CN115310355A - Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system - Google Patents
Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system Download PDFInfo
- Publication number
- CN115310355A CN115310355A CN202210924624.7A CN202210924624A CN115310355A CN 115310355 A CN115310355 A CN 115310355A CN 202210924624 A CN202210924624 A CN 202210924624A CN 115310355 A CN115310355 A CN 115310355A
- Authority
- CN
- China
- Prior art keywords
- load
- load prediction
- loads
- different loads
- historical
- 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 55
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 32
- 230000008878 coupling Effects 0.000 claims abstract description 26
- 238000010168 coupling process Methods 0.000 claims abstract description 26
- 238000005859 coupling reaction Methods 0.000 claims abstract description 26
- 230000004186 co-expression Effects 0.000 claims abstract description 16
- 238000003012 network analysis Methods 0.000 claims abstract description 15
- 238000013528 artificial neural network Methods 0.000 claims description 10
- 238000005065 mining Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000013277 forecasting method Methods 0.000 abstract description 2
- 239000011159 matrix material Substances 0.000 description 9
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000001816 cooling Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000010438 heat treatment Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000010170 biological method Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 108091008053 gene clusters Proteins 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- UEUXEKPTXMALOB-UHFFFAOYSA-J tetrasodium;2-[2-[bis(carboxylatomethyl)amino]ethyl-(carboxylatomethyl)amino]acetate Chemical compound [Na+].[Na+].[Na+].[Na+].[O-]C(=O)CN(CC([O-])=O)CCN(CC([O-])=O)CC([O-])=O UEUXEKPTXMALOB-UHFFFAOYSA-J 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/04—Power grid distribution networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/06—Power analysis or power optimisation
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Public Health (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Water Supply & Treatment (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
Abstract
The utility model provides a comprehensive energy system multi-element load forecasting method and system considering multi-energy coupling, which belongs to the technical field of the load forecasting of the comprehensive energy system in a park, and the scheme comprises the following steps: acquiring historical load prediction related data in the park comprehensive energy system, and performing corresponding pretreatment; based on the historical load prediction related data, adopting a weighted gene co-expression network analysis method to mine nonlinear relations among multiple loads and between the loads and the corresponding influence factors, and determining the influence factors strongly related to different loads; simultaneously inputting load historical data of different loads and characteristics corresponding to strongly related influence factors of the load historical data into a pre-trained load prediction model to obtain load prediction results corresponding to different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
Description
Technical Field
The disclosure belongs to the technical field of load prediction of a park comprehensive energy system, and particularly relates to a multivariate load prediction method and a multivariate load prediction system of the comprehensive energy system considering multi-energy coupling.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
In recent years, energy and environmental problems have attracted wide attention in all countries around the world and all communities, and an Integrated Energy System (IES) as a physical carrier of a new generation of Energy can realize cascade utilization among Energy sources, effectively improve the comprehensive utilization efficiency of the Energy sources and reduce the carbon emission of a park. A park-level Integrated Energy System (RIES) is small in user scale, poor in System robustness and large in load fluctuation. In the system energy distribution and utilization links, accurate prediction of loads is a key factor of operation scheduling and energy management of the combined cooling, heating and power system. Therefore, how to effectively solve the fluctuation and coupling of the multi-element load and accurately realize the prediction of the multi-energy load has become a current research focus.
The inventor finds that the short-term load prediction technology of the current power system is relatively mature, however, various energy sources in the RIES are deeply coupled and mutually influenced, the dynamic characteristics of the various energy sources are greatly different, and the load prediction method aiming at single energy source cannot be popularized to the field of multi-energy prediction; and the load prediction method of single energy can not accurately describe the strong coupling relation among multiple energy sources, so that the prediction result is greatly discounted. In addition, during load prediction, too many input factors can reduce the calculation efficiency of the model, and the existing method screens the input factors of the model by using a correlation coefficient method, but the method has a certain relation with the data quantity and cannot deeply mine the characteristic of nonlinearity among variables, so that important characteristics of the model are omitted, even non-relevant characteristics are selected, and thus, too much noise is generated, and the prediction accuracy is influenced.
Disclosure of Invention
In order to solve the problems, the present disclosure provides a multivariate load prediction method and system for a comprehensive energy system considering multi-energy coupling, in which a WGCNA is used to analyze the correlation between features and screen out appropriate input features; the technical scheme can effectively mine the nonlinear relevance among the multiple loads, can jointly predict the multiple loads, improves the prediction precision and efficiency, and is beneficial to improving the energy utilization rate of a park.
According to a first aspect of the embodiments of the present disclosure, there is provided a multivariate load prediction method for an integrated energy system considering multi-energy coupling, including:
acquiring historical load prediction related data in the park comprehensive energy system, and performing corresponding pretreatment;
based on the historical load prediction related data, adopting a weighted gene co-expression network analysis method to mine nonlinear relations among multiple loads and between the loads and the corresponding influence factors, and determining the influence factors strongly related to different loads;
simultaneously inputting load historical data of different loads and characteristics corresponding to strongly related influence factors of the load historical data into a pre-trained load prediction model to obtain load prediction results corresponding to different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
Further, the mining of the nonlinear relationship among the multiple loads and between the loads and the corresponding influence factors by using the weighted gene co-expression network analysis method specifically comprises the following steps: based on the obtained historical load prediction related data, influence factors which are strongly related to different loads are obtained through a weighted gene co-expression network analysis method, wherein the influence factors include but are not limited to meteorological factors and calendar factors which are strongly related to the loads.
Further, the load prediction model comprises an input layer, a sharing layer and an output layer which are sequentially connected, wherein the sharing layer adopts a BilSTM neural network.
Further, the corresponding preprocessing specifically includes removing abnormal values, filling missing values, and performing normalization processing on the data.
Further, the historical load prediction related data comprises corresponding load historical data under different loads, and corresponding meteorological factors and calendar factors of the load historical data.
Further, the different loads include a cold energy load, a heat energy load, and an electric energy load.
Further, for the pre-trained load prediction model, the performance of the pre-trained load prediction model is evaluated by using the root mean square error, the average absolute error or R2.
According to a second aspect of the embodiments of the present disclosure, there is provided a multivariate load prediction system of an integrated energy system considering multi-energy coupling, comprising:
the data acquisition unit is used for acquiring historical load prediction related data in the park integrated energy system and carrying out corresponding pretreatment;
the influence factor determining unit is used for mining nonlinear relations among multiple loads and between the loads and the corresponding influence factors by adopting a weighted gene co-expression network analysis method based on the historical load prediction related data, and determining the influence factors strongly related to different loads;
the load prediction unit is used for inputting the load historical data of different loads and the characteristics corresponding to the strongly related influence factors into a pre-trained load prediction model to obtain load prediction results corresponding to the different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the memory, wherein the processor executes the computer program to implement a method for predicting a multivariate load of an integrated energy system considering multi-energy coupling.
According to a fourth aspect of embodiments of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method for integrated energy system multivariate load prediction that considers multi-energy coupling.
Compared with the prior art, the beneficial effect of this disclosure is:
(1) The utility model provides a comprehensive energy system multi-load forecasting method and system considering multi-energy coupling, which adopts WGCNA to analyze the correlation between characteristics and screen out proper input characteristics; the technical scheme can effectively mine the nonlinear relevance among the multiple loads, can jointly predict the multiple loads, improves the prediction precision and efficiency, and is beneficial to improving the energy utilization rate of a park.
(2) According to the scheme disclosed by the invention, the load is considered as a time sequence and has autocorrelation, the internal rules of forward and backward historical data can be effectively learned by adopting a BilSTM algorithm, and the prediction precision is improved; meanwhile, multivariate load joint prediction is carried out based on the constructed MTL framework, cold, heat and electric loads can be simultaneously predicted, and the prediction efficiency is improved.
Advantages of additional aspects of the disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosure.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to be construed as limiting the disclosure.
Fig. 1 is a flow chart of the basic WGCNA analysis described in an embodiment of the present disclosure;
FIG. 2 is a flow chart of the execution of the BilSTM-MTL model according to an embodiment of the present disclosure;
fig. 3 is a flowchart of a multivariate load prediction method of an integrated energy system considering multi-energy coupling according to an embodiment of the present disclosure.
Detailed Description
The present disclosure is further described with reference to the following drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
Interpretation of terms:
module (Model): is a highly interconnected gene cluster;
adjacency matrix (Adjacency matrix): a matrix of weighted correlation values between genes;
TOM (polar overlay matrix): converting the adjacency matrix into a topological overlapping matrix to reduce noise and false correlation, wherein the information can be used for constructing a network or drawing a TOM (time of arrival) graph;
LSTM: the long-short term memory neural network is a neural network used for processing time series data and has the capacity of memorizing historical information.
The first embodiment is as follows:
the embodiment aims to provide a multivariate load prediction method of an integrated energy system considering multi-energy coupling.
A multivariate load prediction method of an integrated energy system considering multi-energy coupling comprises the following steps:
acquiring historical load prediction related data in the park comprehensive energy system, and performing corresponding pretreatment;
based on the historical load prediction related data, adopting a weighted gene co-expression network analysis method to mine nonlinear relations among multiple loads and between the loads and the corresponding influence factors, and determining the influence factors strongly related to different loads;
simultaneously inputting load historical data of different loads and characteristics corresponding to strongly related influence factors of the load historical data into a pre-trained load prediction model to obtain load prediction results corresponding to different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
Further, the mining of the nonlinear relationship between the multiple loads and between the loads and the corresponding influence factors by using the weighted gene co-expression network analysis method specifically comprises the following steps: based on the obtained historical load prediction related data, influence factors which are strongly related to different loads are obtained through a weighted gene co-expression network analysis method, wherein the influence factors include but are not limited to meteorological factors and calendar factors which are strongly related to the loads.
Further, the load prediction model comprises an input layer, a sharing layer and an output layer which are sequentially connected, wherein the sharing layer adopts a BilSTM neural network.
Further, the corresponding preprocessing specifically includes removing abnormal values, filling missing values, and performing normalization processing on the data.
Further, the historical load prediction related data comprises corresponding load historical data under different loads, and corresponding meteorological factors and calendar factors of the load historical data.
Further, the different loads include a cold energy load, a heat energy load, and an electric energy load.
Further, for the pre-trained load prediction model, the performance of the pre-trained load prediction model is evaluated by using the root mean square error, the average absolute error or R2.
Specifically, for the convenience of understanding, the scheme of the present embodiment is described in detail below with reference to the accompanying drawings:
in order to solve the problems existing in the prior art:
(1) The traditional correlation coefficient method cannot be applied to the nonlinear relation between the multivariate load of the park comprehensive energy system and various influence factors, and can not screen truly suitable input factors, so that the problem of low model prediction precision is caused;
(2) The load prediction of single energy can not accurately describe the deep coupling characteristic among multiple loads, so that the model prediction efficiency is not high.
The embodiment provides a comprehensive energy system multi-element load prediction method considering multi-energy coupling, and the method mainly adopts the technical concept that: analyzing the correlation among the characteristics by adopting WGCNA, and screening out proper input characteristics; and (3) adopting an MTL framework, taking the BilSTM as a sharing layer of the MTL, sharing information by each subtask, performing combined cooling, heating and power prediction, and obtaining a prediction result. The method can improve the prediction precision and the prediction efficiency, and therefore, the method can well solve the problems in the background technology. Specifically, as shown in fig. 3, the method of the present embodiment includes the following steps:
step 1: preprocessing original data;
firstly, the load data volume of the park integrated energy system is large, abnormality is easy to occur during data measurement and storage, and abnormal values and missing values need to be identified and filled in order to improve the reliability of data and reduce unnecessary noise generated by abnormal data in the model training process.
Secondly, in order to prevent model learning from deviating due to overlarge magnitude difference between variables, a data normalization method is adopted in the original data set and is scaled to the range of [0,1], so that the prediction method is more effective. The normalized formula is as follows:
in the formula: y' is normalized data; y is original data; y is max 、y min The maximum and minimum values of the data.
Step 2: considering the multipotential coupling of the park comprehensive energy system, adopting WGCNA screening characteristics to determine an input/output characteristic set;
due to the deep coupling of the multi-load of the park comprehensive energy system, the traditional correlation Analysis method cannot deeply excavate the complex nonlinear relationship between loads and between the loads and other factors, and therefore, weighted Gene Co-expression Network Analysis (WGCNA) is adopted as a means for exploring the multi-load correlation of the park comprehensive energy system.
WGCNA is a systematic biological method used to describe patterns of genetic associations between different samples and can be used to identify highly co-varying gene sets. The method is an algorithm for excavating genes with similar expression modes according to different gene expression modes, and is defined as a module. The flow chart of the analysis of WGCNA is shown in FIG. 1. Firstly, constructing selection of a weighting correlation gene network soft threshold, taking a specific value beta (soft threshold) for the power of a correlation coefficient between each two pairs of genes i and j to form an adjacency matrix, wherein the calculation formula is as follows:
a ij ＝|cor(i,j)| β (1)
in the formula: a is ij Is the correlation between genes a ij (ii) a i, j are two different genes; beta is weight and defaults to 1-30.
After selecting a proper beta value, constructing a weighted co-expression network and module identification, and calculating the similarity of two genes through Topological Overlap (TOM), wherein the calculation formula is as follows:
in the formula: and u is a ergon, traversing all genes except i, j in the gene list, and calculating by the formula. TOM ij =0, representing a network of genes i and j without common contiguous genes; TOM ij =1 represents genes i and j having identical network adjacent genes.
Finally, the gene modules are associated with external information, WGCNA provides a function for visualizing the correlation of the gene modules, and finally, a correlation coefficient matrix of each module and the external characteristics is obtained, so that the modules highly related to the external characteristics can be found.
WGCNA has been successfully applied in the fields of biology, genetics and the like, and is less applied in the engineering field. The invention applies WGCNA to multi-load prediction to screen characteristics having strong influence on load change, sample points of each characteristic in a data set form a gene set, and a process of weighted gene co-expression network analysis is realized by using WGCNA packet of R language.
After a visual result is obtained, the correlation coefficient of each module is W ij Selecting constant q as division standard, when W ij <When q is greater than q, the correlation is considered to be weak and is not used as an input factor; if W ij When q is more than or equal to q, the strong correlation is determined as an input factor. Determining the input feature set of each task as k by the step n ＝{k 1 ,k 2 ,…,k l L is the number of input factors, and n is the number of subtasks. The input feature set is generally historical data of cooling, heating and power loads, meteorological factors and calendar factors strongly related to the loads, and the output feature set is the cooling, heating and power loads to be predicted.
And 3, step 3: multivariate load prediction model based on BilSTM-MTL
The BilSTM-MTL prediction model proposed in this embodiment is divided into an input layer, a shared layer and an output layer. The model flow diagram is shown in fig. 2.
An input layer: after the first step and the second step are carried out, determining an input feature set of each subtask as an input layer input = { k } of the model 1 ,…,k n -wherein n =3;
sharing layer: the neural network layer is transmitted to a sharing layer of the model, a plurality of LSTM neural elements are combined together to form a BilSTM neural network layer in the sharing layer, and then the BiLSTM neural network layer is linearly connected with a plurality of BilTM network layers with the same structure to form a multilayer sharing network in the MTL;
an output layer: outputting the label value of each subtask, wherein the output layer can comprise a dropout layer, a dense layer and the like, and the load value output = { o } of the cold, heat and electricity is output 1 ,…,o n H, where n =3.
The principles of multi-task learning (MTL) and bidirectional long-short term memory network (BilSTM) in the model are as follows.
(1)MTL
Compared with single-task Learning, multi-task Learning (MTL) can complete a plurality of different tasks in parallel, and the generalization capability of each subtask is improved through the correlation between the tasks. By adopting a hard parameter sharing mechanism in the multi-task learning, the data volume and the parameter scale of the whole model can be reduced, so that the model has higher efficiency.
(2)BiLSTM
When the comprehensive energy load prediction of the park adopts the unidirectional neural network, the propagation training is carried out from front to back according to the time sequence, the load sequence has autocorrelation, the data utilization rate of the unidirectional neural network to the long-time sequence is low, and the inherent characteristics of the data cannot be effectively mined. BilSTM is composed of forward LSTM and backward LSTM, and can better capture the dependency of time series, so that the BiLSTM is adopted as a sharing layer in multi-task learning to share information parameters in order to further mine the related information of the load itself between the past and future time and further improve the prediction accuracy of the model.
The network parameters of the LSTM are calculated as follows:
f t ＝σ(W f [h t-1 ,x t ]+b f ) (4)
i t ＝σ(W i [h t-1 ,x t ]+b i ) (5)
o t ＝σ(W o [h t-1 ,x t ]+b o ) (6)
c t ＝f t c t-1 +i t tanh(W c [h t-1 ,x t ]+b c ) (7)
h t ＝σ(o t tanhc t ) (8)
in the formula: f. of t 、i t 、o t 、c t The states of the forgetting gate, the input gate, the output gate and the state unit at the current time t are respectively; h is t-1 The state of the hidden layer at the previous moment; x is the number of t Inputting at the current time t; w is a group of f 、W i 、W o 、W c And b f 、b i 、b o 、b c Respectively corresponding weight coefficient matrix and bias item; σ denotes Sigmoid activation function.
Hidden layer state h of BilSTM t From the input a at the present moment t Hidden layer output state h at the previous moment of forward propagation t-1 And the output state h of the preceding instant of the back propagation i-1 And (4) forming. The hidden layer state is shown below.
h t ＝LSTM(x t ,h t-1 ) (9)
h i ＝LSTM(x t ,h i-1 ) (10)
h t ＝a t h t +b t h i +c t (11)
In the formula: LSTM is the operation process of the formulas (4) - (8); h is t Is a forward hidden layer state; h is i Is in a backward hidden layer state; a is t Outputting weights for the hidden layers of the forward propagation unit; b is a mixture of t Outputting weights for the hidden layers of the backward propagation unit; c. C t And optimizing parameters for the hidden layer bias at the current moment.
And 4, step 4: model evaluation
Comparing the combined prediction result of the cold, heat and electricity output by the model with the true value by using Root Mean Square Error (RMSE), mean Absolute Error (MAE) and R 2 As an evaluation index of the model, the following formula is given:
in the formula: n is the number of samples; y is i Is an actual value representing time i;
is an actual value representing time i;
is the average of the samples.
When RMSE and MAE are used as error evaluation indices, the smaller the value, the better; using R 2 When the model performance evaluation index is used, the larger the value is, the better the fitting degree of the model is represented.
Example two:
the embodiment aims to provide the comprehensive energy system multivariate load prediction system considering the multi-energy coupling.
An integrated energy system multivariate load prediction system considering multi-energy coupling, comprising:
the data acquisition unit is used for acquiring historical load prediction related data in the park comprehensive energy system and performing corresponding pretreatment;
the influence factor determining unit is used for mining nonlinear relations among multiple loads and between the loads and the corresponding influence factors by adopting a weighted gene co-expression network analysis method based on the historical load prediction related data, and determining the influence factors strongly related to different loads;
the load prediction unit is used for inputting the load historical data of different loads and the characteristics corresponding to the strongly related influence factors into a pre-trained load prediction model to obtain load prediction results corresponding to the different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
Further, the system described in this embodiment corresponds to the method described in the first embodiment, and the technical details thereof have been described in detail in the first embodiment, so that details are not repeated herein.
In further embodiments, there is also provided:
an electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, the computer instructions when executed by the processor performing the method of embodiment one. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processor, a digital signal processor DSP, an application specific integrated circuit ASIC, an off-the-shelf programmable gate array FPGA or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
A computer readable storage medium storing computer instructions which, when executed by a processor, perform the method of embodiment one.
The method in the first embodiment may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements, i.e., algorithm steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.
The comprehensive energy system multi-element load prediction method and system considering the multi-energy coupling can be realized, and have wide application prospects.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (10)
1. A multivariate load prediction method for an integrated energy system considering multi-energy coupling is characterized by comprising the following steps:
acquiring historical load prediction related data in the park comprehensive energy system, and performing corresponding pretreatment;
based on the historical load prediction related data, adopting a weighted gene co-expression network analysis method to mine nonlinear relations among multiple loads and between the loads and the corresponding influence factors, and determining the influence factors strongly related to different loads;
simultaneously inputting load historical data of different loads and characteristics corresponding to strongly related influence factors of the load historical data into a pre-trained load prediction model to obtain load prediction results corresponding to different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
2. The method for forecasting the multivariate load of the comprehensive energy system considering the multi-energy coupling as claimed in claim 1, wherein the mining of the nonlinear relationship among the multivariate loads and between the loads and the corresponding influencing factors by the weighted gene co-expression network analysis method specifically comprises the following steps: based on the obtained historical load prediction related data, influence factors which are strongly related to different loads are obtained through a weighted gene co-expression network analysis method, wherein the influence factors include but are not limited to meteorological factors and calendar factors which are strongly related to the loads.
3. The method of claim 1, wherein the load prediction model comprises an input layer, a sharing layer and an output layer which are connected in sequence, wherein the sharing layer adopts a BilSTM neural network.
4. The method according to claim 1, wherein the preprocessing comprises removing outliers, filling missing values, and normalizing the data.
5. The method of claim 1, wherein the historical load forecast related data includes historical load data corresponding to different loads, and weather factors and calendar factors corresponding to the historical load forecast related data.
6. The method of claim 1, wherein the different loads comprise a cold energy load, a heat energy load, and an electrical energy load.
7. The method for multi-load prediction of an integrated energy system considering multi-energy coupling as claimed in claim 1, wherein the pre-trained load prediction model is evaluated for performance by using root mean square error, mean absolute error or R2.
8. An integrated energy system multivariate load prediction system considering multi-energy coupling, comprising:
the data acquisition unit is used for acquiring historical load prediction related data in the park integrated energy system and carrying out corresponding pretreatment;
the influence factor determining unit is used for mining nonlinear relations among multiple loads and between the loads and the corresponding influence factors by adopting a weighted gene co-expression network analysis method based on the historical load prediction related data, and determining the influence factors strongly related to different loads;
the load prediction unit is used for inputting the load historical data of different loads and the characteristics corresponding to the strongly related influence factors into a pre-trained load prediction model to obtain load prediction results corresponding to the different loads; the load prediction model adopts an MTL framework, a BilSTM is used as a sharing layer of the MTL, and prediction tasks under different loads share information through the sharing layer.
9. An electronic device comprising a memory, a processor and a computer program stored and executed on the memory, wherein the processor executes the program to implement a method of multivariate load prediction for an integrated energy system considering multi-energy coupling according to any of claims 1-7.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method for multivariate load prediction of an integrated energy system considering multi-energy coupling according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924624.7A CN115310355A (en) | 2022-08-02 | 2022-08-02 | Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210924624.7A CN115310355A (en) | 2022-08-02 | 2022-08-02 | Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115310355A true CN115310355A (en) | 2022-11-08 |
Family
ID=83857920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210924624.7A Pending CN115310355A (en) | 2022-08-02 | 2022-08-02 | Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115310355A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116070804A (en) * | 2023-04-06 | 2023-05-05 | 国网冀北电力有限公司 | Power system load prediction method and device based on knowledge graph and data driving |
-
2022
- 2022-08-02 CN CN202210924624.7A patent/CN115310355A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116070804A (en) * | 2023-04-06 | 2023-05-05 | 国网冀北电力有限公司 | Power system load prediction method and device based on knowledge graph and data driving |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Al-Janabi et al. | A new method for prediction of air pollution based on intelligent computation | |
| CN113962364B (en) | Multi-factor power load prediction method based on deep learning | |
| CN109754113B (en) | Load prediction method based on dynamic time warping and long-and-short time memory | |
| Khafaga et al. | Forecasting Energy Consumption Using a Novel Hybrid Dipper Throated Optimization and Stochastic Fractal Search Algorithm | |
| Yao et al. | Short-term load forecasting method based on feature preference strategy and LightGBM-XGboost | |
| Hu et al. | Electrochemical-theory-guided modeling of the conditional generative adversarial network for battery calendar aging forecast | |
| CN113554466A (en) | Short-term power consumption prediction model construction method, prediction method and device | |
| CN111985719A (en) | Power load prediction method based on improved long-term and short-term memory network | |
| CN111311001B (en) | Bi-LSTM network short-term load prediction method based on DBSCAN algorithm and feature selection | |
| CN114239991A (en) | Building heat supply load prediction method, device and equipment based on data driving | |
| CN113591368A (en) | Comprehensive energy system multi-energy load prediction method and system | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN115018193A (en) | Time series wind energy data prediction method based on LSTM-GA model | |
| CN115310355A (en) | Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system | |
| CN111178641A (en) | Short-term power load prediction method based on feature extraction and multi-core RSVR combined model | |
| CN117439053A (en) | Method, device and storage medium for predicting electric quantity of Stacking integrated model | |
| CN112990530A (en) | Regional population number prediction method and device, electronic equipment and storage medium | |
| CN117390550A (en) | Low-carbon park carbon emission dynamic prediction method and system considering emission training set | |
| CN116578858A (en) | Air compressor fault prediction and health degree evaluation method and system based on graphic neural network | |
| CN114065646B (en) | Energy consumption prediction method based on hybrid optimization algorithm, cloud computing platform and system | |
| CN116090757A (en) | Method for evaluating capability demand satisfaction of information guarantee system | |
| CN115423146A (en) | Self-adaptive runoff forecasting method based on multi-factor nearest neighbor sampling regression and support vector machine | |
| CN114707422A (en) | Intelligent power check method based on load prediction | |
| Shen et al. | An interval analysis scheme based on empirical error and mcmc to quantify uncertainty of wind speed | |
| CN114021847A (en) | Short-term power load prediction method and system based on ISGU (inverse transformation and aggregation) hybrid model |
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 |