CN108022014A - A kind of Load Prediction In Power Systems method and system - Google Patents
A kind of Load Prediction In Power Systems method and system Download PDFInfo
- Publication number
- CN108022014A CN108022014A CN201711269680.7A CN201711269680A CN108022014A CN 108022014 A CN108022014 A CN 108022014A CN 201711269680 A CN201711269680 A CN 201711269680A CN 108022014 A CN108022014 A CN 108022014A
- Authority
- CN
- China
- Prior art keywords
- mrow
- data
- msub
- load
- msup
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000012706 support-vector machine Methods 0.000 claims abstract description 20
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims description 12
- 210000000349 chromosome Anatomy 0.000 claims description 9
- 238000009825 accumulation Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 230000002087 whitening effect Effects 0.000 claims description 2
- 239000002994 raw material Substances 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- 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)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a kind of Load Prediction In Power Systems method and system, including：Obtain electric system historical load data；The historical load data got is pre-processed；Trend is carried out to pretreated data to handle, obtain trend term using gray theory；Carry out spectrum analysis and determine Decomposition order, the data after going trend processing are decomposed using variation mode decomposition；The support vector cassification summation that data after decomposition are optimized using improved NSGA II is reconstructed；Trend term is added to obtain final prediction result according to the result after reconstruct；The predicted value of each load component is overlapped, determines actual prediction result.Beneficial effect of the present invention：Load Prediction In Power Systems method and system provided by the present invention, the support vector machine method optimized based on gray theory-variation mode decomposition and NSGA II, can successfully managing load prediction, to fluctuate big accuracy not high, the shortcomings that being easily trapped into local optimum.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a power system load prediction method and system.
Background
The power load prediction is an important content, premise and basis of power system planning and power grid operation. Under the situation that energy conservation and environmental protection are vigorously advocated in China to save the existing energy consumption, the accuracy of power load prediction is related to the economic and efficient operation of the whole power grid enterprise and the safe operation of the whole power generation power grid, namely the current situation puts forward a higher standard requirement on the accuracy of power load prediction.
The conventional methods for short-term load prediction commonly used at present include a classical prediction method represented by a time series method and a regression analysis method, and an artificial intelligence method represented by an expert system method, a neural network and a fuzzy logic method. Because the power load change process is a highly complex nonlinear process, the traditional method is difficult to establish an effective mathematical model, so that the accuracy of a prediction result is not high.
The gray model method in the existing power system prediction has larger error when predicting the power load with large fluctuation change and has poor prediction precision on discrete data.
The mature theoretical basis of artificial intelligence is applied to an actual power system, but the artificial intelligence and the mature theoretical basis of artificial intelligence have respective defects, and the actual prediction effect is not ideal.
Disclosure of Invention
The invention aims to solve the problems and provides a method and a system for predicting the load of a power system, wherein the method and the system are based on a support vector machine method of grey theory-variational modal decomposition and NSGA-II optimization, and can effectively overcome the defects that the load prediction is large in fluctuation, low in accuracy and easy to fall into local optimization.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses a load prediction method of a power system, which comprises the following steps:
(1) acquiring historical load data of a power system;
(2) preprocessing the acquired historical load data;
(3) performing trend removing processing on the preprocessed data by using a grey theory to obtain a trend item;
(4) performing frequency spectrum analysis to determine the number of decomposition layers, and decomposing the data subjected to trend removing processing by using variational modal decomposition;
(5) reconstructing decomposed data by utilizing the classified summation of an improved NSGA-II optimized support vector machine;
(6) adding a trend item according to the reconstructed result to obtain a final prediction result;
(7) and superposing the predicted values of the load components to determine an actual prediction result.
Further, the preprocessing the acquired historical load data specifically includes:
the data is subjected to equal spacing and outlier removal processing, the data containing the edge values is equally spaced, and the edge data is processed.
Further, the trend removing processing on the preprocessed data by using the gray theory specifically comprises:
and accumulating or subtracting the preprocessed data to generate a gray model, determining parameters of a differential equation by using time series data, gradually whitening the gray quantity, and predicting a future state.
Further, solving parameters a and u according to the definition of the grey system theory;
obtaining a prediction model according to the obtained parameters a and u as follows:
by subtraction reduction, the final prediction results were obtained as follows:
wherein x is(1)Is x(0)A new sequence is generated by the first accumulation; a and u are both parameters, and u is a control item; x is the number of(0)(1) Is initial data; x is the number of(0)(t) is data at time t, x(1)(t) is x(0)(t) accumulating the generated new sequence for one time;the data at the time of t +1,generated for one accumulationA new sequence.
Further, decomposing the data after trend removing processing by using variational modal decomposition specifically comprises:
decomposing the input signal into k finite bandwidths with center frequencies, called modes u, by using variational mode decomposition datakThe sum of the bandwidth estimates for each modality is minimized.
Further, the decomposition model is specifically:
in the formula, delta (t) is an impulse function;calculating partial derivative for t; f is the primary signal, ukA modal component of the kth limited broadband; omegakThe center frequency of the modal component of the kth limited broadband; u. ofk(t) is the modal component of the kth finite wide band at time t.
Further, reconstructing the decomposed data by using the classified summation of the support vector machine optimized by the improved NSGA-II, specifically comprising the following steps:
according to a support vector machine model, introducing a Lagrange multiplier algorithm to obtain an optimized objective function:
introducing non-linear mappingRn→ H, map the samples to a new data set:
the above problem translates into:
wherein, αiIs a lagrange multiplier.
Further, a normalization method is used for preprocessing operation, parameters of the support vector machine are optimized, and the method specifically comprises the following steps:
in the formula: x'ijIs a normalized value, x 'of the ith input dimension j'ij∈[0,1](ii) a The original value of j dimension for the ith input; x is the number ofjminThe minimum value in the j-th dimension is taken as all the input; x is the number ofjmaxThe maximum value in the j-th dimension for all inputs;
setting population size PpopIterative algebra GgenEncoding a chromosome by using floating-point number encoding, wherein the 1 st bit of the chromosome is a multiplication factor C, the 2 nd bit is a radial basis kernel function parameter gamma, and the 3 rd bit is an EmseThe 4-position is 1-R and the 5-position is RrankI.e. the hierarchy of each chromosome, the smaller the number the greater the fitness, and D at position 6disThe method is a basis for judging sparsity when the levels are the same;
randomly generating an initial population with EmseAnd R is an objective function, and rapid non-dominated sorting, selection, crossing and variation are carried out;
when the iteration times reach the maximum iteration times, the optimal parameters C and gamma are obtained.
The invention also discloses a system for forecasting the load of the power system, which comprises the following components:
the data acquisition module is used for acquiring historical load data of the power system;
the preprocessing module is used for carrying out equal-interval processing and outlier removal processing on the data;
the trend removing processing module is used for removing the trend of the original data by utilizing a grey theory;
the training optimization module is used for carrying out spectrum analysis to determine the number of decomposition layers, decomposing input data by using variational modal decomposition, and carrying out classified summation reconstruction on the processed data by using an improved NSGA-II optimized support vector machine;
the prediction module is used for adding a trend item according to the reconstructed result to obtain a prediction result;
and the result determining module is used for superposing the predicted values of the load components to determine an actual prediction result.
Further, still include:
and the processing module is used for carrying out normalization processing on the historical load data of the power system after the historical load data of the power system is obtained.
The invention has the beneficial effects that:
the invention provides a load prediction method and a load prediction system of a power system, which are characterized in that historical load data of a point system is obtained, a grey theory is utilized to perform trend removing processing on the historical load data, the number of decomposition layers is determined by utilizing frequency spectrum analysis, input data are decomposed by variational modal decomposition, then a support vector machine optimized by improved NSGA-II is utilized to perform classification summation reconstruction, and finally a result is added with a trend item and is superposed to obtain a prediction result.
Drawings
FIG. 1 is a flow chart of an embodiment of a method for load prediction in an electrical power system according to the present invention;
FIG. 2 is a diagram of a prediction model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of SVM processing data;
FIG. 4 is a graph of the amplitude variation of sampled data and predicted using gray theory;
FIG. 5 is a graph of sampled data amplitude change after detrending;
FIG. 6 is a plot of spectral analysis after sample data detrending;
FIG. 7 is a diagram of analysis of the modes and spectra after the decomposition of the variation modes of the sampled data;
fig. 8(a) -8 (c) show the prediction results of the modal components of the sample data;
FIG. 9 is a graph showing the comparison result between the final prediction results of the method GM-VMD-SVM and the final prediction results of the other two methods EMD-SVM and GM-SVM;
fig. 10 is a block diagram of a power system load prediction apparatus according to an embodiment of the present invention.
The specific implementation mode is as follows:
the invention is further illustrated by the following examples in conjunction with the accompanying drawings:
fig. 1 shows a flowchart of an embodiment of a method for loading a power system, where the method includes:
step S101: acquiring historical load data of a power system;
the historical load data of the power system can be historical data collected by the data collecting and monitoring device. After the historical load data is obtained, normalization preprocessing can be further performed on the historical load data of the power system.
Step S102: preprocessing the data;
the data is subjected to equal interval processing and outlier removal processing, the data containing the edge values are subjected to equal interval processing, and the edge data are processed to improve the prediction accuracy.
Step S103: the data were detrended using the grey theory.
The definition of the gray system theory is
x(0)(t)+az(1)(t)＝u t＝2,3,…,N；
z(1)(t)＝0.5x(1)(t-1)+0.5x(1)(t)；
Wherein x(1)Is x(0)A new sequence is generated by the first accumulation; a and u are parameters, wherein u is a control term, and the two equations are obtained:
substituting the obtained a and u into the following equation:
can obtain a prediction model of
By subtraction reduction, the final prediction results were obtained as follows:
x(1)is x(0)A new sequence is generated by the first accumulation; a and u are both parameters, and u is a control item; x is the number of(0)(1) Is initial data; x is the number of(0)(t) is data at time t, x(1)(t) is x(0)(t) accumulating the generated new sequence for one time;the data at the time of t +1,generated for one accumulationA new sequence.
Step S104: and performing spectral analysis to determine the decomposition layer number, decomposing the input data by using variational modal decomposition, and classifying, summing and reconstructing the processed data by using an improved NSGA-II optimized support vector machine.
Decomposing the input signal into k finite bandwidths with center frequencies, called modes u, by using variational mode decomposition datakThe sum of the bandwidth estimates for each modality is minimized, with the decomposition model as follows:
in the formula: { uk}:＝{u1,…,ukIs the modal component; { omega [ [ omega ] ]k}:＝{ω1,ω2,…,ωkThe center frequency of each modal component is multiplied by the frequency; delta (t) is an impulse function;calculating partial derivative for t; f is the original signal.
In order to obtain the optimal solution of the constraint problem, a Lagrange multiplication operator lambda (t) is introduced, and the constraint variable problem is converted into an unconstrained variable problem:
the optimal solution can be obtained by solving the saddle point of the Lagrangian function through an alternating direction multiplier algorithm.
And (3) carrying out classified summation reconstruction by using a support vector machine optimized by the improved NSGA-II, wherein the support vector machine model is as follows:
where omega is hyperplane normal vector, C is penalty factor, n is sample number, ξ is relaxation factor representing allowable error rate under linear irreducible condition, y isiIs the sample output, and yi∈{-1,+1}；xiThe sample input quantity is; b is a threshold value.
Introducing a Lagrange multiplier algorithm to obtain an optimized objective function:
formula (III) αiIs a lagrange multiplier.
Introducing non-linear mappingRn→ H, map the samples to a new data set
The above problem translates into:
the prediction accuracy and performance of the SVM regression model are generally characterized by the mean square error Emse and the correlation coefficient R, and the formula is as follows:
preprocessing operation is carried out by utilizing a normalization method, parameters of the support vector machine are optimized, and the formula is as follows:
in the formula: x'ijIs a normalized value, x 'of the ith input dimension j'ij∈[0,1](ii) a For the ith inputjThe original value of the dimension; x is the number ofjminThe minimum value in the j-th dimension is taken as all the input; x is the number ofjmaxThe maximum value in the j-th dimension for all inputs.
Setting population size PpopIterative algebra GgenEncoding a chromosome by using floating-point number encoding, wherein the 1 st bit of the chromosome is a multiplication factor C, the 2 nd bit is a radial basis kernel function parameter gamma, and the 3 rd bit is an EmseThe 4-position is 1-R and the 5-position is RrankI.e. the hierarchy of each chromosome, the smaller the number the greater the fitness, and D at position 6disThe criterion is to determine the sparsity when the levels are the same.
Randomly generating an initial population with EmseAnd R is an objective function, and rapid non-dominated sorting, selection, intersection and mutation are carried out.
When the iteration times reach the maximum iteration times, the optimal parameters C and gamma are obtained.
And step S105, obtaining a prediction result according to the reconstructed result and the trend item. The trend term is a trend simulation of the data obtained by the gray theoretical fitting mentioned in step S103, i.e., the fitting value in fig. 4.
And step S106, overlapping the predicted values of the load components to determine an actual prediction result.
Fig. 4-8 illustrate a process for predicting the acquired load history data in one embodiment. The raw data of fig. 4 is first detrended by using a gray model to obtain the data shown in fig. 5, and the dominant frequency is obtained as 6, i.e., fig. 6. Then, the data in fig. 6 is decomposed by using the variation modal decomposition, the number of decomposition layers is 6, and each mode and the spectrum analysis thereof as shown in fig. 7 are obtained. And then, carrying out classification summation reconstruction by using an improved NSGA-II optimized support vector machine, and then adding a trend term to obtain prediction results of each mode shown in figures 8(a) to 8 (c). And finally, superposing the modal prediction results to obtain a load prediction curve shown by the GM-VMD-SVM in the figure 9, wherein the other two curves in the figure 9 are the load prediction results of the EMD-SVM and the GM-SVM which are the other two methods respectively. The comparison shows that the prediction effect of the method used in the patent is better.
The present invention also provides a power system load prediction apparatus, as shown in fig. 10, including:
the data acquisition module is used for acquiring historical load data of the power system;
the preprocessing module is used for carrying out equal-interval processing and outlier removal processing on the data;
the trend removing processing module is used for removing the trend of the original data by utilizing a grey theory;
the training optimization module is used for carrying out spectrum analysis to determine the number of decomposition layers, decomposing input data by using variational modal decomposition, and carrying out classified summation reconstruction on the processed data by using an improved NSGA-II optimized support vector machine;
the processing module is used for carrying out normalization processing on the historical load data of the power system after the historical load data of the power system is obtained;
the prediction module is used for adding a trend item according to the reconstructed result to obtain a prediction result;
and the result determining module is used for superposing the predicted values of the load components to determine an actual prediction result.
The method utilizes a support vector machine method which is subjected to grey theory-variational modal decomposition and NSGA-II optimization, and can better deal with the nonlinearity and the non-stationarity of the power load signal, namely when a signal with high volatility is processed, the variational modal decomposition can be well processed; when the edge value, namely the redundant item, is responded, the grey theoretical model can better utilize the known data to predict; NSGA-II has better global search capability and can improve the prediction precision.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A method for predicting a load of an electric power system, comprising the steps of:
(1) acquiring historical load data of a power system;
(2) preprocessing the acquired historical load data;
(3) performing trend removing processing on the preprocessed data by using a grey theory to obtain a trend item;
(4) performing frequency spectrum analysis to determine the number of decomposition layers, and decomposing the data subjected to trend removing processing by using variational modal decomposition;
(5) reconstructing decomposed data by utilizing the classified summation of an improved NSGA-II optimized support vector machine;
(6) adding a trend item according to the reconstructed result to obtain a final prediction result;
(7) and superposing the predicted values of the load components to determine an actual prediction result.
2. The method for predicting the load of the power system according to claim 1, wherein the preprocessing is performed on the acquired historical load data, and specifically comprises:
the data is subjected to equal spacing and outlier removal processing, the data containing the edge values is equally spaced, and the edge data is processed.
3. The method for predicting the load of the power system according to claim 1, wherein the detrending process of the preprocessed data by using the gray theory is specifically as follows:
and accumulating or subtracting the preprocessed data to generate a gray model, determining parameters of a differential equation by using time series data, gradually whitening the gray quantity, and predicting a future state.
4. A power system load prediction method as claimed in claim 3,
solving parameters according to the definition of gray system theoryaAndu；
according to the obtained parametersaAnduthe prediction model obtained is:
<mrow> <msup> <mover> <mi>x</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mrow> <mo>(</mo> <msup> <mi>x</mi> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> </msup> <mo>(</mo> <mn>1</mn> <mo>)</mo> <mo>-</mo> <mfrac> <mi>u</mi> <mi>a</mi> </mfrac> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>a</mi> <mi>t</mi> </mrow> </msup> <mo>+</mo> <mfrac> <mi>u</mi> <mi>a</mi> </mfrac> <mo>;</mo> </mrow>
by subtraction reduction, the final prediction results were obtained as follows:
<mrow> <msup> <mover> <mi>x</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <msup> <mover> <mi>x</mi> <mo>^</mo> </mover> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>-</mo> <msup> <mi>x</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msup> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>;</mo> </mrow>
wherein,x (1)is composed ofx (0)A new sequence is generated by the first accumulation;aanduare all parameters which are used as the raw materials,uis a control item; x is the number of(0)(1) Is initial data; x is the number of(0)(t) is data at time t, x(1)(t) is x(0)(t) accumulating the generated new sequence for one time;the data at the time of t +1,generated for one accumulationA new sequence.
5. The method for predicting the load of the power system according to claim 1, wherein decomposing the data after the trend removing process by using the variational modal decomposition specifically comprises:
decomposing the input signal into k finite bandwidths with center frequencies, called modes u, by using variational mode decomposition datakThe sum of the bandwidth estimates for each modality is minimized.
6. The method of claim 5, wherein the decomposition model is specifically:
<mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>min</mi> <mo>{</mo> <mo>&Sigma;</mo> <mo>|</mo> <mo>|</mo> <msub> <mo>&part;</mo> <mi>t</mi> </msub> <mo>+</mo> <mfrac> <mi>j</mi> <mrow> <mi>&pi;</mi> <mi>t</mi> </mrow> </mfrac> <msub> <mi>u</mi> <mi>k</mi> </msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <msup> <mi>e</mi> <mrow> <mo>-</mo> <msub> <mi>j&omega;</mi> <mi>k</mi> </msub> <mi>t</mi> </mrow> </msup> <mo>|</mo> <msubsup> <mo>|</mo> <mn>2</mn> <mn>2</mn> </msubsup> <mo>}</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mtable> <mtr> <mtd> <mrow> <mi>s</mi> <mo>.</mo> <mi>t</mi> <mo>.</mo> </mrow> </mtd> <mtd> <mrow> <munder> <mo>&Sigma;</mo> <mi>k</mi> </munder> <msub> <mi>u</mi> <mi>k</mi> </msub> <mo>=</mo> <mi>f</mi> </mrow> </mtd> </mtr> </mtable> </mtd> </mtr> </mtable> </mfenced>
in the formula, delta (t) is an impulse function;calculating partial derivative for t; f is the primary signal, ukA modal component of the kth limited broadband; omegakThe center frequency of the modal component of the kth limited broadband; u. ofk(t) is the modal component of the kth finite wide band at time t.
7. The method according to claim 1, wherein the decomposed data is reconstructed by using a support vector machine classified sum optimized by an improved NSGA-II, and specifically comprises:
according to a support vector machine model, introducing a Lagrange multiplier algorithm to obtain an optimized objective function:
<mrow> <mi>max</mi> <mi> </mi> <mi>L</mi> <mo>=</mo> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>&alpha;</mi> <mi>i</mi> </msub> <mo>-</mo> <mfrac> <mn>1</mn> <mn>2</mn> </mfrac> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>&alpha;</mi> <mi>i</mi> </msub> <msub> <mi>&alpha;</mi> <mi>j</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <msub> <mi>y</mi> <mi>j</mi> </msub> <msubsup> <mi>x</mi> <mi>i</mi> <mi>T</mi> </msubsup> <msub> <mi>x</mi> <mi>j</mi> </msub> <mo>;</mo> </mrow>
<mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <msub> <mi>&alpha;</mi> <mi>i</mi> </msub> <msub> <mi>y</mi> <mi>i</mi> </msub> <mo>=</mo> <mn>0</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mn>0</mn> <mo>&le;</mo> <msub> <mi>&alpha;</mi> <mi>i</mi> </msub> <mo>&le;</mo> <mi>C</mi> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>n</mi> <mo>;</mo> </mrow>
introducing non-linear mappingMapping the samples to a new data set:
the above problem translates into:
wherein, αiIs a lagrange multiplier.
8. The method according to claim 7, wherein the preprocessing operation is performed by using a normalization method to optimize the parameters of the support vector machine, specifically:
<mrow> <msubsup> <mi>x</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> <mo>&prime;</mo> </msubsup> <mo>=</mo> <mfrac> <mrow> <msub> <mi>x</mi> <mi>j</mi> </msub> <mo>-</mo> <msub> <mi>x</mi> <mrow> <mi>j</mi> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> </mrow> <mrow> <msub> <mi>x</mi> <mrow> <mi>j</mi> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>x</mi> <mrow> <mi>j</mi> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> </mrow> </mfrac> <mo>;</mo> </mrow>
in the formula: x'ijIs a normalized value, x 'of the ith input dimension j'ij∈[0,1](ii) a The original value of j dimension for the ith input; x is the number ofjminThe minimum value in the j-th dimension is taken as all the input; x is the number ofjmaxThe maximum value in the j-th dimension for all inputs;
setting population size PpopIterative algebra GgenEncoding a chromosome by using floating-point number encoding, wherein the 1 st bit of the chromosome is a multiplication factor C, the 2 nd bit is a radial basis kernel function parameter gamma, and the 3 rd bit is an EmseThe 4-position is 1-R and the 5-position is RrankI.e. the hierarchy of each chromosome, the smaller the number the greater the fitness, and D at position 6disThe method is a basis for judging sparsity when the levels are the same;
randomly generating an initial population with EmseAnd R is an objective function, and rapid non-dominated sorting, selection, crossing and variation are carried out;
when the iteration times reach the maximum iteration times, the optimal parameters C and gamma are obtained.
9. A power system load prediction system, comprising:
the data acquisition module is used for acquiring historical load data of the power system;
the preprocessing module is used for carrying out equal-interval processing and outlier removal processing on the data;
the trend removing processing module is used for removing the trend of the original data by utilizing a grey theory;
the training optimization module is used for carrying out spectrum analysis to determine the number of decomposition layers, decomposing input data by using variational modal decomposition, and carrying out classified summation reconstruction on the processed data by using an improved NSGA-II optimized support vector machine;
the prediction module is used for adding a trend item according to the reconstructed result to obtain a prediction result;
and the result determining module is used for superposing the predicted values of the load components to determine an actual prediction result.
10. The power system load prediction system of claim 9, further comprising:
and the processing module is used for carrying out normalization processing on the historical load data of the power system after the historical load data of the power system is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711269680.7A CN108022014B (en) | 2017-12-05 | 2017-12-05 | Power system load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711269680.7A CN108022014B (en) | 2017-12-05 | 2017-12-05 | Power system load prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108022014A true CN108022014A (en) | 2018-05-11 |
| CN108022014B CN108022014B (en) | 2021-06-01 |
Family
ID=62078389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711269680.7A Active CN108022014B (en) | 2017-12-05 | 2017-12-05 | Power system load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108022014B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960486A (en) * | 2018-06-12 | 2018-12-07 | 郑州航空工业管理学院 | Interactive set evolvement method based on grey support vector regression prediction adaptive value |
| CN109376921A (en) * | 2018-10-15 | 2019-02-22 | 河南理工大学 | Based on hereditary artificial fish school optimization RBF neural short-term load forecasting method |
| CN109449974A (en) * | 2018-11-16 | 2019-03-08 | 大连理工大学 | A kind of integrated energy system Optimization Scheduling based on change Mode Decomposition and sample entropy theory |
| CN109816164A (en) * | 2019-01-16 | 2019-05-28 | 国网信通亿力科技有限责任公司 | A kind of Methods of electric load forecasting |
| CN111178575A (en) * | 2019-11-12 | 2020-05-19 | 国网山西省电力公司晋中供电公司 | Grey model power load prediction method |
| CN113406525A (en) * | 2021-06-15 | 2021-09-17 | 安庆师范大学 | Lithium battery pack residual life prediction method based on optimized variational modal decomposition |
| CN114707710A (en) * | 2022-03-22 | 2022-07-05 | 湖南大学 | Method for evaluating day-by-day net load standby demand of power system and computer device |
| CN116227087A (en) * | 2023-03-29 | 2023-06-06 | 南京理工大学 | Method for predicting tensile characteristics of welded joint of thin-wall part based on finite element method |
| CN117878927A (en) * | 2024-03-11 | 2024-04-12 | 国网黑龙江省电力有限公司绥化供电公司 | Power system load trend analysis method based on time sequence analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
| CN103544544A (en) * | 2013-10-29 | 2014-01-29 | 广东工业大学 | Energy consumption forecasting method and device |
| CN103729501A (en) * | 2013-12-18 | 2014-04-16 | 国网山西省电力公司晋中供电公司 | Short-term power load predicting method based on grey theory |
| CN104376368A (en) * | 2014-08-19 | 2015-02-25 | 上海交通大学 | Wind power generation short-term load forecasting method and device based on frequency domain decomposition |
| DE102014010117A1 (en) * | 2014-07-08 | 2016-01-14 | Evohaus Gmbh | Forecasting and control system for the electricity purchase of households |
| CN107392364A (en) * | 2017-07-12 | 2017-11-24 | 河海大学 | The short-term load forecasting method of variation mode decomposition and depth belief network |
-
2017
- 2017-12-05 CN CN201711269680.7A patent/CN108022014B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
| CN103544544A (en) * | 2013-10-29 | 2014-01-29 | 广东工业大学 | Energy consumption forecasting method and device |
| CN103729501A (en) * | 2013-12-18 | 2014-04-16 | 国网山西省电力公司晋中供电公司 | Short-term power load predicting method based on grey theory |
| DE102014010117A1 (en) * | 2014-07-08 | 2016-01-14 | Evohaus Gmbh | Forecasting and control system for the electricity purchase of households |
| CN104376368A (en) * | 2014-08-19 | 2015-02-25 | 上海交通大学 | Wind power generation short-term load forecasting method and device based on frequency domain decomposition |
| CN107392364A (en) * | 2017-07-12 | 2017-11-24 | 河海大学 | The short-term load forecasting method of variation mode decomposition and depth belief network |
Non-Patent Citations (1)
| Title |
|---|
| 路雪艳、赵超凡、吴晓峰、韩晓霞: "基于改进的NSGA_Ⅱ多目标优化方法研究", 《计算机应用研究》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108960486B (en) * | 2018-06-12 | 2021-10-29 | 郑州航空工业管理学院 | Interactive set evolution method for predicting adaptive value based on gray support vector regression |
| CN108960486A (en) * | 2018-06-12 | 2018-12-07 | 郑州航空工业管理学院 | Interactive set evolvement method based on grey support vector regression prediction adaptive value |
| CN109376921A (en) * | 2018-10-15 | 2019-02-22 | 河南理工大学 | Based on hereditary artificial fish school optimization RBF neural short-term load forecasting method |
| CN109449974A (en) * | 2018-11-16 | 2019-03-08 | 大连理工大学 | A kind of integrated energy system Optimization Scheduling based on change Mode Decomposition and sample entropy theory |
| CN109449974B (en) * | 2018-11-16 | 2021-06-08 | 大连理工大学 | Comprehensive energy system optimization scheduling method based on variable mode decomposition and sample entropy theory |
| CN109816164A (en) * | 2019-01-16 | 2019-05-28 | 国网信通亿力科技有限责任公司 | A kind of Methods of electric load forecasting |
| CN111178575A (en) * | 2019-11-12 | 2020-05-19 | 国网山西省电力公司晋中供电公司 | Grey model power load prediction method |
| CN113406525A (en) * | 2021-06-15 | 2021-09-17 | 安庆师范大学 | Lithium battery pack residual life prediction method based on optimized variational modal decomposition |
| CN113406525B (en) * | 2021-06-15 | 2023-06-02 | 金陵科技学院 | Lithium battery pack residual life prediction method based on optimization variation modal decomposition |
| CN114707710A (en) * | 2022-03-22 | 2022-07-05 | 湖南大学 | Method for evaluating day-by-day net load standby demand of power system and computer device |
| CN114707710B (en) * | 2022-03-22 | 2024-05-28 | 湖南大学 | Method for evaluating daily payload standby requirement of power system and computer device |
| CN116227087A (en) * | 2023-03-29 | 2023-06-06 | 南京理工大学 | Method for predicting tensile characteristics of welded joint of thin-wall part based on finite element method |
| CN117878927A (en) * | 2024-03-11 | 2024-04-12 | 国网黑龙江省电力有限公司绥化供电公司 | Power system load trend analysis method based on time sequence analysis |
| CN117878927B (en) * | 2024-03-11 | 2024-05-28 | 国网黑龙江省电力有限公司绥化供电公司 | Power system load trend analysis method based on time sequence analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108022014B (en) | 2021-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108022014B (en) | Power system load prediction method and system | |
| Li et al. | Short-term load-forecasting method based on wavelet decomposition with second-order gray neural network model combined with ADF test | |
| Ekonomou et al. | A short-term load forecasting method using artificial neural networks and wavelet analysis | |
| CN112990556A (en) | User power consumption prediction method based on Prophet-LSTM model | |
| CN112434848B (en) | Nonlinear weighted combination wind power prediction method based on deep belief network | |
| CN109255728B (en) | Photovoltaic power generation power neural network prediction method based on chaotic phase space optimization reconstruction | |
| CN116010900A (en) | Multi-scale feature fusion gearbox fault diagnosis method based on self-attention mechanism | |
| CN105354643A (en) | Risk prediction evaluation method for wind power grid integration | |
| CN104966161A (en) | Electric energy quality recording data calculating analysis method based on Gaussian mixture model | |
| CN113837434A (en) | Solar photovoltaic power generation prediction method and device, electronic equipment and storage medium | |
| CN112069666B (en) | Power grid short-term reliability evaluation method based on probabilistic power flow method | |
| CN114266416A (en) | Photovoltaic power generation power short-term prediction method and device based on similar days and storage medium | |
| CN110837934B (en) | Smart grid short-term residential load prediction method based on deep learning | |
| CN104239689B (en) | Short-term wind speed forecasting method based on optimization Method Using Relevance Vector Machine | |
| CN112018755A (en) | Photovoltaic power distribution network reactive voltage prediction method and system based on cyclic neural network | |
| Sun et al. | Short-term power load prediction based on VMD-SG-LSTM | |
| CN117689082A (en) | Short-term wind power probability prediction method, system and storage medium | |
| CN117592615A (en) | Short-term power load prediction method based on ICEEMDAN and integrated LSTM-TCN model | |
| CN116706895A (en) | Ultra-short-term wind power prediction method considering deep features and capability constraint | |
| CN116663745A (en) | LSTM drainage basin water flow prediction method based on PCA_DWT | |
| CN113705885A (en) | Power distribution network voltage prediction method and system integrating VMD, XGboost and optimized TCN | |
| Rao et al. | Time series anomaly detection based on CEEMDAN and LSTM | |
| Razif et al. | Application of empirical mode decomposition in improving group method of data handling | |
| Fang et al. | Research on Short-Term and Medium-Term Power Load Forecasting Based on STL-LightGBM | |
| Xia et al. | Research on Solar Radiation Estimation based on Singular Spectrum Analysis-Deep Belief 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 |