forming a node admittance matrix.
And secondly, according to the first node and the last node of the branch, acquiring the connection relation of any node, and further forming an integral tree-shaped relation structure. Meanwhile, the network topology structure is fully utilized, hierarchical relations are formed through breadth-first search of multi-time layer-by-layer traversal, and node calculation sequences of the forward-backward generation tide algorithm are screened. The method comprises the following specific steps:
a) Searching for an end node as a first layer node;
b) Searching a parent node of the end node as a second-layer node;
c) Continuing to search the father node of the second layer node as the third layer node, repeatedly searching until all the father nodes of a certain layer node are root nodes, and stopping searching;
d) And deleting the nodes in the previous hierarchy, which are repeated in the subsequent hierarchy, forming a real hierarchy relationship, and determining the sequence of the power flow calculation nodes.
The forward push back power flow algorithm based on the branch current firstly presumes that the voltage amplitude of each node is 1 and the phase angle is 0, and the branch current forward push back power flow algorithm fused with the PMU utilizes the PMU as a device for directly measuring the voltage phasor and the current phasor of the node, so that the measurement data of the system are greatly enriched, and the method comprises the following specific steps:
step 1: starting from the layer node, obtaining branch current according to kirchhoff current law;
in the formula ,
phase current from node i to node j; />
The j-point phase voltage; />
The j point potential is at power; p (P) j 、Q j Active power and reactive power are injected for point j. />
Step 2: gradually calculating the injection current of the non-end node from the second layer, wherein the injection current is equal to the sum of the current flowing out of the node according to kirchhoff current law;
in the formula ,
phase current from node i to node j; m is all branches directly connected with the node j, I jT Is the T-th branch current connected to node j.
Step 3: step 1, step 2 can calculate branch currents of all branches, and then utilize known root node voltage to sequentially calculate each load node voltage from the root node backwards;
step 4: calculating the voltage amplitude correction quantity and the maximum voltage amplitude correction quantity of each node;
step 5: judging convergence conditions:
wherein Λ is iteration times, if the maximum voltage amplitude correction is smaller than a threshold value psi, jumping out of the loop, and outputting a voltage calculation result; otherwise, repeating the steps 1 to 5 until the conditions of the above formula are satisfied;
step 6: after the voltage and current of each node are obtained, the potential of each node at the power can be calculated
As a result of tidal currents.
The invention considers a tide calculation method for fusing PMU collected data under the background of completing PMU optimization distribution. In view of the characteristics of high sampling frequency, high measured data density and the like of PMUs, the measured data needs to be detected, identified, screened and the like, and adverse effects on the prediction of the running trend of the power distribution network caused by the use of error data are avoided. In addition, different voltage classes have certain requirements on voltage errors, and different generators have certain requirements on current errors. Preferably, distributed power sources (Distributed Generation, DG) may be connected to the distribution network such that the system power flow is changed from unidirectional flow to bidirectional flow.
And fourthly, analyzing a power flow result at each prediction moment to obtain a power distribution network operation trend prediction result containing PMU. And respectively calculating multi-dimensional gray-neural network mixed coordination prediction results at different prediction moments, and obtaining voltage and current at different prediction moments and power distribution network operation trend results by utilizing a forward push back power flow algorithm of the branch current of the fusion PMU. After the predicted running trend result of the power distribution network is obtained, the reliability index of the power distribution network can be calculated to adjust the current state of the power distribution network, or a rush-repair work order is established, so that the power distribution network can be effectively overhauled, and the like.
Compared with the prior art, the invention has the following advantages:
1. according to the invention, a power distribution network operation measurement database is established by combining power distribution network PMU measurement data, SCADA data, marketing data and the like, a multidimensional gray-neural network mixed coordination prediction model is established, the output condition of each node of a system at the next moment is predicted, the power distribution network flow algorithm fused with PMU is adopted to calculate the system flow at each prediction moment on the basis, the operation trend of the power distribution network in the future state is further obtained, and the operation trend of the power distribution network is reflected more accurately by considering various factors of prediction influence precision;
2. the invention predicts the output condition of each node of the system at the next moment by utilizing a multi-dimensional gray-neural network mixed coordination prediction model, combines the multi-dimensional gray prediction model and a residual prediction model based on a BP neural network, limits the smoothing condition of GM (1, N) in the prediction process by introducing a smoothing control parameter c into GM (1, N) in the multi-dimensional gray prediction model, and corrects the residual prediction model of the BP neural network by utilizing the residual error of the BP neural network system, thereby further improving the prediction precision;
3. according to the method, the power flow algorithm of the power distribution network fused with the PMU is adopted to calculate the system power flow at each predicted moment, and the PMU is added to reduce iteration times and improve convergence speed.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of a prediction flow using a multi-dimensional gray-neural network hybrid coordinated prediction model in the method of the present invention;
FIG. 3 is a schematic flow chart of a forward push back power flow algorithm for merging branch currents of PMUs in the method of the present invention;
FIG. 4 is a diagram of an IEEE33 system with 23 and 29 nodes connected to a photovoltaic system in an embodiment of the invention;
FIG. 5 is a graph of active force prediction at nodes 22 and 24 hours according to an embodiment of the present invention, wherein FIG. 5 (a) is a graph of the prediction result of a multidimensional gray prediction model, and FIG. 5 (b) is a graph of the prediction result of a multidimensional gray-neural network hybrid coordination prediction model;
FIG. 6 is a diagram of the result of the IEEE33 system power flow using the forward-push-back replacement power flow algorithm of the branch current with the integrated PMU according to the embodiment of the present invention, wherein FIG. 6 (a) is a diagram of the result of the IEEE33 system power flow with the PMU, and FIG. 6 (b) is a diagram of the result of the IEEE33 system power flow without the PMU;
FIG. 7 is a diagram of the result of the IEEE33 system power flow using the branch current forward push back power flow algorithm with PMU fusion under DG access conditions, where FIG. 7 (a) is a diagram of the result of the IEEE33 system power flow with PMU, and FIG. 7 (b) is a diagram of the result of the IEEE33 system power flow without PMU;
fig. 8 is a graph showing voltage variation trend of node 22 in future days according to an embodiment of the present invention.
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
The invention relates to a PMU-based power distribution network operation trend prediction method, which comprises the following steps:
1. and acquiring PMU measurement data, SCADA data and marketing data of the power distribution network, and establishing a power distribution network operation measurement database and preprocessing bad measurement data.
2. And (3) taking the measurement data of the power distribution network provided in the step one as input data, establishing a multi-dimensional gray-neural network hybrid coordination prediction model, and predicting the output condition of each node of the system at the next moment.
The multidimensional gray-neural network hybrid coordination prediction model is a multidimensional gray prediction model established by utilizing uncertainty and gray among factors. The specific flow of the data of the running measurement database of the power distribution network is read, and the data is input into the multi-dimensional gray-neural network hybrid coordination prediction model for prediction, as shown in figure 2.
The system is provided with the following characteristic sequences:
……
wherein ,
is->
G=1, 2, …, N, then +.>
Is->
The immediately adjacent mean sequence of (2) is:
the GM (1, n) model is obtained as follows:
wherein ,
for gray derivative->
Is background value, -a is development coefficient, +.>
For driving terms, the multidimensional gray GM (1, n) model specifically calculates the following steps:
11): for a pair of
Generates +.>
wherein ：
12): for a pair of
And (5) performing quasi-smoothness inspection and exponential law inspection. The accuracy of gray predictions relates to the smoothness of the original sequence, the better the smoothness, the higher the accuracy of the predictions:
wherein ρ (k) is the sequence
Is a smooth ratio of (c). If->
The sequence is as follows:
ρ(k)∈[0,0.5],k＝2,3,…,n；
then call for
For quasi-smooth sequences, grey modeling can be performed directly, otherwise the original sequence needs to be preprocessed, if k>3 satisfies the following formula:
if the above is true, then determine
Meets the rule of index.
13): from the following components
Generating sequence next to the mean->
And the gray scale parameters are calculated.
wherein ：
is provided with
Is a parameter column, and:
14): and determining a model, solving, and obtaining a predicted value through reducing and reducing.
solving for
Is a simulation of the values of:
when (when)
When the amplitude change is small, the patient is->
The approximate corresponding time for the ash constant GM (1, n) is:
wherein ,
get->
Reduction determination
Is a predicted value of (1):
the improved multidimensional gray model provided by the invention introduces a smoothing control parameter c into GM (1, N), ensures that an original sequence meets exponential smoothing, and increases prediction accuracy positively. The smoothing conditions of GM (1, n) may be limited during prediction:
where k=4, 5, …, n, adding the original signature sequence addition factor c to GM (1, n) such that the new sequence satisfies the following equation:
from the above formula, it can be found that
The appropriate c must be found to satisfy the above formula, expressed in detail as follows:
further, the following steps are obtained:
from the above formula, the appropriate c-addition sequence can be selected. In addition, for sequences to which the appropriate influencing factor c has been added
An ash differential equation may be established such as:
estimating parameters by using a least square method to obtain:
[b 1 ,b2,…b n ,μ] T ＝(B T B) -1 B T Y H
wherein ：
under the initial conditions
Performing accumulation sequenceThe prediction can be obtained: />
Pair sequence
The predicted reduction values of (2) are:
then finally pair the sequences
The predictions for each term of (a) are expressed as:
and the residual prediction model based on the BP neural network is corrected by utilizing the residual of the BP neural network system, so that the prediction precision is improved. Taking the residual error of the multivariable gray prediction as the input quantity of the BP neural network, predicting the residual error, and making a difference between the input quantity of the BP neural network and the predicted value to obtain the residual value of the residual error, wherein the process is circulated twice and comprises the following steps:
1) The power sequence to be predicted is represented as S, the preprocessed load measurement data is taken as a characteristic sequence, voltage and current are taken as influencing factors, smoothness of the characteristic sequence is considered, a smoothing factor is introduced, a GM (1, N) model is determined, a predicted value is obtained through accumulation and subtraction, and the predicted result is represented as S GM . Calculate the sequences S and S GM The difference is the residual epsilon, where epsilon=s-S GM . Predicting epsilon sequence by BP neural network, and the predicted result sequence is expressed as epsilon BP 。
2) Calculating the predicted result sequence epsilon of the residual sequence epsilon and the BP neural network pair epsilon BP Is epsilon in the difference sequence 2 Having epsilon 2 ＝ε-ε BP For epsilon 2 BP neural network prediction is performed, and the predicted result is expressed as epsilon 2BP 。
3) Adding the prediction results of the step 1) and the step 2) to obtain a final prediction result sequence S GM1 ＝S GM +ε BP +ε 2BP 。
The multi-dimensional gray-neural network hybrid coordination prediction model is a combination form of a multi-dimensional gray prediction model and a residual prediction model based on a BP neural network, and can improve prediction precision.
3. And (3) establishing a forward push back power flow algorithm for merging the branch currents of the PMU, carrying out power flow calculation on the prediction result of the step two by using the algorithm to obtain the voltage of each prediction moment, and then realizing the prediction of the operation trend of the power distribution network based on the PMU.
The forward push back power flow algorithm based on the branch current firstly presumes that the voltage amplitude of each node is 1 and the phase angle is 0, and the branch current forward push back power flow algorithm fused with the PMU utilizes the PMU as a device for directly measuring the voltage phasor and the current phasor of the node, so that the measurement data of the system are greatly enriched, and the flow is shown in a figure 3, and the specific steps are as follows:
1) Starting from the layer node, obtaining branch current according to kirchhoff current law;
in the formula ,
phase current from node i to node j; />
The j-point phase voltage; />
The j point potential is at power; p (P) j 、Q j Active power and reactive power are injected for point j.
2) Gradually calculating the injection current of the non-end node from the second layer, wherein the injection current is equal to the sum of the current flowing out of the node according to kirchhoff current law;
in the formula ,
phase current from node i to node j; m is all branches directly connected with the node j, I jT Is the T-th branch current connected to node j.
3) Branch currents of all branches can be obtained in the first step and the second step, and then the known root node voltage is utilized to sequentially obtain the load node voltages from the root node to the back;
4) Calculating the voltage amplitude correction quantity and the maximum voltage amplitude correction quantity of each node;
5) Judging convergence conditions:
wherein Λ is the iteration number, if the maximum voltage correction is smaller than the threshold value psi, jumping out of the loop, and outputting a voltage calculation result; otherwise, repeating the steps 1 to 5 until the conditions of the above formula are satisfied;
6) After the voltage and current of each node are obtained,the potential of each node at power can be calculated
As a result of tidal currents.
The patent considers the load flow calculation method for fusing the PMU collected data under the background of completing PMU optimization distribution. In view of the characteristics of high sampling frequency, high measured data density and the like of PMUs, the measured data needs to be detected, identified, screened and the like, and adverse effects on the prediction of the running trend of the power distribution network caused by the use of error data are avoided. In addition, different voltage classes have certain requirements on voltage errors, and different generators have certain requirements on current errors. Taking a 10kV voltage class and a three-phase asynchronous generator as an example, the voltage deviation of the 10kV voltage class and the three-phase asynchronous generator is +/-7%, the inter-phase current deviation is 5%, and the measurement error of 5% of the PMU measurement data is comprehensively considered in the invention and is applied to the analysis of the embodiment.
The distributed power supplies (Distributed Generation, DG) are connected into the power distribution network, so that system power flow is changed from unidirectional flow to bidirectional flow, and in order to verify the influence of PMU addition on the power flow of the power distribution network containing DG, the invention provides a branch current forward push back power flow algorithm containing DG fusion PMU, and analyzes the influence of PMU errors on the DG branch current forward push back power flow result.
4. And analyzing the power flow result at each prediction moment to obtain a power distribution network operation trend prediction result containing PMU. And respectively calculating multi-dimensional gray-neural network mixed coordination prediction results at different prediction moments, and obtaining voltage and current at different prediction moments and power distribution network operation trend results by utilizing a forward push back power flow algorithm of the branch current of the fusion PMU. After the predicted running trend result of the power distribution network is obtained, the reliability index of the power distribution network can be calculated to adjust the current state of the power distribution network, or a rush-repair work order is established, so that the power distribution network can be effectively overhauled, and the like.
In the embodiment of the invention, an IEEE33 system is taken as an example, 12.66kV is taken as a reference voltage (as shown in figure 4), and the operation trend prediction research of the power distribution network based on PMU is carried out. According to the PMU measurement data, the SCADA data and the marketing data and the like, the PMU is distributed at the node 5 and the node 20, and under the condition of not considering errors, the measured voltage of the node 5PMU is 11.59kV, and the measured voltage of the node 20PMU is 12.56kV. Taking PMU measurement and historical voltage and current of each node as influence factors, calculating the system output condition at the future moment by using a multidimensional gray-neural network model, and plotting the output condition at the node 22 as shown in FIG. 5. Table 1 shows the relative errors at different predicted times for node 22.
Table 1 relative error at different predicted times for node 22
Fig. 5 (a) shows a prediction result by using only a multidimensional gray model, the residual error is not repaired, fig. 5 (b) shows a multidimensional gray-neural network hybrid coordination prediction result, and the residual error of the multidimensional prediction result is secondarily corrected by using a neural network, so that the prediction accuracy is improved. The multi-dimensional gray-neural network hybrid coordination prediction result can be seen to well show the change trend of the load along with time. Under the condition of considering multiple influencing factors, the residual error is corrected by using the BP neural network, and the residual error is closer to a true value, so that data support is provided for load flow calculation of a subsequent system.
And D, taking the multidimensional gray-neural network mixed coordination prediction result as the input of the analysis result in the step four, directly substituting current data in PMU measurement data into a power flow calculation process, substituting PMU measurement voltage data into a power flow result, and obtaining the system power flow at the prediction moment. Fig. 6 is a current result of a branch current forward-push substitution method of the power distribution network in which PMUs are integrated, in which fig. 6 (a) calculates a current calculation result of the branch current forward-push substitution method when the PMUs are added and not, considering whether the PMUs are connected to the system current, on the premise of not considering the PMU error, it can be seen that after the PMUs are added, the measured value replaces the iteration value, the iteration number and the iteration time are reduced, and the convergence rate is improved, as shown in table 2:
TABLE 2 comparison of the number of iterations of the flow calculation and the convergence time with and without PMU addition
| Category(s) | Iteration times (times) | Convergence time(s) |
| Without addition of |
3 | 0.054881 |
| Adding |
2 | 0.035376 |
When the PMU measurement error is considered to be 5%, namely +5% is considered at a low voltage point and-5% is considered at a high voltage point, the actual voltage of the node 5 is 11.038kV, the voltage of the node 20 is 13.2211kV, the voltage measurement is directly substituted into a power flow result and an iteration process to obtain the power flow result of the graph (b), the iteration times are unchanged, the convergence time is not obviously improved, and therefore the influence of the error of 5% on the system power flow is considered to be larger. The iteration number and convergence time of the power flow calculation under the condition of considering the PMU measurement error are shown in table 3:
table 3 number of iterations and convergence time considering PMU measurement errors
| Category(s) | Iteration times (times) | Convergence time(s) |
| Without addition of |
3 | 0.054881 |
| Adding |
3 | 0.052716 |
When PMU measurement errors are not considered, constant-power photovoltaic units are connected to the node 23 and the node 29, and the system power flow situation is obtained as shown in fig. 7. Fig. 7 (a) shows that the PMU access effect on the system power flow without considering the measurement error, and the test result is similar to that when DG is not added, the addition of PMU reduces the iteration number and increases the convergence rate. Fig. 7 (b) shows that when the measurement error is considered, the PMU measurement error affects the system power flow, and the test result is similar to that when DG is not added, the iteration number is unchanged, and the convergence time is not obviously improved.
In summary, the embodiment of the invention adopts a multidimensional gray-neural network hybrid coordination prediction model, combines PMU measurement data and historical data, predicts the output condition of each node at the next moment of the power distribution network, and carries out the calculation of the forward-push substitution trend of the branch current fused with the PMU on the future state of the power distribution network on the basis of the prediction to obtain the running trend of the power distribution network voltage. FIG. 8 depicts the trend of the voltage at node 22 over the next 24 hours; fig. 5 (a) depicts the trend of the active power of node 22 over the next 24 hours.
According to the method, the fact that the prediction accuracy is influenced by various factors is considered, the predicted operation trend result of the power distribution network can accurately reflect the operation trend of the power distribution network, and according to the predicted result of the method, the reliability index of the power distribution network can be calculated to adjust the current state of the power distribution network, or a rush repair work order is built, measures such as effective overhaul of the power distribution network are achieved, and the overall construction of the power distribution network is facilitated.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions may be made without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (5)
1. A PMU-based power distribution network operation trend prediction method is characterized by comprising the following steps:
s1: acquiring measurement data of a power distribution network including PMU measurement data, SCADA data and marketing data, establishing a power distribution network operation measurement database, and preprocessing bad measurement data;
s2: establishing a multidimensional gray-neural network hybrid coordination prediction model, taking measurement data of a power distribution network as input data, and predicting the output condition of each node of a system at the next moment;
s3: establishing a forward push back power flow algorithm of the branch current of the fusion PMU, and calculating the voltage at each prediction moment for the prediction result of the step S2 by using the algorithm to obtain a power flow result;
s4: the power flow result of each prediction moment is obtained, and the power distribution network operation trend prediction based on PMU is realized;
the specific steps of predicting by the multi-dimensional gray-neural network hybrid coordination prediction model in the step S2 include:
21 The system feature sequence is:
in the formula ,
as a feature element, the relevant multifactor sequence is:
……
wherein ,
is->
G=1, 2, …, N, then +.>
Is->
The immediately adjacent mean sequence of (2) is:
the GM (1, n) model is obtained as:
wherein ,
for gray derivative->
For background value, a is the coefficient of development, +.>
Is a driving item;
22 Calculating a multidimensional gray GM (1, n) model:
for a pair of
Generates +.>
wherein ：
the accuracy of gray predictions is related to the smoothness of the original sequence, the better the smoothness, the higher the accuracy of the predictions, for
Performing quasi-smoothness test, exponential rule test, and sequence +.>
The expression of the smoothness ratio ρ (k) is:
wherein if it
The sequence is as follows:
ρ(k)∈[0,0.5],k＝2,3,…,n；
then call for
For quasi-smooth sequences, grey modeling can be performed directly, otherwise the original sequence needs to be preprocessed, if k>3 and satisfies the following formula:
then it is determined that
Meets the index rule;
from the following components
Generating sequence next to the mean->
And solving gray scale parameters;
wherein ：
is provided with
Is a parameter column, and:
23 Determining a model and solving, and obtaining a predicted value through reducing and subtracting:
order the
Is GM (1, N) model->
The expression of the whitening equation is:
solving for
Is a simulation of the values of:
when (when)
When the amplitude change is small, the patient is->
The approximate corresponding time for the ash constant GM (1, n) is:
wherein ,
get->
Reduction to determine->
The predicted values of (2) are: />
The multidimensional gray-neural network hybrid coordination prediction model introduces a smoothing control parameter c into GM (1, N), limits the smoothing condition of GM (1, N) in the prediction process, and the expression of the limiting condition is as follows:
wherein: k=4, 5, …, n, adding the original signature sequence addition factor c to GM (1, n) to make the new sequence satisfy the following formula:
from the above, it can be seen that
Obtaining a proper c to enable the proper c to meet the above formula, wherein the specific expression is as follows:
further, the following steps are obtained:
selecting proper c adding sequence from the above formula, and adding proper influencing factor c sequence
The ash differential equation is established as follows:
estimating parameters by using a least square method to obtain:
[b 1 ,b2,…b n ,μ] T ＝(B T B) -1 B T Y H
wherein ：
under the initial conditions
The prediction of the accumulated sequence can be obtained:
pair sequence
The predicted reduction values of (2) are:
final pair sequence
The predicted expression for each term of (a) is:
the multidimensional gray-neural network hybrid coordination prediction model is corrected by utilizing BP neural network system residual errors, and the method specifically comprises the following steps:
a) Acquiring a power sequence to be predicted, predicting by using the preprocessed load measurement data as a characteristic sequence and using a multivariate gray algorithm to calculate the difference between a predicted result and the power sequence as residual epsilon, predicting the epsilon sequence by using a BP neural network, and expressing the predicted result sequence as epsilon BP ；
b) Calculating the predicted result sequence epsilon of the residual sequence epsilon and the BP neural network pair epsilon BP Difference sequence epsilon 2 I.e. epsilon 2 ＝ε-ε BP For epsilon 2 BP neural network prediction is carried out, and the predicted result is expressed as epsilon 2BP ；
c) Adding the predicted results of the step a) and the step b) to obtain a final predicted result sequence S GM1 ＝S GM +ε BP +ε 2BP ；
In step S3, the step of obtaining the current result by using the forward push back power flow algorithm of the branch current of the fusion PMU specifically includes the following steps:
31 Starting from the layer node, obtaining branch current according to kirchhoff current law;
in the formula ,
for node i to node j point phaseCurrent (I)>
For j-point phase voltage +.>
At the power of j point potential, P j 、Q j Injecting active power and reactive power for the point j;
32 Gradually calculating the injection current of the non-end node from the second layer, wherein the injection current is equal to the sum of the injection current of the node according to kirchhoff current law;
wherein m is all branches directly connected with the node j, I jT Is the T branch current connected with the node j;
33 According to the branch currents of all branches obtained in the steps 31) and 32), utilizing the known root node voltage to sequentially obtain the load node voltages from the root node backwards;
34 Calculating a voltage amplitude correction amount and a maximum voltage amplitude correction amount for each node:
35 Judging convergence conditions:
wherein Λ is the iteration number, if the maximum voltage correction is smaller than the threshold value psi, jumping out of the loop, and outputting a voltage calculation result; otherwise, repeating the steps 31) to 35) until the condition of the above formula is satisfied;
36 After the voltage and current of each node are obtained, the potential of each node is calculated to be in power
As a result of tidal currents.
2. The PMU-based power distribution network operational trend prediction method according to claim 1, wherein the preprocessing of the bad measurement data in the step S1 includes a measurement data defect processing and a load measurement data jump processing.
3. The PMU-based power distribution network operational trend prediction method according to claim 2, wherein the measurement data defect processing specifically comprises:
suppose the 24-point load on a day is:
L＝{l(1),l(2),…l(24)}
judging the defect of the measured data:
l (h) -l (h-1) | < lambdah=2, 3, …,24, λ is a constant
If the difference between any two adjacent load measurement data in the day is smaller than lambda, judging the load measurement data in the day as abnormal data, and deleting the corresponding measurement data; if the load measurement data of any two hours and more in the day is less than lambda, the processing is performed according to the following principle:
for holidays: according to the characteristic that the load measurement data has periodicity and annual similarity, the load measurement data of the day of the last year is searched to be used as a reference load L (y-1, d), the closest load measurement data is found before the correction date, the difference delta between the load measurement data and the reference load L is used as the annual increment of the load measurement data, and then the expression of the load measurement data correction value L (y, d) corresponding to the day to be corrected is as follows:
L(y,d)＝L(y-1,d)+Δ
for the common day: according to the load day similarity principle, seeking normal operation data before an abnormal day, and processing according to the following formula:
in the formula ,PR (h, r) is the h normal measurement value, P R (f, r) is the f-th normal measurement value,
is the average value of f-1 normal measurement values, < >>
The average value of K normal points is the normal measurement value.
4. The PMU-based power distribution network operational trend prediction method according to claim 3, wherein the specific contents of the load measurement data jump processing are:
judging whether the load measurement data between any two points in the day is jump data or not, wherein a judgment equation is as follows:
if the judgment equation is satisfied, judging the load measurement data as jump data, correcting the head end load measurement data by adopting a right-adjacent-stage ratio generating method, correcting the tail end load measurement data by adopting a left-adjacent-stage ratio generating method, and averaging the intermediate load measurement data, wherein the expression for averaging the intermediate load measurement data is as follows:
l(1)＝[l(2)] 2 /l(3)
l(h)＝0.5×[l(h-1)+l(h+1)]h＝2,3,…23
l(n)＝[l(n-1)] 2 /l(n-2) n＝24
5. the PMU-based power distribution network operational trend prediction method according to claim 1, wherein the specific contents of the prediction using the multivariate gray algorithm are: and taking the voltage and the current as multivariable influence factors, introducing a smoothing factor, ensuring the smoothness of the characteristic sequence, determining a GM (1, N) model, and obtaining a predicted value of the characteristic sequence through accumulation and subtraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910251522.1A CN109993361B (en) | 2019-03-29 | 2019-03-29 | PMU-based power distribution network operation trend prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910251522.1A CN109993361B (en) | 2019-03-29 | 2019-03-29 | PMU-based power distribution network operation trend prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109993361A CN109993361A (en) | 2019-07-09 |
| CN109993361B true CN109993361B (en) | 2023-06-13 |
Family
ID=67131127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910251522.1A Active CN109993361B (en) | 2019-03-29 | 2019-03-29 | PMU-based power distribution network operation trend prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109993361B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110490378A (en) * | 2019-08-07 | 2019-11-22 | 中国南方电网有限责任公司 | The calculation method of Power Network Status Estimation precision based on cloud SCADA big data |
| CN110866652B (en) * | 2019-11-21 | 2023-02-28 | 国网四川省电力公司电力科学研究院 | Online PMU data error correction method and system based on LSTM model |
| CN110988594B (en) * | 2019-12-06 | 2022-04-08 | 国网上海市电力公司 | Power distribution network fault position determining method and determining device |
| CN111144638B (en) * | 2019-12-24 | 2023-12-12 | 东南大学 | Power distribution network operation situation prediction method based on big data |
| CN113783179B (en) * | 2020-06-09 | 2023-10-24 | 北京映翰通网络技术股份有限公司 | Power grid load prediction and optimization method |
| CN112819067A (en) * | 2021-01-28 | 2021-05-18 | 广东电网有限责任公司广州供电局 | Method, device, equipment and storage medium for processing bad data of power distribution network |
| CN116231650A (en) * | 2022-11-09 | 2023-06-06 | 南方电网数字电网研究院有限公司 | Power system online power flow calculation method supporting multisource real-time data fusion |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103633739A (en) * | 2013-11-28 | 2014-03-12 | 中国科学院广州能源研究所 | Microgrid energy management system and method |
| CN108508863A (en) * | 2017-10-12 | 2018-09-07 | 上海智容睿盛智能科技有限公司 | A kind of electromechanical equipment method for diagnosing faults based on gray model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101661069B (en) * | 2009-09-25 | 2011-07-20 | 北京四方继保自动化股份有限公司 | Dynamic process real-time estimation method of weak observable non-PMU measuring point independent of state matrix |
-
2019
- 2019-03-29 CN CN201910251522.1A patent/CN109993361B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103633739A (en) * | 2013-11-28 | 2014-03-12 | 中国科学院广州能源研究所 | Microgrid energy management system and method |
| CN108508863A (en) * | 2017-10-12 | 2018-09-07 | 上海智容睿盛智能科技有限公司 | A kind of electromechanical equipment method for diagnosing faults based on gray model |
Non-Patent Citations (1)
| Title |
|---|
| 区域电力***安全风险评估研究;高攀;《CNKI中国知网》;20170215(第02期);第1-53页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109993361A (en) | 2019-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109993361B (en) | PMU-based power distribution network operation trend prediction method | |
| CN108846517B (en) | Integration method for predicating quantile probabilistic short-term power load | |
| Mohn et al. | Tracing PV and QV curves with the help of a CRIC continuation method | |
| CN109308571B (en) | Distribution line variable relation detection method | |
| CN107292502B (en) | Power distribution network reliability assessment method | |
| CN110570108B (en) | Optimal load reduction algorithm based on Lagrange multiplier and application thereof | |
| CN110543720B (en) | State estimation method based on SDAE-ELM pseudo-measurement model | |
| CN109088407B (en) | Power distribution network state estimation method based on deep belief network pseudo-measurement modeling | |
| CN110363334B (en) | Grid line loss prediction method of photovoltaic grid connection based on gray neural network model | |
| CN109711662A (en) | A kind of power grid robust state estimation method based on multisource data fusion | |
| CN114997566A (en) | Power grid blocking risk assessment method and system considering node connectivity loss | |
| CN115600722A (en) | Long-term prediction method for electric carbon factor of enterprise | |
| CN115877312A (en) | Electric energy meter informatization evaluation calibration model based on station area electric energy conservation | |
| Wang et al. | A multi-objective approach to determine time series aggregation strategies for optimal design of multi-energy systems | |
| CN112232570A (en) | Forward active total electric quantity prediction method and device and readable storage medium | |
| CN117335425A (en) | Tidal current calculation method based on GA-BP neural network | |
| CN117335426A (en) | Key feature identification method suitable for short-term voltage stability evaluation | |
| CN111967652A (en) | Double-layer cooperative real-time correction photovoltaic prediction method | |
| CN111861256A (en) | Active power distribution network reconstruction decision method and system | |
| CN110752622A (en) | Power distribution network affine state estimation method | |
| CN116085685A (en) | Method and system for guaranteeing stable gas supply of natural gas in peak period of gas consumption | |
| CN113111588B (en) | NO of gas turbine X Emission concentration prediction method and device | |
| Ju et al. | Short-term hybrid forecasting model of wind power based on EEMD-ARIMA-LSTM | |
| CN110633917A (en) | New energy policy implementation effect comprehensive evaluation method based on electric power big data | |
| Liu et al. | A Method for Power System Reliability Assessment Based on BP Neural Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |