CN115471017A - Regional microgrid interconnection optimization method and system based on mutual power assistance - Google Patents
Regional microgrid interconnection optimization method and system based on mutual power assistance Download PDFInfo
- Publication number
- CN115471017A CN115471017A CN202211420989.2A CN202211420989A CN115471017A CN 115471017 A CN115471017 A CN 115471017A CN 202211420989 A CN202211420989 A CN 202211420989A CN 115471017 A CN115471017 A CN 115471017A
- Authority
- CN
- China
- Prior art keywords
- target
- energy
- prediction model
- result
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 113
- 238000013528 artificial neural network Methods 0.000 claims abstract description 64
- 230000015654 memory Effects 0.000 claims abstract description 61
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 43
- 238000012360 testing method Methods 0.000 claims abstract description 24
- 230000005611 electricity Effects 0.000 claims description 46
- 238000012545 processing Methods 0.000 claims description 23
- 238000004458 analytical method Methods 0.000 claims description 22
- 230000000306 recurrent effect Effects 0.000 claims description 19
- 230000003044 adaptive effect Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000013507 mapping Methods 0.000 claims description 4
- 230000000737 periodic effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 11
- 230000008901 benefit Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000004793 poor memory Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 239000013589 supplement Substances 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- 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
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Power Engineering (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a regional micro-grid interconnection optimization method and system based on mutual power assistance, and relates to the technical field of computers, wherein the method comprises the following steps: acquiring target energy information of a target area; dividing to obtain target training data and target test data; decomposing by using a set empirical mode decomposition principle to obtain a plurality of target subsequences; extracting any two target subsequences; training by utilizing a long-short term memory model principle to obtain a long-short term memory prediction model, and training by utilizing a generalized regression neural network principle to obtain a generalized regression neural network prediction model; a combined prediction model is obtained by combining a dynamic self-adaptive variable weight optimization theory; and obtaining a target energy source prediction result of the target test data through the combined prediction model, and performing energy source scheduling optimization. The problem of prior art can't accurate prediction regional energy, and then can't in time energy dispatch is solved. The effects of improving the energy prediction accuracy and optimizing the distributed energy scheduling operation of the regional microgrid are achieved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a regional micro-grid interconnection optimization method and system based on mutual electric energy coordination.
Background
As a new technical means, the micro-grid can effectively integrate energy sources by utilizing a flexible and diversified structure and a control mode in the system, so that the optimal configuration of resources is realized, and the micro-grid becomes an important component part for effectively supplying electric energy. Further, as the number of local area micro-grids increases, individual micro-grid individuals with the same benefits or objectives tend to reach cooperative alliances to form regional micro-grid interconnection systems. The micro-grid group can more effectively utilize the complementary characteristics of energy and load between the sub-micro-grids in the system to mutually supplement electric energy by adopting an interconnection form in a region, thereby improving the reliability and economy of power consumption of users and realizing the cooperative and mutual assistance between the sub-micro-grids in the interconnection system. The intelligent prediction of regional energy in the prior art has limitations, so that the accuracy of energy prediction is influenced, further, energy scheduling cannot be timely performed on regions with energy quantities which do not meet energy use requirements, and the normal use of regional energy is finally influenced. Therefore, the research of analyzing and optimizing the dispatching operation of the regional micro-grid interconnection system with mutual electric energy assistance by using the computer technology has important significance.
However, there are technical problems in the prior art that regional energy cannot be predicted accurately, and then energy scheduling cannot be performed on a region where the energy amount does not meet the energy use requirement through a regional microgrid in time, so that normal use of regional electric energy is affected.
Disclosure of Invention
The invention aims to provide a regional micro-grid interconnection optimization method and system based on electric energy mutual aid, which are used for solving the technical problems that regional energy cannot be accurately predicted in the prior art, and further, energy scheduling cannot be timely performed on a region with energy quantity not meeting energy use requirements through a regional micro-grid, so that normal use of regional electric energy is influenced.
In view of the above problems, the present invention provides a regional microgrid interconnection optimization method and system based on mutual energy coordination.
In a first aspect, the present invention provides a regional microgrid interconnection optimization method based on mutual power, which is implemented by a regional microgrid interconnection optimization system based on mutual power, wherein the method includes: acquiring initial energy information of a target area by collection, and preprocessing the initial energy information to obtain target energy information, wherein the target energy information is information with a time identifier; dividing the target energy information to obtain a target energy information division result, wherein the target energy information division result comprises target training data and target test data; decomposing the target training data by using a set empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences; extracting any two target subsequences in the plurality of target subsequences, and respectively recording the two target subsequences as a first target subsequence and a second target subsequence; training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model; based on the long-short term memory prediction model and the generalized regression neural network prediction model, a combined prediction model is obtained by combining dynamic adaptive variable weight optimization theory analysis; and processing the target test data through the combined prediction model to obtain a target energy prediction result, and performing energy scheduling optimization on the target area according to the target energy prediction result.
In a second aspect, the present invention further provides a regional microgrid interconnection optimization system based on mutual energy, configured to execute the regional microgrid interconnection optimization method based on mutual energy according to the first aspect, wherein the system includes: the system comprises an information acquisition module, a time identification module and a data processing module, wherein the information acquisition module is used for acquiring initial energy information of a target area and preprocessing the initial energy information to obtain target energy information, and the target energy information is information with a time identification; the information dividing module is used for dividing the target energy information to obtain a target energy information dividing result, wherein the target energy information dividing result comprises target training data and target test data; the data decomposition module is used for decomposing the target training data by using an ensemble empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences; a sequence obtaining module, configured to extract any two target subsequences from the multiple target subsequences, and record the extracted two target subsequences as a first target subsequence and a second target subsequence; the model training module is used for training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model; the model obtaining module is used for obtaining a combined prediction model by combining dynamic adaptive variable weight optimization theory analysis based on the long-short term memory prediction model and the generalized regression neural network prediction model; and the scheduling execution module is used for processing the target test data through the combined prediction model to obtain a target energy prediction result and performing energy scheduling optimization on the target area according to the target energy prediction result.
One or more technical schemes provided by the invention at least have the following technical effects or advantages:
acquiring initial energy information of a target area by collection, and preprocessing the initial energy information to obtain target energy information, wherein the target energy information is information with a time identifier; dividing the target energy information to obtain a target energy information division result, wherein the target energy information division result comprises target training data and target test data; decomposing the target training data by using an ensemble empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences; extracting any two target subsequences in the plurality of target subsequences, and respectively recording the target subsequences as a first target subsequence and a second target subsequence; training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model; based on the long-short term memory prediction model and the generalized regression neural network prediction model, a combined prediction model is obtained by combining dynamic adaptive variable weight optimization theory analysis; and processing the target test data through the combined prediction model to obtain a target energy prediction result, and performing energy scheduling optimization on the target area according to the target energy prediction result. Energy prediction of the target area is finally obtained by decomposing and analyzing energy information in the target area, performing energy independent prediction by using different models, and weighting each model by combining a dynamic adaptive variable weight optimization theory. The combined prediction model is obtained by effectively combining different single prediction models, the limitation of the single model is broken, the overall prediction performance of the comprehensive prediction model is improved, the advantages of the different models are fully utilized, and the technical goal of improving the prediction precision of the model is further improved. The target energy prediction result is obtained through combined prediction model prediction, more accurate and reliable prediction reference data are provided for the interconnection operation of the micro-grid in the subsequent region, the actual energy demand of the target region is combined, energy scheduling between the regions is realized by interconnection of the micro-grids, and the technical effects of improving energy optimization configuration and optimizing the scheduling operation of the micro-grid in the regions are achieved.
The above description is only an overview of the technical solutions of the present invention, and the present invention can be implemented in accordance with the content of the description so as to make the technical means of the present invention more clearly understood, and the above and other objects, features, and advantages of the present invention will be more clearly understood.
Drawings
In order to more clearly illustrate the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only exemplary, and for those skilled in the art, other drawings can be obtained according to the provided drawings without inventive effort.
FIG. 1 is a schematic flow chart of a regional microgrid interconnection optimization method based on electric energy mutual aid of the invention;
FIG. 2 is a schematic flow chart of the energy scheduling optimization of the target area according to the target power consumption prediction result and the target energy prediction result in the area micro-grid interconnection optimization method based on mutual power assistance of electric energy of the present invention;
FIG. 3 is a schematic flow chart of a power consumption stage-power consumption list constructed in the method for optimizing regional micro-grid interconnection based on mutual power assistance according to the present invention;
FIG. 4 is a schematic flow chart of obtaining a plurality of target subsequences in the method for optimizing interconnection of local micro-grids based on electric energy mutual aid of the invention;
FIG. 5 is a schematic structural diagram of a regional microgrid interconnection optimization system based on electric energy mutual aid.
Description of reference numerals:
the system comprises an information acquisition module M100, an information division module M200, a data decomposition module M300, a sequence acquisition module M400, a model training module M500, a model acquisition module M600 and a scheduling execution module M700.
Detailed Description
The invention provides a regional micro-grid interconnection optimization method and system based on mutual energy assistance, and solves the technical problem that the normal use of regional electric energy is influenced because regional energy cannot be accurately predicted and further energy scheduling cannot be performed on a region with energy quantity not meeting the energy use requirement through a regional micro-grid in the prior art. The technical effects of improving the energy prediction accuracy and further optimizing the distributed energy scheduling operation of the regional microgrid are achieved.
In the technical scheme of the invention, the acquisition, storage, use, processing and the like of the data all accord with relevant regulations of national laws and regulations.
In the following, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments of the present invention, and it should be understood that the present invention is not limited by the example embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention. It should be further noted that, for the convenience of description, only some but not all of the features relevant to the present invention are shown in the drawings.
Example one
Referring to fig. 1, the present invention provides a regional microgrid interconnection optimization method based on mutual energy, wherein the method is applied to a regional microgrid interconnection optimization system based on mutual energy, and the method specifically includes the following steps:
step S100: acquiring initial energy information of a target area, and preprocessing the initial energy information to obtain target energy information, wherein the target energy information is information with a time identifier;
specifically, the regional micro-grid interconnection optimization method based on mutual energy economy is applied to the regional micro-grid interconnection optimization system based on mutual energy economy, and can predict the energy of a target region by collecting and analyzing the energy information of the target region, so as to realize the aim of providing a data basis for regional micro-grid interconnection optimization. The target area refers to any area which utilizes the regional microgrid interconnection optimization system to conduct regional internal energy source scheduling guidance and optimization. And acquiring energy information in the target area, exemplarily including energy type, corresponding energy distribution and the like, eliminating energy information irrelevant to electric energy, and only reserving the energy information relevant to the electric energy as the target energy information. Exemplary wind energy such as wind power generation, and the like. In addition, specific energy information is in continuous change due to the fact that energy in the target area is migrated along with actual use, energy scheduling and the like, and therefore the target energy information obtained through collection and analysis is subjected to time marking, namely the target energy information is information with time identification. By obtaining the basic condition data such as the type and distribution of the existing energy in the target area, the technical goal of providing basic data for the energy in the follow-up intelligent prediction target area is realized.
Step S200: dividing the target energy information to obtain a target energy information division result, wherein the target energy information division result comprises target training data and target test data;
specifically, the target energy information of the target area is acquired, the target energy information is manually divided based on the information scale, and the divided two parts are the target training data and the target testing data respectively. The target training data are obtained through division, a training data basis is provided for subsequent independent training of each prediction model, target test data are obtained through division, and a detection data basis is provided for subsequent detection and analysis of the prediction performance of the comprehensive prediction model.
Step S300: decomposing the target training data by using a set empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences;
further, as shown in fig. 2, step S300 of the present invention further includes:
step S310: extracting the time identification of the target training data and generating a target training data sequence;
step S320: obtaining target white noise, and adding the target white noise to the target training data sequence to obtain a new target training data sequence;
step S330: obtaining a preset iteration result, and carrying out iterative decomposition on the new target training data sequence until the new target training data sequence meets the preset iteration result to obtain a plurality of target residual error sequences;
further, step S330 of the present invention further includes:
step S331: analyzing to obtain a plurality of local extrema in the new target training data sequence;
step S332: according to the plurality of local extreme values, sequentially establishing an upper envelope and a lower envelope of the new target training data sequence;
step S333: calculating to obtain a mean sequence according to the upper envelope curve and the lower envelope curve, wherein a calculation formula of the mean sequence is as follows:
step S334: wherein, the
Refers to the mean sequence, the
Refers to the upper layer envelope, the
The reference is the lower envelope line, and the d is the new target training data sequence;
step S335: performing residual error operation on the new target training data sequence and the mean value sequence to obtain a target residual error sequence;
step S336: and continuing to iterate until the preset iteration result is met, and generating the plurality of target residual error sequences.
Step S340: and taking the plurality of target residual difference sequences as the plurality of target subsequences.
Specifically, the target training data is preprocessed before training each individual prediction model based on the target training data.
Firstly, extracting the time mark of the target training data, and generating the time sequence of the target training data according to the time mark, namely generating the target training data sequence. And then, mixing the target white noise to the target training data sequence to obtain a new target training data sequence after mixing. Wherein the target white noise is used to assist in analyzing noisy data while facilitating the mixing of subsequent individual predictive models. And then, carrying out iterative decomposition on the new target training data sequence by using a set empirical mode decomposition principle until a decomposition result meets an inherent mode function, namely the preset iteration result, and obtaining a plurality of target residual error sequences. The empirical mode decomposition is a self-adaptive data processing tool, and can extract and analyze feature information from the new target training data sequence and decompose the feature information to obtain a series of inherent mode functions.
Further, the ensemble empirical mode decomposition firstly analyzes a plurality of local extrema in the new target training data sequence, and establishes an upper envelope and a lower envelope of the new target training data sequence according to the plurality of local extrema. Then, a mean sequence is obtained through calculation according to the upper envelope curve and the lower envelope curve, wherein a calculation formula of the mean sequence is as follows:
wherein, the
Refers to the mean sequence, the
Is the upper layer envelope of the above
Refers to the lower envelope, and d refers to the new target training data sequence.
And finally, performing residual error operation on the new target training data sequence and the mean value sequence to obtain a target residual error sequence, and continuously iterating until an inherent modal function is met to generate a plurality of target residual error sequences. Wherein the plurality of target residual error sequences are a plurality of components of a natural modal function. And finally, taking the target residual difference sequences as the target subsequences, namely forming a target training data decomposition result.
By preprocessing the target training data before model training, the influence of the intermittence and randomness of the energy data in the target training data on the accuracy of subsequent prediction is avoided, and the technical effect of reducing the influence of the instability of the energy data on the prediction model is achieved.
Step S400: extracting any two target subsequences in the plurality of target subsequences, and respectively recording the target subsequences as a first target subsequence and a second target subsequence;
step S500: training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model;
step S600: based on the long-short term memory prediction model and the generalized regression neural network prediction model, a combined prediction model is obtained by combining dynamic adaptive variable weight optimization theory analysis;
specifically, based on the multiple target subsequences obtained after information decomposition processing, data training is performed through the principle of each individual prediction model, and then each individual prediction model is obtained.
Firstly, any two target subsequences in the plurality of target subsequences are extracted and respectively recorded as a first target subsequence and a second target subsequence. Wherein the first target subsequence refers to any one of a plurality of target subsequences, and the second target subsequence is different from the first target subsequence. And then training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model. The long-term and short-term memory model is a model established based on a recurrent neural network, and the problem of poor memory of the recurrent neural network on historical information is solved by adding long time lag to the recurrent neural network. And further, training the second target subsequence by using a generalized regression neural network principle to obtain a generalized regression neural network prediction model. The generalized regression neural network model is based on nonlinear regression, can be trained to obtain a good regression effect when the number of samples is small, and is suitable for solving the problem of nonlinear characteristics.
Further, after the long-short term memory prediction model and the generalized regression neural network prediction model are obtained through training respectively, in order to improve the model prediction accuracy, the weight dynamic update is carried out on each model through a dynamic adaptive variable weight optimization theory, and therefore the combined prediction model is obtained. The technical goal of fully utilizing the advantages of each model to carry out comprehensive energy prediction is achieved, the advantage of processing specific data by utilizing a single model is utilized, the goal of fine management of energy data is achieved, and the technical effect of providing a model basis for subsequent energy prediction is achieved.
Step S700: and processing the target test data through the combined prediction model to obtain a target energy prediction result, and performing energy scheduling optimization on the target area according to the target energy prediction result.
Further, as shown in fig. 3, step S700 of the present invention further includes:
step S710: collecting historical electricity utilization data of the target area;
step S720: analyzing the historical electricity utilization data, and obtaining an electricity utilization stage-electricity consumption list according to an analysis result;
further, as shown in fig. 4, step S720 of the present invention further includes:
step S721: preprocessing the historical electricity utilization data to obtain a historical electricity utilization data processing result;
step S722: acquiring a preset power utilization period, and performing period division on the historical power utilization data processing result according to the preset power utilization period to obtain a period division result, wherein the period division result comprises a plurality of periods;
step S723: extracting any one of the multiple periods, and performing stage division on the any one period to obtain a stage division result, wherein the stage division result comprises multiple stages;
step S724: sequentially carrying out power consumption analysis on each stage in the plurality of stages, and calculating to obtain a plurality of power consumption according to analysis results;
step S725: and constructing the electricity utilization stage-electricity consumption list according to the mapping relation between the plurality of stages and the plurality of electricity consumptions.
Step S730: extracting a time identifier in the target energy information to obtain a target time stage;
step S740: traversing the target time stage in the electricity utilization stage-electricity consumption list, and analyzing to obtain a target electricity consumption prediction result;
step S750: and performing energy scheduling optimization on the target area according to the target power consumption prediction result and the target energy prediction result.
Specifically, before performing energy scheduling optimization on the target area according to the target energy prediction result, the actual required electric quantity of the target area is predicted and estimated, and then the targeted energy scheduling is performed by combining the electric quantity demand and the energy prediction of the target area.
Firstly, historical electricity utilization data of the target area are collected, missing data, error data and the like in the historical electricity utilization data are screened and removed, and therefore a historical electricity utilization data processing result is obtained. Exemplarily, the power consumption data in special time periods such as power limit and power failure are removed, the power consumption requirement data are guaranteed to be in line with reality, and the data are guaranteed to be real and reliable. And then, acquiring a preset power utilization period, and periodically dividing the historical power utilization data processing result according to the preset power utilization period to obtain a periodic division result. Wherein the period division result includes a plurality of periods. Then, any one of the multiple periods is extracted, and the any one period is subjected to stage division to obtain a stage division result, wherein the stage division result comprises multiple stages. And then, carrying out power consumption analysis on each stage in the plurality of stages in sequence, and calculating to obtain a plurality of power consumption according to the analysis result. For example, if the electricity consumption data of a certain area in the past year is analyzed, when the preset electricity consumption period is one day, the electricity consumption data processing results of any one day in the electricity consumption data of the area in the past year are divided into stages, if twenty-four stages are obtained by dividing according to hours, the electricity consumption of the same stage in the past year electricity consumption data of the area in each day is subjected to statistical analysis, for example, the electricity consumption of the stage of 7-8 points in each day is subjected to statistics, and the average value is calculated to obtain the average total electricity consumption of the stage of 7-8 points in each day in history, namely the corresponding electricity consumption. And finally, constructing the electricity utilization stage-electricity consumption list according to the mapping relation between the plurality of stages and the plurality of electricity consumptions.
Further, extracting a time identifier in the target energy information to obtain a target time stage, wherein the target time stage is a stage at a moment when the target energy information is collected, traversing the target time stage in the electricity utilization stage-electricity consumption list, and analyzing to obtain a target electricity consumption prediction result. The target electricity consumption prediction result refers to an estimation result of the electricity consumption demand of the target area in the target time phase. And finally, performing energy scheduling optimization on the target area according to the target power consumption prediction result and the target energy prediction result. Exemplarily, when the target energy prediction result in the target area is smaller than the target power consumption prediction result, it indicates that the energy in the target area cannot meet the energy of the actual power consumption demand, and thus energy is called from other areas; when the target energy prediction result in the target area is greater than the target power consumption prediction result, the energy in the target area can meet the energy of the actual power consumption demand, and the surplus of the energy exists, so that when the energy in other areas is not enough, the energy can be called out from the target area.
Energy scheduling among regions is realized by utilizing interconnection of the micro-grids, and the technical effects of improving energy optimal configuration and optimizing regional micro-grid scheduling operation are achieved.
Further, step S750 of the present invention further includes:
step S751: extracting a time identifier of the target energy information to obtain target time;
step S752: sequentially analyzing the prediction results of the long-short term memory prediction model and the generalized regression neural network prediction model based on the target time and the target test data to respectively obtain the prediction results of the long-short term memory prediction model and the generalized regression neural network prediction model;
step S753: according to a dynamic adaptive variable weight optimization theory, sequentially analyzing the weight coefficients of the long-short term memory prediction model and the generalized regression neural network prediction model to respectively obtain a long-short term memory prediction model coefficient and a generalized regression neural network prediction model coefficient;
step S754: and calculating to obtain the target energy prediction result according to the long-short term memory prediction model coefficient, the generalized regression neural network prediction model coefficient, the long-short term memory prediction model prediction result and the generalized regression neural network prediction model prediction result.
Further, the invention also comprises the following steps:
step S7551: obtaining an actual energy observation result of the target time;
step S7552: and calculating to obtain a relative prediction error of the combined prediction model according to the target energy prediction result and the actual energy observation result, wherein a calculation formula of the relative prediction error is as follows:
step S7553: wherein, the
Is the target energy prediction result, the
Refers to the actual energy observation, the t refers to the target time,the above-mentioned
Is the long-short term memory prediction model coefficient, the
Refers to the coefficient of the generalized regression neural network prediction model, the
Is the prediction result of the long-short term memory prediction model, and the
The method refers to the prediction result of the generalized regression neural network prediction model.
Specifically, after a target energy prediction result is obtained by using the combined prediction model, relevant data are collected for predicting error calculation in order to quantify the prediction performance of the model, and therefore the superiority of the combined prediction model is visually evaluated.
Firstly, extracting the time identification of the target energy information to obtain the target time. And then intelligently analyzing the target test data at the target time by respectively utilizing the long-short term memory prediction model and the generalized regression neural network prediction model, and respectively obtaining a long-short term memory prediction model prediction result and a generalized regression neural network prediction model prediction result. And then, according to a dynamic adaptive variable weight optimization theory, sequentially analyzing the weight coefficients of the long-short term memory prediction model and the generalized regression neural network prediction model to respectively obtain a long-short term memory prediction model coefficient and a generalized regression neural network prediction model coefficient. The dynamic self-adaptive variable weight optimization is to perform targeted adjustment on the weight of each model at the next moment according to the historical prediction performance of each pre-individual sub-prediction model forming the combined prediction model in the prediction process. And obtaining the target energy prediction result by weighted calculation according to the long-short term memory prediction model coefficient, the generalized regression neural network prediction model coefficient, the long-short term memory prediction model prediction result and the generalized regression neural network prediction model prediction result.
Further, actually acquiring an actual energy observation result of the target area at the target time, and calculating a relative prediction error of the combined prediction model according to the target energy prediction result and the actual energy observation result, wherein a calculation formula of the relative prediction error is as follows:
wherein, the
Is the predicted result of the target energy, the
Is the actual energy observation, t is the target time, the
Is the long-short term memory prediction model coefficient, the
Refers to the coefficient of the generalized regression neural network prediction model, the
Is the prediction result of the long-short term memory prediction model, and the
The method refers to the prediction result of the generalized regression neural network prediction model.
The actual energy condition of the target area is compared with the model prediction result, and the prediction error data of the model is obtained through calculation, so that the model prediction quality is quantitatively evaluated based on the relative prediction error obtained through calculation, and the technical effect of objectively quantifying the model prediction quality is achieved.
In summary, the regional microgrid interconnection optimization method based on mutual energy coordination provided by the invention has the following technical effects:
acquiring initial energy information of a target area by collection, and preprocessing the initial energy information to obtain target energy information, wherein the target energy information is information with a time identifier; dividing the target energy information to obtain a target energy information division result, wherein the target energy information division result comprises target training data and target test data; decomposing the target training data by using a set empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences; extracting any two target subsequences in the plurality of target subsequences, and respectively recording the target subsequences as a first target subsequence and a second target subsequence; training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model; based on the long-short term memory prediction model and the generalized regression neural network prediction model, a combined prediction model is obtained by combining dynamic adaptive variable weight optimization theory analysis; and processing the target test data through the combined prediction model to obtain a target energy prediction result, and performing energy scheduling optimization on the target area according to the target energy prediction result. Energy prediction of the target area is finally obtained by decomposing and analyzing energy information in the target area, performing energy independent prediction by using different models, and weighting each model by combining a dynamic adaptive variable weight optimization theory. The combined prediction model is obtained by effectively combining different single prediction models, the limitation of the single model is broken, the overall prediction performance of the comprehensive prediction model is improved, the advantages of the different models are fully utilized, and the technical goal of improving the prediction precision of the model is further improved. The target energy prediction result is obtained through combined prediction model prediction, more accurate and reliable prediction reference data are provided for the interconnection operation of the micro-grid in the subsequent region, the actual energy demand of the target region is combined, energy scheduling between the regions is realized by interconnection of the micro-grids, and the technical effects of improving energy optimization configuration and optimizing the scheduling operation of the micro-grid in the regions are achieved.
Example two
Based on the same inventive concept as the method for optimizing the interconnection of the regional microgrid based on mutual energy coordination in the foregoing embodiment, the present invention further provides a system for optimizing the interconnection of the regional microgrid based on mutual energy coordination, please refer to fig. 5, where the system includes:
the information acquisition module M100 is configured to acquire initial energy information of a target area, and preprocess the initial energy information to obtain target energy information, where the target energy information is information with a time identifier;
the information dividing module M200 is configured to divide the target energy information to obtain a target energy information dividing result, where the target energy information dividing result includes target training data and target test data;
the data decomposition module M300 is configured to decompose the target training data by using a set empirical mode decomposition principle to obtain a target training data decomposition result, where the target training data decomposition result includes a plurality of target subsequences;
a sequence obtaining module M400, where the sequence obtaining module M400 is configured to extract any two target subsequences in the multiple target subsequences, and record the two target subsequences as a first target subsequence and a second target subsequence;
the model training module M500 is used for training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model;
a model obtaining module M600, wherein the model obtaining module M600 is used for obtaining a combined prediction model by combining dynamic adaptive variable weight optimization theory analysis based on the long-short term memory prediction model and the generalized regression neural network prediction model;
a scheduling execution module M700, where the scheduling execution module M700 is configured to process the target test data through the combined prediction model to obtain a target energy prediction result, and perform energy scheduling optimization on the target area according to the target energy prediction result.
Further, the scheduling execution module M700 in the system is further configured to:
collecting historical electricity utilization data of the target area;
analyzing the historical electricity utilization data, and obtaining an electricity utilization stage-electricity consumption list according to an analysis result;
extracting a time identifier in the target energy information to obtain a target time stage;
traversing the target time stage in the electricity utilization stage-electricity consumption list, and analyzing to obtain a target electricity consumption prediction result;
and performing energy scheduling optimization on the target area according to the target power consumption prediction result and the target energy prediction result.
Further, the scheduling execution module M700 in the system is further configured to:
preprocessing the historical electricity utilization data to obtain a historical electricity utilization data processing result;
acquiring a preset power utilization period, and performing period division on the historical power utilization data processing result according to the preset power utilization period to obtain a period division result, wherein the period division result comprises a plurality of periods;
extracting any one of the multiple periods, and performing stage division on the any one period to obtain a stage division result, wherein the stage division result comprises multiple stages;
sequentially carrying out power consumption analysis on each stage in the plurality of stages, and calculating to obtain a plurality of power consumption according to analysis results;
and constructing the electricity utilization stage-electricity consumption list according to the mapping relation between the plurality of stages and the plurality of electricity consumptions.
Further, the data decomposition module M300 in the system is further configured to:
extracting the time identification of the target training data and generating a target training data sequence;
obtaining target white noise, and adding the target white noise to the target training data sequence to obtain a new target training data sequence;
obtaining a preset iteration result, and carrying out iterative decomposition on the new target training data sequence until the new target training data sequence meets the preset iteration result to obtain a plurality of target residual error sequences;
and taking the plurality of target residual difference sequences as the plurality of target subsequences.
Further, the data decomposition module M300 in the system is further configured to:
analyzing to obtain a plurality of local extreme values in the new target training data sequence;
according to the plurality of local extrema, sequentially establishing an upper envelope line and a lower envelope line of the new target training data sequence;
calculating to obtain a mean sequence according to the upper envelope curve and the lower envelope curve, wherein a calculation formula of the mean sequence is as follows:
wherein, the
Refers to the mean sequence, the
Is the upper layer envelope of the above
Refers to the lower envelope, and d refers to the new targetTraining a data sequence;
performing residual error operation on the new target training data sequence and the mean value sequence to obtain a target residual error sequence;
and continuing to iterate until the preset iteration result is met, and generating the target residual error sequences.
Further, the scheduling execution module M700 in the system is further configured to:
extracting a time identifier of the target energy information to obtain target time;
sequentially analyzing the prediction results of the long-short term memory prediction model and the generalized regression neural network prediction model based on the target time and the target test data to respectively obtain the prediction results of the long-short term memory prediction model and the generalized regression neural network prediction model;
according to a dynamic adaptive variable weight optimization theory, sequentially analyzing the weight coefficients of the long-short term memory prediction model and the generalized regression neural network prediction model to respectively obtain a long-short term memory prediction model coefficient and a generalized regression neural network prediction model coefficient;
and calculating to obtain the target energy prediction result according to the long-short term memory prediction model coefficient, the generalized regression neural network prediction model coefficient, the long-short term memory prediction model prediction result and the generalized regression neural network prediction model prediction result.
Further, the scheduling execution module M700 in the system is further configured to:
obtaining an actual energy observation result of the target time;
and calculating to obtain a relative prediction error of the combined prediction model according to the target energy prediction result and the actual energy observation result, wherein a calculation formula of the relative prediction error is as follows:
wherein, the
Is the predicted result of the target energy, the
Means the actual energy observation result, the t means the target time, the
Is the long-short term memory prediction model coefficient, the
Refers to the coefficient of the generalized regression neural network prediction model, the
Is the prediction result of the long-short term memory prediction model, and the
The prediction result of the generalized regression neural network prediction model is referred to.
In the present specification, the embodiments are described in a progressive manner, and each embodiment focuses on differences from other embodiments, and the aforementioned regional microgrid interconnection optimization method based on power coordination in the first embodiment of fig. 1 and the specific example are also applicable to the regional microgrid interconnection optimization system based on power coordination in this embodiment, and through the foregoing detailed description of the regional microgrid interconnection optimization method based on power coordination, those skilled in the art can clearly know a regional microgrid interconnection optimization system based on power coordination in this embodiment, so for the sake of brevity of the description, detailed descriptions are not repeated here. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the present invention and its equivalent technology, it is intended that the present invention also include such modifications and variations.
Claims (8)
1. A regional micro-grid interconnection optimization method based on electric energy mutual aid is characterized by comprising the following steps:
acquiring initial energy information of a target area, and preprocessing the initial energy information to obtain target energy information, wherein the target energy information is information with a time identifier;
dividing the target energy information to obtain a target energy information division result, wherein the target energy information division result comprises target training data and target test data;
decomposing the target training data by using a set empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences;
extracting any two target subsequences in the plurality of target subsequences, and respectively recording the target subsequences as a first target subsequence and a second target subsequence;
training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model;
based on the long-short term memory prediction model and the generalized regression neural network prediction model, a combined prediction model is obtained by combining dynamic adaptive variable weight optimization theory analysis;
and processing the target test data through the combined prediction model to obtain a target energy prediction result, and performing energy scheduling optimization on the target area according to the target energy prediction result.
2. The method of claim 1, wherein the optimizing energy scheduling for the target area based on the target energy prediction comprises:
collecting historical electricity utilization data of the target area;
analyzing the historical electricity utilization data, and obtaining an electricity utilization stage-electricity consumption list according to an analysis result;
extracting a time identifier in the target energy information to obtain a target time stage;
traversing the target time stage in the electricity utilization stage-electricity consumption list, and analyzing to obtain a target electricity consumption prediction result;
and performing energy scheduling optimization on the target area according to the target power consumption prediction result and the target energy prediction result.
3. The method of claim 2, wherein obtaining a list of electricity usage phases-electricity usage based on the analysis comprises:
preprocessing the historical electricity utilization data to obtain a historical electricity utilization data processing result;
obtaining a preset power utilization period, and periodically dividing the historical power utilization data processing result according to the preset power utilization period to obtain a periodic division result, wherein the periodic division result comprises a plurality of periods;
extracting any one of the multiple periods, and performing stage division on the any one period to obtain a stage division result, wherein the stage division result comprises multiple stages;
sequentially carrying out power consumption analysis on each stage in the plurality of stages, and calculating to obtain a plurality of power consumption according to analysis results;
and constructing the electricity utilization stage-electricity consumption list according to the mapping relation between the plurality of stages and the plurality of electricity consumptions.
4. The method according to claim 1, wherein decomposing the target training data using a set empirical mode decomposition principle to obtain a target training data decomposition result comprises:
extracting the time identification of the target training data and generating a target training data sequence;
obtaining target white noise, and adding the target white noise to the target training data sequence to obtain a new target training data sequence;
obtaining a preset iteration result, and carrying out iterative decomposition on the new target training data sequence until the new target training data sequence meets the preset iteration result to obtain a plurality of target residual error sequences;
and taking the plurality of target residual difference sequences as the plurality of target subsequences.
5. The method of claim 4, wherein the iteratively decomposing the new target training data sequence until the preset iteration result is satisfied to obtain a plurality of target residual error sequences comprises:
analyzing to obtain a plurality of local extreme values in the new target training data sequence;
according to the plurality of local extreme values, sequentially establishing an upper envelope and a lower envelope of the new target training data sequence;
calculating to obtain a mean sequence according to the upper envelope and the lower envelope, wherein a calculation formula of the mean sequence is as follows:
wherein, the
Refers to the mean sequence, the
Is the upper layer envelope of the above
Refers to the lower envelope, and d refers to the new target training data sequence;
performing residual error operation on the new target training data sequence and the mean value sequence to obtain a target residual error sequence;
and continuing to iterate until the preset iteration result is met, and generating the target residual error sequences.
6. The method of claim 2, further comprising, prior to the optimizing energy scheduling for the target area based on the target power usage prediction and the target energy prediction, performing energy scheduling for the target area:
extracting a time identifier of the target energy information to obtain target time;
based on the target time and the target test data, sequentially analyzing the prediction results of the long-short term memory prediction model and the generalized regression neural network prediction model to respectively obtain the prediction result of the long-short term memory prediction model and the prediction result of the generalized regression neural network prediction model;
according to a dynamic adaptive variable weight optimization theory, sequentially analyzing the weight coefficients of the long-short term memory prediction model and the generalized regression neural network prediction model to respectively obtain a long-short term memory prediction model coefficient and a generalized regression neural network prediction model coefficient;
and calculating to obtain the target energy prediction result according to the long-short term memory prediction model coefficient, the generalized regression neural network prediction model coefficient, the long-short term memory prediction model prediction result and the generalized regression neural network prediction model prediction result.
7. The method of claim 6, further comprising:
obtaining an actual energy observation result of the target time;
and calculating to obtain a relative prediction error of the combined prediction model according to the target energy prediction result and the actual energy observation result, wherein a calculation formula of the relative prediction error is as follows:
wherein, the
Is the target energy prediction result, the
Is the actual energy observation, t is the target time, the
Is the long-short term memory prediction model coefficient, the
Refers to the coefficient of the generalized regression neural network prediction model, the
Is the prediction result of the long-short term memory prediction model, and the
The method refers to the prediction result of the generalized regression neural network prediction model.
8. A regional micro-grid interconnection optimization system based on mutual electric energy economy is characterized by comprising:
the system comprises an information acquisition module, a data processing module and a data processing module, wherein the information acquisition module is used for acquiring initial energy information of a target area and preprocessing the initial energy information to obtain target energy information, and the target energy information is information with a time identifier;
the information dividing module is used for dividing the target energy information to obtain a target energy information dividing result, wherein the target energy information dividing result comprises target training data and target test data;
the data decomposition module is used for decomposing the target training data by utilizing a set empirical mode decomposition principle to obtain a target training data decomposition result, wherein the target training data decomposition result comprises a plurality of target subsequences;
a sequence obtaining module, configured to extract any two target subsequences from the multiple target subsequences, and record the extracted two target subsequences as a first target subsequence and a second target subsequence;
the model training module is used for training the first target subsequence by using a long-short term memory model principle to obtain a long-short term memory prediction model, and training the second target subsequence by using a generalized recurrent neural network principle to obtain a generalized recurrent neural network prediction model;
the model obtaining module is used for obtaining a combined prediction model by combining dynamic adaptive variable weight optimization theory analysis based on the long-short term memory prediction model and the generalized regression neural network prediction model;
and the scheduling execution module is used for processing the target test data through the combined prediction model to obtain a target energy prediction result and performing energy scheduling optimization on the target area according to the target energy prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211420989.2A CN115471017A (en) | 2022-11-15 | 2022-11-15 | Regional microgrid interconnection optimization method and system based on mutual power assistance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211420989.2A CN115471017A (en) | 2022-11-15 | 2022-11-15 | Regional microgrid interconnection optimization method and system based on mutual power assistance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115471017A true CN115471017A (en) | 2022-12-13 |
Family
ID=84338188
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211420989.2A Pending CN115471017A (en) | 2022-11-15 | 2022-11-15 | Regional microgrid interconnection optimization method and system based on mutual power assistance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115471017A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109462231A (en) * | 2018-11-15 | 2019-03-12 | 合肥工业大学 | Optimization of Load Dispatching method, system and the storage medium of resident's micro-capacitance sensor |
| CN113435923A (en) * | 2021-06-15 | 2021-09-24 | 北京百度网讯科技有限公司 | Power consumption prediction method and device and electronic equipment |
| CN113869556A (en) * | 2021-05-19 | 2021-12-31 | 国家电网有限公司 | Power consumption prediction method, device and equipment |
| CN113869590A (en) * | 2021-09-30 | 2021-12-31 | 上海许继电气有限公司 | EEMD-LSTM-based regional energy internet load prediction method and system |
-
2022
- 2022-11-15 CN CN202211420989.2A patent/CN115471017A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109462231A (en) * | 2018-11-15 | 2019-03-12 | 合肥工业大学 | Optimization of Load Dispatching method, system and the storage medium of resident's micro-capacitance sensor |
| US20200161867A1 (en) * | 2018-11-15 | 2020-05-21 | Hefei University Of Technology | Method, system and storage medium for load dispatch optimization for residential microgrid |
| CN113869556A (en) * | 2021-05-19 | 2021-12-31 | 国家电网有限公司 | Power consumption prediction method, device and equipment |
| CN113435923A (en) * | 2021-06-15 | 2021-09-24 | 北京百度网讯科技有限公司 | Power consumption prediction method and device and electronic equipment |
| CN113869590A (en) * | 2021-09-30 | 2021-12-31 | 上海许继电气有限公司 | EEMD-LSTM-based regional energy internet load prediction method and system |
Non-Patent Citations (1)
| Title |
|---|
| 杨磊等: "基于集合经验模态分解和套索算法的短期风速组合变权预测模型研究", 《电力系统保护与控制》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110610280B (en) | Short-term prediction method, model, device and system for power load | |
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| Ahmad et al. | A data-driven deep sequence-to-sequence long-short memory method along with a gated recurrent neural network for wind power forecasting | |
| Chang et al. | A memory-network based solution for multivariate time-series forecasting | |
| Niska et al. | Data analytics approach to create waste generation profiles for waste management and collection | |
| CN108388962B (en) | Wind power prediction system and method | |
| Liu et al. | Heating load forecasting for combined heat and power plants via strand-based LSTM | |
| CN115293326A (en) | Training method and device of power load prediction model and power load prediction method | |
| Massaoudi et al. | Short-term electric load forecasting based on data-driven deep learning techniques | |
| Brucke et al. | A non-intrusive load monitoring approach for very short-term power predictions in commercial buildings | |
| Rahman et al. | Deep learning based ensemble method for household energy demand forecasting of smart home | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| CN111754029A (en) | Community load prediction system | |
| CN115545333A (en) | Method for predicting load curve of multi-load daily-type power distribution network | |
| Cai et al. | Gray wolf optimization-based wind power load mid-long term forecasting algorithm | |
| CN115238854A (en) | Short-term load prediction method based on TCN-LSTM-AM | |
| Johannesen et al. | Comparing recurrent neural networks using principal component analysis for electrical load predictions | |
| Al-Ja’afreh et al. | An enhanced CNN-LSTM based multi-stage framework for PV and load short-term forecasting: DSO scenarios | |
| Eren et al. | A comprehensive review on deep learning approaches for short-term load forecasting | |
| Stamatescu et al. | Reporting interval impact on deep residential energy measurement prediction | |
| Zi-hao et al. | A thermal energy usage prediction method for electric thermal storage heaters based on deep learning | |
| CN115471017A (en) | Regional microgrid interconnection optimization method and system based on mutual power assistance | |
| CN113177366B (en) | Comprehensive energy system planning method and device and terminal equipment | |
| Rostum et al. | A combined effective time series model based on clustering and whale optimization algorithm for forecasting smart meters electricity consumption | |
| Deng et al. | Medium-term rolling load forecasting based on seasonal decomposition and long short-term memory 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20221213 |