CN113869590A - EEMD-LSTM-based regional energy internet load prediction method and system - Google Patents
EEMD-LSTM-based regional energy internet load prediction method and system Download PDFInfo
- Publication number
- CN113869590A CN113869590A CN202111161973.XA CN202111161973A CN113869590A CN 113869590 A CN113869590 A CN 113869590A CN 202111161973 A CN202111161973 A CN 202111161973A CN 113869590 A CN113869590 A CN 113869590A
- Authority
- CN
- China
- Prior art keywords
- data
- energy internet
- regional energy
- prediction
- eemd
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 55
- 238000005070 sampling Methods 0.000 claims abstract description 29
- 238000013528 artificial neural network Methods 0.000 claims abstract description 18
- 230000007774 longterm Effects 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 41
- 238000010606 normalization Methods 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 7
- 238000009825 accumulation Methods 0.000 claims description 6
- 238000013277 forecasting method Methods 0.000 claims 4
- 238000004422 calculation algorithm Methods 0.000 abstract description 5
- 238000010248 power generation Methods 0.000 abstract description 5
- 238000013135 deep learning Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 210000004027 cell Anatomy 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a regional energy Internet load prediction method and a system based on EEMD-LSTM, wherein the method comprises the following steps: acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads; performing component decomposition on historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments; and predicting a plurality of data components through the long-term and short-term neural network, and accumulating the prediction results to obtain the prediction result of the regional energy Internet load. Aiming at the problem of accuracy of short-term load prediction of power generation of the regional energy Internet, by combining the randomness characteristics of the regional energy Internet and applying a deep learning algorithm to predict regional energy Internet load data on the basis of analyzing load characteristics, uncertain factors in original signals are effectively overcome, and the requirement of the short-term load prediction of the regional energy Internet is met.
Description
Technical Field
The invention relates to the technical field of power equipment detection, in particular to a regional energy Internet load prediction method and system based on EEMD-LSTM.
Background
The short-term load prediction is an important component of power grid planning and is a precondition for ensuring the safe and efficient operation of a power grid, and the short-term load prediction is used as an important index to reflect the power generation state of the regional energy Internet; load prediction is carried out on the system at the future moment through the load state of each sub-device of the regional energy Internet system, so that the overall operation reliability of the transformer substation is improved, and the aims of safe and economic operation of a power grid are finally achieved.
Aiming at the research problem of load prediction, a neural network is mostly adopted to process the nonlinear change problem of short-term load prediction data at present, however, most prediction models have more consideration factors, complex model construction and longer training time, and cannot acquire load prediction result data in time, so that difficulty is brought to practical application; the accuracy of a single neural network is not ideal enough, the optimal effect cannot be achieved frequently, and the relatively stable prediction accuracy of the whole system cannot be ensured.
Disclosure of Invention
The embodiment of the invention aims to provide an EEMD-LSTM-based regional energy Internet load prediction method and system, aiming at the problem of accuracy of regional energy Internet power generation short-term load prediction, by combining the randomness characteristic of the regional energy Internet and applying a deep learning algorithm to predict regional energy Internet load data on the basis of analyzing load characteristics, the uncertain factors in original signals are effectively overcome, the requirement of regional energy Internet short-term load prediction is met, and the smooth operation of an energy Internet power system is improved.
In order to solve the above technical problem, a first aspect of an embodiment of the present invention provides a method for predicting a regional energy internet load based on EEMD-LSTM, including the following steps:
acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads;
performing component decomposition on the historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments;
and predicting the data components through a long-term and short-term neural network, and accumulating the prediction results to obtain the prediction result of the regional energy Internet load.
Further, the component decomposition of the historical operating data at different time instants by a collective empirical mode decomposition method includes:
preprocessing historical operating data of the regional energy Internet load sub-equipment, and then adding a random white noise sequence of a first preset value group to obtain a to-be-processed sequence added with the white noise sequence;
performing component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling point;
and accumulating the IMF components and calculating an average value to obtain the data components of the sampling points.
Further, the standard deviation of the random white noise sequence of the first preset value set is a second preset value.
Further, the predicting the number of data components by the long-short term neural network comprises:
respectively carrying out normalization processing on the plurality of data components;
predicting the data component after the normalization processing through an LSTM model;
and performing inverse normalization processing on the prediction result.
Further, the preprocessing the historical operation data of the regional energy internet load sub-equipment comprises:
and screening the historical operating data of the sub-equipment to remove unreasonable data.
Further, after the prediction result of the regional energy internet load is obtained, the method further includes:
and comparing the prediction result with the historical operation data, and carrying out error analysis.
Accordingly, a second aspect of the embodiments of the present invention provides an EEMD-LSTM-based regional energy internet load prediction system, including:
the data acquisition module is used for acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads;
the data decomposition module is used for carrying out component decomposition on the historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments;
and the data prediction module is used for predicting the data components through a long-term and short-term neural network and accumulating prediction results to obtain the prediction result of the regional energy Internet load.
Further, the data decomposition module comprises:
the white noise processing unit is used for preprocessing the historical operating data of the regional energy Internet load sub-equipment, and then adding a random white noise sequence of a first preset value group to obtain a to-be-processed sequence added with the white noise sequence;
the data decomposition unit is used for carrying out component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling points;
and the data accumulation unit is used for accumulating the IMF components and calculating an average value to obtain the data components of the sampling points.
Further, the standard deviation of the random white noise sequence of the first preset value set is a second preset value.
Further, the data prediction module comprises:
a normalization processing unit, configured to perform normalization processing on the plurality of data components respectively; a data prediction unit for predicting the normalized data component by an LSTM model;
and the anti-normalization processing unit is used for carrying out anti-normalization processing on the prediction result.
Further, the white noise processing unit is further configured to perform data screening on the historical operating data of the sub-device to remove unreasonable data.
Further, the system for predicting the internet load of the EEMD-LSTM-based regional energy source further comprises:
and the error analysis module is used for comparing the prediction result with the historical operating data and carrying out error analysis.
Accordingly, a third aspect of an embodiment of the present invention provides an electronic device, including: at least one processor; and a memory coupled to the at least one processor; wherein the memory stores instructions executable by the one processor to cause the at least one processor to perform the method for EEMD-LSTM based regional energy Internet load prediction as described above.
Accordingly, a fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having stored thereon computer instructions, which when executed by a processor, implement the above described EEMD-LSTM based regional energy Internet load prediction method.
The technical scheme of the embodiment of the invention has the following beneficial technical effects:
aiming at the problem of accuracy of short-term load forecasting of power generation of the regional energy Internet, by combining the randomness characteristics of the regional energy Internet and applying a deep learning algorithm to forecast the load data of the regional energy Internet on the basis of analyzing the load characteristics, uncertain factors in original signals are effectively overcome, the requirement of the short-term load forecasting of the regional energy Internet is met, and the stable operation of an energy Internet power system is improved.
Drawings
FIG. 1 is a flowchart of a method for predicting the Internet load of regional energy sources based on EEMD-LSTM according to an embodiment of the present invention;
FIG. 2 is a flow chart of regional energy Internet prediction provided by an embodiment of the invention;
FIG. 3 is an exploded view of an EEMD process provided by an embodiment of the present invention;
FIG. 4 is a schematic diagram of a prediction process for a single LSTM neuron according to an embodiment of the present invention;
FIG. 5 is a block diagram of a system for forecasting the internet load of the regional energy based on EEMD-LSTM according to the embodiment of the present invention;
FIG. 6 is a block diagram of a data decomposition module provided by an embodiment of the present invention;
fig. 7 is a block diagram of a data prediction module according to an embodiment of the present invention.
Reference numerals:
1. the device comprises a data acquisition module, a data decomposition module, a white noise processing unit, a data decomposition unit, a data accumulation unit, a data prediction module, a normalization processing unit, a data prediction unit, a data analysis module and a data analysis module, wherein the white noise processing unit is 21, the data analysis module is 22, the data decomposition unit is 23, the data accumulation unit is 3, the data prediction module is 31, the normalization processing unit is 32, the data prediction unit is 33, the anti-normalization processing unit is 4, and the error analysis module is 4.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings in conjunction with the following detailed description. It should be understood that the description is intended to be exemplary only, and is not intended to limit the scope of the present invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
Referring to fig. 1 and 2, a first aspect of an embodiment of the present invention provides a method for predicting internet load of regional energy based on EEMD-LSTM, including the following steps:
s100, historical operating data of a plurality of sub-devices in regional energy Internet loads are obtained.
S200, performing component decomposition on the historical operating data of the sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments.
Specifically, referring to fig. 3, in step S200, component decomposition is performed on historical operating data at different times by using a collective empirical mode decomposition method, and the method further includes the steps of:
s210, preprocessing historical operating data of the regional energy Internet load sub-equipment, and then adding a random white noise sequence of a first preset value group to obtain a to-be-processed sequence added with the white noise sequence; .
Further, the standard deviation of the random white noise sequence of the first preset value group is a second preset value.
And S220, performing component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling point.
And S230, accumulating the IMF components and calculating an average value to obtain data components of the sampling points.
White noise is added into an original signal to be decomposed by an ensemble empirical mode decomposition method, signals with different time scales are automatically distributed to a proper reference scale, and a group of components with respective rules are generated. And due to the zero-mean characteristic of the noise, after several averaging, the added noise can be mutually cancelled, and the result of the integrated mean can be used as the final result. In the model building process, an original load sequence of the energy Internet is added into a random white noise sequence to obtain a sequence to be processed after a noise signal is added, 100 groups of random white noise sequences with difference are added by using an EMD method, and the standard deviation is taken as 0.2. And decomposing the original load sequence of the regional energy Internet once, adding a white noise sequence with the same root mean square, and performing EMD (empirical mode decomposition) for N times to reconstruct a signal to achieve a better effect.
And S300, predicting a plurality of data components through the long-term and short-term neural network, and accumulating the prediction results to obtain the prediction result of the regional energy Internet load.
Further, referring to fig. 4, in step S300, predicting a plurality of data components through the long-term and short-term neural network includes:
s310, normalization processing is respectively carried out on the data components.
And S320, predicting the normalized data component through the LSTM model.
And S330, performing inverse normalization processing on the prediction result.
In order to solve the problem that the gradient generated by the recurrent neural network RNN when processing long-term dependence disappears, an improved algorithm LSTM based on the recurrent neural network RNN is provided, and the method introduces the concept of grid and storage unit in each hidden layer. An LSTM memory block mainly comprises an input gate, a forgetting gate, an output gate and a self-connected memory unit. The input gate controls the activation input to the memory cell and can calculate the load power data input value, the power output value at the previous time and the output value of the memory channel at the previous time. The forgetting gate can help a network to forget past input data and reset the state of the memory cell, and the forgetting gate and the output gate act together to determine the state of the memory unit at the current moment and finally determine the output power value of the output gate. In addition, the application of the multiplication gate can ensure that the storage unit can access and store information for a long time. Each cell contains memory cells that store information from a previous time. Therefore, the LSTM as a deep neural network has more hidden layers, and the accuracy advantage of calculation is increased.
The model is based on an EEMD-LSTM regional energy Internet system day-ahead hour load prediction model. For the historical regional energy Internet hourly output sequence, firstly, a group of components are obtained by decomposition through an EEMD method, data of each component is normalized to be in a [0, 1] range, then a corresponding LSTM model is established for prediction, results are subjected to inverse normalization, each component, namely the hourly load prediction value of the energy Internet system in the region before the day, is superposed, and finally, the output prediction result is compared with the original load data, and error analysis is carried out.
In addition, the historical operation data of the regional energy Internet load sub-equipment is preprocessed, and the preprocessing comprises the following steps: and (4) screening the historical operating data of the sub-equipment to remove unreasonable data.
Besides, after the prediction result of the regional energy internet load is obtained, the method further comprises the following steps: and comparing the prediction result with historical operation data, and performing error analysis.
The data prediction of the regional energy Internet load comes from the operation data of each sub-device to which the regional energy Internet load belongs, and the output data of the sub-device system can be cleaned by carrying out big data processing on the historical data of the sub-devices and screening out unreasonable data. Processing the obtained data by using an Ensemble Empirical Mode Decomposition (EEMD) method, performing component decomposition on the data at each moment, predicting each component by using a long-short-term neural network (LSTM), and accumulating the values of each component to obtain a prediction result.
According to the technical scheme, the EEMD method is used for overcoming uncertain factors in original load data signals, when data preprocessing is carried out on regional energy Internet load data, due to the fact that the regional energy Internet load is high in randomness, low in similarity of historical load curves, basically no rules can be followed, the user capacity is limited, load characteristics among users are not obvious enough, and modal aliasing can be effectively avoided by decomposing the original data through the EEMD (ensemble empirical mode decomposition) method.
In addition, compared with the traditional neural network, the same data sampling points are used for modeling, the fitting condition of the prediction result curve of each model is compared, and the result is subjected to error analysis; aiming at the irregular fluctuation characteristic of the short-term load sequence of the regional energy Internet system, the EEMD-LSTM can predict the load more accurately; the EEMD-LSTM can basically meet the requirement of short-term load prediction of the regional energy Internet and provide certain reference for smooth operation of an energy Internet power system. And the LSTM storage unit is used for learning and retaining useful information in the historical data of the power load for a long time, so that reference is made for improving the prediction accuracy of the regional energy Internet load.
Accordingly, referring to fig. 5, a second aspect of the embodiment of the present invention provides an EEMD-LSTM-based regional energy internet load prediction system, including: the device comprises a data acquisition module 1, a data decomposition module 2 and a data prediction module 3.
Wherein: the data acquisition module 1 is used for acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads; the data decomposition module 2 is used for performing component decomposition on historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components corresponding to the sampling points at different moments respectively; the data prediction module 3 is used for predicting a plurality of data components through a long-term and short-term neural network, and accumulating prediction results to obtain a prediction result of the regional energy Internet load.
Specifically, referring to fig. 6, the data decomposition module 2 includes: a white noise processing unit 21, a data decomposition unit 22, and a data accumulation unit 23.
The white noise processing unit 21 is used for preprocessing historical operation data of the regional energy internet load sub-equipment, and then adding a random white noise sequence of the first preset value group to obtain a to-be-processed sequence added with the white noise sequence; the data decomposition unit 22 is configured to perform component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling points; the data accumulation unit 23 is configured to accumulate a plurality of IMF components and calculate an average value to obtain a data component of a sampling point.
Further, the standard deviation of the random white noise sequence of the first preset value group is a second preset value.
Specifically, referring to fig. 7, the data prediction module 3 includes: a normalization processing unit 31, a data prediction unit 32, and an inverse normalization processing unit 33.
The normalization processing unit 31 is configured to perform normalization processing on the plurality of data components respectively; the data prediction unit 32 is configured to predict the normalized data component through an LSTM model; the denormalization processing unit 33 is configured to perform denormalization processing on the prediction result.
Further, the white noise processing unit 21 is further configured to perform data screening on historical operating data of the sub-device to remove unreasonable data.
Further, the EEMD-LSTM-based regional energy Internet load prediction system further comprises: and the error analysis module 5 is used for comparing the prediction result with the historical operation data and carrying out error analysis.
Accordingly, a third aspect of an embodiment of the present invention provides an electronic device, including: at least one processor; and a memory coupled to the at least one processor; wherein the memory stores instructions executable by a processor to cause the at least one processor to perform the method for EEMD-LSTM based regional energy Internet load prediction as described above.
Accordingly, a fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having stored thereon computer instructions, which when executed by a processor, implement the above described EEMD-LSTM based regional energy Internet load prediction method.
The embodiment of the invention aims to protect an EEMD-LSTM-based regional energy Internet load prediction method and system, wherein the method comprises the following steps: acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads; performing component decomposition on historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments; and predicting a plurality of data components through the long-term and short-term neural network, and accumulating the prediction results to obtain the prediction result of the regional energy Internet load. The technical scheme has the following effects:
aiming at the problem of accuracy of short-term load forecasting of power generation of the regional energy Internet, by combining the randomness characteristics of the regional energy Internet and applying a deep learning algorithm to forecast the load data of the regional energy Internet on the basis of analyzing the load characteristics, uncertain factors in original signals are effectively overcome, the requirement of the short-term load forecasting of the regional energy Internet is met, and the stable operation of an energy Internet power system is improved.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (12)
1. A regional energy Internet load prediction method based on EEMD-LSTM is characterized by comprising the following steps:
acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads;
performing component decomposition on the historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments;
and predicting the data components through a long-term and short-term neural network, and accumulating the prediction results to obtain the prediction result of the regional energy Internet load.
2. The EEMD-LSTM based regional energy Internet load forecasting method of claim 1, wherein said component decomposition of said historical operating data at different times by collective empirical mode decomposition comprises:
preprocessing historical operating data of the regional energy Internet load sub-equipment, and then adding a random white noise sequence of a first preset value group to obtain a to-be-processed sequence added with the white noise sequence;
performing component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling point;
and accumulating the IMF components and calculating an average value to obtain the data components of the sampling points.
3. The EEMD-LSTM based regional energy Internet load forecasting method of claim 2,
and the standard deviation of the random white noise sequence of the first preset value group is a second preset value.
4. The EEMD-LSTM based regional energy Internet load forecasting method of claim 1, wherein said forecasting said data components via long and short term neural networks comprises:
respectively carrying out normalization processing on the plurality of data components;
predicting the data component after the normalization processing through an LSTM model;
and performing inverse normalization processing on the prediction result.
5. The EEMD-LSTM based regional energy Internet load forecasting method of claim 2, wherein pre-processing historical operational data of the regional energy Internet load sub-equipment comprises:
and screening the historical operating data of the sub-equipment to remove unreasonable data.
6. The method of claim 1, wherein the obtaining of the prediction result of the regional energy internet load further comprises:
and comparing the prediction result with the historical operation data, and carrying out error analysis.
7. An EEMD-LSTM-based regional energy Internet load prediction system, comprising:
the data acquisition module is used for acquiring historical operating data of a plurality of sub-devices in regional energy Internet loads;
the data decomposition module is used for carrying out component decomposition on the historical operating data of sampling points at different moments by an ensemble empirical mode decomposition method to obtain a plurality of data components respectively corresponding to the sampling points at different moments;
and the data prediction module is used for predicting the data components through a long-term and short-term neural network and accumulating prediction results to obtain the prediction result of the regional energy Internet load.
8. The EEMD-LSTM based regional energy Internet load prediction system of claim 7, wherein said data decomposition module comprises:
the white noise processing unit is used for preprocessing the historical operating data of the regional energy Internet load sub-equipment, and then adding a random white noise sequence of a first preset value group to obtain a to-be-processed sequence added with the white noise sequence;
the data decomposition unit is used for carrying out component decomposition on the sequence to be processed by an ensemble empirical mode decomposition method to obtain a plurality of IMF components of the sampling points;
and the data accumulation unit is used for accumulating the IMF components and calculating an average value to obtain the data components of the sampling points.
9. The EEMD-LSTM based regional energy Internet load prediction system of claim 8,
and the standard deviation of the random white noise sequence of the first preset value group is a second preset value.
10. The EEMD-LSTM based regional energy Internet load prediction system of claim 7, wherein said data prediction module comprises:
a normalization processing unit, configured to perform normalization processing on the plurality of data components respectively; a data prediction unit for predicting the normalized data component by an LSTM model;
and the anti-normalization processing unit is used for carrying out anti-normalization processing on the prediction result.
11. The EEMD-LSTM based regional energy Internet load prediction system of claim 8,
and the white noise processing unit is used for screening the historical operating data of the sub-equipment to remove unreasonable data.
12. The EEMD-LSTM based regional energy Internet load prediction system of claim 7, further comprising:
and the error analysis module is used for comparing the prediction result with the historical operating data and carrying out error analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111161973.XA CN113869590A (en) | 2021-09-30 | 2021-09-30 | EEMD-LSTM-based regional energy internet load prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111161973.XA CN113869590A (en) | 2021-09-30 | 2021-09-30 | EEMD-LSTM-based regional energy internet load prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113869590A true CN113869590A (en) | 2021-12-31 |
Family
ID=79001349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111161973.XA Pending CN113869590A (en) | 2021-09-30 | 2021-09-30 | EEMD-LSTM-based regional energy internet load prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113869590A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115099502A (en) * | 2022-06-29 | 2022-09-23 | 四川大学 | Short-term power load prediction method based on inter-user power consumption behavior similarity |
| CN115471017A (en) * | 2022-11-15 | 2022-12-13 | 浙江大学 | Regional microgrid interconnection optimization method and system based on mutual power assistance |
-
2021
- 2021-09-30 CN CN202111161973.XA patent/CN113869590A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115099502A (en) * | 2022-06-29 | 2022-09-23 | 四川大学 | Short-term power load prediction method based on inter-user power consumption behavior similarity |
| CN115471017A (en) * | 2022-11-15 | 2022-12-13 | 浙江大学 | Regional microgrid interconnection optimization method and system based on mutual power assistance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110969285B (en) | Prediction model training method, prediction device, prediction equipment and medium | |
| CN110348622B (en) | Temperature prediction method and system based on machine learning and electronic equipment | |
| CN107506868B (en) | Method and device for predicting short-time power load | |
| CN113869590A (en) | EEMD-LSTM-based regional energy internet load prediction method and system | |
| CN109687447B (en) | Electric power energy consumption prediction method and device | |
| Zhang et al. | Interval prediction of ultra-short-term photovoltaic power based on a hybrid model | |
| CN116383773A (en) | Data center energy efficiency anomaly detection method, system and medium based on self-adaptive prediction interval | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| CN111697560B (en) | Method and system for predicting load of power system based on LSTM | |
| CN114548586A (en) | Short-term power load prediction method and system based on hybrid model | |
| CN113988398A (en) | Wind turbine generator power prediction method and device, electronic equipment and storage medium | |
| CN116757378A (en) | Demand response evaluation model training method, application evaluation method and related equipment | |
| Wang et al. | Multienergy load forecasting for regional integrated energy systems considering multienergy coupling of variation characteristic curves | |
| CN116739172A (en) | Method and device for ultra-short-term prediction of offshore wind power based on climbing identification | |
| CN115222138A (en) | Photovoltaic short-term power interval prediction method based on EEMD-LSTM microgrid | |
| Sun et al. | Short-term power load prediction based on VMD-SG-LSTM | |
| Yang et al. | Software defect prediction: an ensemble learning approach | |
| CN116859255A (en) | Method, device, equipment and medium for predicting state of health of energy storage battery | |
| CN111080037A (en) | Short-term power load prediction method and device based on deep neural network | |
| CN110659442B (en) | System, method and device for short-term prediction of data of system, and storage medium | |
| CN115759343A (en) | E-LSTM-based user electric quantity prediction method and device | |
| CN115907194A (en) | Photovoltaic power generation power prediction method, system and product | |
| CN115564115A (en) | Wind power plant power prediction method and related equipment | |
| CN110689168B (en) | Power load prediction method and prediction device | |
| Wang et al. | Short-term probability density function forecasting of industrial loads based on ConvLSTM-MDN |
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 |