CN113609762A - Electric cooling and heating load combined prediction method and system based on GRU-MTL - Google Patents
Electric cooling and heating load combined prediction method and system based on GRU-MTL Download PDFInfo
- Publication number
- CN113609762A CN113609762A CN202110850013.8A CN202110850013A CN113609762A CN 113609762 A CN113609762 A CN 113609762A CN 202110850013 A CN202110850013 A CN 202110850013A CN 113609762 A CN113609762 A CN 113609762A
- Authority
- CN
- China
- Prior art keywords
- load
- heating
- data
- cooling
- electric
- 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
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention belongs to the technical field of comprehensive energy system multi-element load prediction, and provides a GRU-MTL-based combined prediction method and system for electric cooling and heating loads. Inputting historical data of an electric load, historical data of a cold load, historical data of a heat load, historical data of weather and calendar information into a trained electric cold and heat load prediction model to obtain electric load data, cold load data and heat load data to be predicted; wherein the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
Description
Technical Field
The invention belongs to the technical field of comprehensive energy system multi-element load prediction, and particularly relates to a GRU-MTL-based combined prediction method and system for electric cooling and heating loads.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
As an important component of an energy system, a traditional power system, a traditional thermodynamic system and the like are independently planned, designed and operated, coupling among different types of energy is broken, and the flexibility of operation of the energy system is limited to a great extent.
An Integrated Energy System (IES) is an important component of a new-generation energy system, covers energy systems such as power supply, cold supply and heat supply, integrates energy supply, energy conversion and energy storage devices in various forms, and realizes coupling of different types of energy sources in different links such as sources, networks and loads. With the increasing coupling of various energy systems and the increasing market for access to large-scale renewable energy, energy production and consumption, these changes put higher demands on the accuracy of energy use prediction.
The load of the comprehensive energy system has diversity, including electric load, cold load, heat load and the like, and accurate short-term prediction of the multi-load is the basis of operation and scheduling of the comprehensive energy system, and the guarantee of the prediction accuracy is particularly important. At present, more researches are carried out on short-term load prediction of an electric power system, and a prediction model has higher prediction precision and is widely applied to dispatching operation of the electric power system. Compared with the prior art, the electrical load, the cold load and the heat load have strong time-varying property and coupling property, so that many relevant influence factors exist, the coupling property among the electrical load, the cold load and the heat load is not fully considered for carrying out comprehensive modeling when the electrical load, the cold load and the heat load are predicted, and the prediction accuracy of the model is influenced.
Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides a GRU-MTL-based combined prediction method and system for electric cooling and heating loads, which can fully simulate the coupling characteristics between loads and between the loads and various input characteristics when performing combined prediction of the electric cooling and heating loads, and improve the accuracy of the combined prediction of the electric cooling and heating loads, thereby well solving the problem of the coupling characteristics in the combined prediction of the electric cooling and heating loads mentioned in the background above.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a GRU-MTL-based combined prediction method for electric heating and cooling loads.
A method for jointly predicting electric cooling and heating loads based on GRU-MTL comprises the following steps:
inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
wherein the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
The invention provides a combined prediction system of electric heating and cooling loads based on GRU-MTL.
A GRU-MTL-based combined prediction system for electric cooling and heating loads comprises:
a prediction module configured to: inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
a model building module configured to: the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
A third aspect of the invention provides a computer-readable storage medium.
A computer readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the GRU-MTL based combined prediction of electrical and thermal load as defined in the first aspect above.
A fourth aspect of the invention provides a computer apparatus.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the GRU-MTL based combined prediction of electrical and thermal load as described in the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
the method can effectively solve the problem of coupling characteristics between loads and between the loads and among various input characteristics in the combined prediction of the electric cooling and heating loads, thereby improving the accuracy of the combined prediction of the electric cooling and heating loads.
Advantages of additional aspects of the invention 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 invention.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a frame diagram of the method for jointly predicting the electrical heating and thermal load based on GRU-MTL.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. 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 invention 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 exemplary embodiments according to the invention. 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.
It is noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present disclosure. It should be noted that each block in the flowchart or block diagrams may represent a module, a segment, or a portion of code, which may comprise one or more executable instructions for implementing the logical function specified in the respective embodiment. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In recent years, existing load prediction methods are mainly classified into a conventional method, a machine learning method, and a deep learning method. The traditional prediction method such as Kalman filtering is only suitable for small-scale data, and the requirement on data stability is high; although the traditional machine learning methods such as support vector regression and the like are improved to a certain extent compared with the traditional methods, long-time sequence information cannot be remembered, and the prediction precision is limited; the long-short term memory neural network most widely used in the deep learning method has poor prediction effect due to the fact that the long-short term memory neural network cannot well simulate the coupling among the multiple loads.
As the multi-element load of the comprehensive energy system is influenced by various factors and presents the characteristics of uncertainty and nonlinearity, the multi-element load such as electricity, cold and heat of the system is not only related to the historical data of the multi-element load, but also has non-negligible influence on the accuracy of multi-element load prediction due to the coupling characteristics between the load and between the load and various input characteristics. Therefore, aiming at the problem, the invention provides an electric heating and cooling load combined prediction method based on the combination of GRU and MTL. The multi-task learning structure is used for training a plurality of load prediction tasks in parallel by using a shared learning mechanism, and the characteristic information hidden in a plurality of related tasks is fully utilized, so that the generalization capability is improved, and the problem of coupling in multi-load prediction is solved; and considering the condition that the input characteristic data set is not very large in the electric heating and cooling load combined prediction research, the electric heating and cooling multi-load prediction is carried out by combining the multi-task learning structure and the LSTM variant GRU neural network, so that the prediction precision is improved.
Multi-task learning (MTL) is a kind of transfer learning, and due to the existence of a hidden layer parameter hard sharing mechanism, a hidden layer of a network can be shared among three subtasks for predicting electric, cold and thermal loads, coupling characteristics among input features of the three subtasks are simulated, and branching is started at a part close to output of the network to obtain target values of the three subtasks, namely predicted values of the electric, cold and thermal loads. Compared with the single use of an LSTM or GRU model for multi-task output, the MTL can share and represent a plurality of tasks in a shallow layer, the risk of network overfitting is reduced, the network generalization effect is improved, and therefore the prediction precision is improved.
Embodiments of the invention are described below by way of a number of examples:
example one
As shown in fig. 1, the embodiment provides a method for jointly predicting electrical, thermal and thermal loads based on GRU-MTL, and the embodiment is exemplified by applying the method to a server, it is understood that the method may also be applied to a terminal, and may also be applied to a system including a terminal and a server, and is implemented by interaction between the terminal and the server. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network server, cloud communication, middleware service, a domain name service, a security service CDN, a big data and artificial intelligence platform, and the like. The terminal may be, but is not limited to, a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and the like. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein. In this embodiment, the method includes the steps of:
inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
wherein the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
As one or more embodiments, the process of training the electric heating and cooling load prediction model includes:
constructing an input feature set and an output feature set, wherein input samples in the input feature set correspond to output samples in the output feature set one by one;
inputting the input samples in the input characteristic set into an electric cold and heat load prediction model, comparing predicted electric load data, cold load data and heat load data output by the model with corresponding output samples in the output characteristic set, and calculating a loss function;
and adjusting the hyper-parameters of the electric heating and cooling load prediction model generation by generation according to the loss function until iteration reaches the preset times to obtain the trained electric heating and cooling load prediction model.
Specifically, in this embodiment, an input feature set and an output feature set are first divided, data preprocessing operations such as abnormal value processing, missing value processing, normalization processing and the like are performed on a historical data set of the input feature, then the preprocessed data set is input into a GRU-MTL electric cooling and heating load prediction model designed by the present invention, operations such as network structure hyper-parameter selection, network model training, network parameter optimization, model performance evaluation and the like are performed on the preprocessed data set, and finally, a joint prediction result of the electric cooling and heating load is obtained.
The specific scheme of the invention is as follows:
(1) input and output feature set partitioning
As one or more implementations, the set of input features includes: electrical load historical data, cold load historical data, thermal load historical data, weather historical data, and calendar information; the set of output features includes: predicted electrical load data, cold load data, and heat load data; the weather history data includes: temperature history data, humidity history data, wind speed history data, and pressure history data.
Specifically, the input feature set and the output feature set are key for determining the performance capability of the model. The input characteristics are various attributes influencing combined prediction of the electric heating and cooling loads, and necessary related information such as historical data, weather information, calendar information and the like needs to be determined when the input characteristics are divided according to the combined prediction problem of the electric heating and cooling loads. Therefore, the present embodiment selects data such as electrical load history data, cold load history data, heat load history data, temperature data, humidity data, wind speed data, pressure data, and calendar information as the input feature set. And the output feature set is the electric load data, the cold load data and the heat load data at the moment to be predicted.
(2) Data pre-processing
As one or more embodiments, before inputting the electrical load historical data, the cooling load historical data, the thermal load historical data, the weather historical data, and the calendar information into the trained electrical cooling and heating load prediction model, the method further comprises: and performing abnormal value processing, missing value processing and normalization processing on the electric load historical data, the cold load historical data, the heat load historical data and the weather historical data.
Specifically, firstly, in view of the fact that the data set is prone to abnormal conditions in the measuring, transmitting and storing processes, if the part of samples is directly discarded, the available information of the established GRU-MTL electric heating and cooling load prediction model is greatly reduced, and the model prediction performance is reduced. Therefore, missing value filling and outlier identification are carried out on the data set so as to ensure the integrity and the goodness of the data set.
Secondly, after the input feature set and the output feature set are determined, normalization processing needs to be carried out on the input feature set and the output feature set so as to prevent the model prediction accuracy from being influenced by large order difference among variables. Therefore, this embodiment uses the Min-Max normalization method to perform linear transformation on the original data, so that the result value is mapped between [0, 1 ]. The transformation function is as follows:
wherein max is the maximum value of the input feature sample data, min is the minimum value of the input feature sample data, and x is the current input feature sample data.
(3) GRU-MTL electric cooling and heating load combined prediction model
Due to the existence of the coupling characteristic problem between the load and between the load and various input characteristics during the electric heating and cooling load joint prediction, the Gate control cycle Unit (GRU) and the Multi-task Learning (MTL) are combined together in the embodiment, and an electric heating and cooling load joint prediction model is built. When the MTL is used for processing a problem, the MTL can learn by using the sharing layer and obtain auxiliary coupling information provided by other related subtasks, the coupling characteristic between input features in electric-cooling-heating load joint prediction can be effectively simulated, and the purposes of improving the output precision of the model and enhancing the generalization capability are further achieved. The MTL contains a set of tasks yt(T ∈ T) and data set
Wherein T is the number of tasks, N is the number of data samples,
the tag of the t task for the ith data point. Defining the prediction function as ft(x；θsh，θt)：x→ytWherein thetashParameters shared for different tasks, thetatFor task-related parameters, the overall optimization penalty function for the MTL can be defined as follows:
in the formula: alpha is alphatIs the weight coefficient of the task;
is a loss function defined as
In one or more embodiments, the MTL in the electric heating and cooling load prediction model adopts a learning mode with hard parameter sharing. The key for constructing the electric heating and cooling load joint prediction model by using the MTL is the selection of a learning mode and the construction of a shared learning layer.
1) MTL can divide learning patterns into hard and soft sharing mechanisms by a feature sharing approach. The hard sharing mechanism is that a plurality of subtasks share the same characteristic sharing layer, and characteristic parameters are completely the same; the soft sharing mechanism is that each subtask has its own characteristic parameter, and different subtasks need to be regularized to achieve the purpose of information sharing. Compared with the prior art, the hard sharing mechanism is simpler in characteristic parameters and model structure, the overfitting problem is less prone to being generated under the conditions that the electric heating and cooling load combined prediction model to be constructed is various in parameters and complex in structure, and the generalization capability of the model is stronger. Meanwhile, the soft sharing mechanism is considered to be suitable for tasks without close relation and is not suitable for solving the problem of complex coupling of combined prediction of electric heating and cooling loads. Therefore, the invention adopts a learning mode of hard parameter sharing in the structural design of the electric heating and cooling load combined prediction model.
2) In the shared learning layer construction process, considering the condition that the input characteristic data set is not very large in the electric heating and cooling load combined prediction research, in order to ensure the prediction precision, a GRU neural network is adopted to construct the shared learning layer. The GRU has a reset gate and an update gate inside each neuron, and the specific calculation process is as follows.
And (4) updating the door:
zt＝σ(W(z)xt+U(z)ht-1)
resetting a gate:
rt＝σ(W(r)xt+U(r)ht-1)
the current memory content is as follows:
h′t＝tanh(Wxt+rt⊙Uht-1)
final memory of the current time step:
ht＝zt⊙ht-1+(1-zt)⊙h′t
in the formula, zt、rtUpdating the gating information of the gate and the reset gate for the current time step, x, respectivelytFor input information at the current time step, ht-1For final memory of the last time step, W, U is a weight matrix, h't、htRespectively memorizing the current memory content and the final memory of the current time step, wherein sigma is a sigmoid activation function.
(4) Model construction
As one or more embodiments, the process of constructing the electric heating and cooling load prediction model includes:
and determining the hyper-parameters of the electric heating and cooling load prediction model by adopting a grid optimization method, and constructing the electric heating and cooling load prediction model by combining a multi-task learning MTL and a GRU neural network based on the hyper-parameters of the electric heating and cooling load prediction model.
Specifically, firstly, a grid optimization method is adopted to determine the super-parameters of a GRU-MTL electric cooling and heating load combined prediction model, and a GRU-MTL electric cooling and heating load combined prediction model is established according to the super-parameters of the GRU-MTL electric cooling and heating load combined prediction model; and then inputting the input characteristic data set into the established model, and training the network model until iteration is carried out to preset times. And converting the low-dimensional features in the original data set into high-dimensional features layer by layer through a plurality of hidden layers, so that the model learns the hidden mapping relation. Inputting the characteristic quantity of the verification set into the trained GRU-MTL network by adopting an Adam optimization algorithm, comparing the output combined prediction result of the electric heating and cooling loads with a real value, calculating a loss function, and adjusting network parameters generation by generation according to the loss function; and finally, using Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) as model evaluation indexes, wherein the specific expressions are as follows:
when the three types of error functions are used as evaluation indexes, the smaller the value is, the better the prediction result is.
Example two
The embodiment provides an electric heating and cooling load combined prediction system based on GRU-MTL.
A GRU-MTL-based combined prediction system for electric cooling and heating loads comprises:
a prediction module configured to: inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
a model building module configured to: the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the GRU-MTL-based combined prediction method of electrical and thermal load.
Example four
The embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the GRU-MTL-based combined electrical and thermal load prediction method according to the first embodiment.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention 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, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A GRU-MTL-based combined prediction method for electric cooling and heating loads is characterized by comprising the following steps:
inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
wherein the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
2. The GRU-MTL-based combined cooling, heating and heating load prediction method according to claim 1, wherein the process of building the cooling, heating and heating load prediction model comprises the following steps:
and determining the hyper-parameters of the electric heating and cooling load prediction model by adopting a grid optimization method, and constructing the electric heating and cooling load prediction model by combining a multi-task learning MTL and a GRU neural network based on the hyper-parameters of the electric heating and cooling load prediction model.
3. The GRU-MTL-based combined cooling, heating and heating load prediction method according to claim 1, wherein the process of training the cooling, heating and heating load prediction model comprises:
constructing an input feature set and an output feature set, wherein input samples in the input feature set correspond to output samples in the output feature set one by one;
inputting the input samples in the input characteristic set into an electric cold and heat load prediction model, comparing predicted electric load data, cold load data and heat load data output by the model with corresponding output samples in the output characteristic set, and calculating a loss function;
and adjusting the hyper-parameters of the electric heating and cooling load prediction model generation by generation according to the loss function until iteration reaches the preset times to obtain the trained electric heating and cooling load prediction model.
4. The GRU-MTL-based combined electrical heating and cooling load prediction method according to claim 3, wherein the set of input features comprises: electrical load historical data, cold load historical data, thermal load historical data, weather historical data, and calendar information; the set of output features includes: predicted electrical load data, cold load data, and heat load data.
5. The GRU-MTL based combined cooling, heating and loading prediction method of claim 1, further comprising, before inputting the electrical loading history data, cooling loading history data, heating loading history data, weather history data and calendar information into the trained electrical cooling, heating and loading prediction model: and performing abnormal value processing, missing value processing and normalization processing on the electric load historical data, the cold load historical data, the heat load historical data and the weather historical data.
6. The GRU-MTL-based combined cooling, heating and loading prediction method according to claim 1, wherein the MTL in the cooling, heating and loading prediction model adopts a learning mode with hard parameter sharing.
7. The GRU-MTL-based combined cooling, heating and heating load prediction method according to any one of claims 1-6, wherein the meteorological historical data comprises: temperature history data, humidity history data, wind speed history data, and pressure history data.
8. A GRU-MTL-based combined prediction system for electric cooling and heating loads is characterized by comprising:
a prediction module configured to: inputting the historical data of the electric load, the historical data of the cold load, the historical data of the heat load, the historical data of the weather and the calendar information into a trained electric cold and heat load prediction model to obtain the electric load data, the cold load data and the heat load data to be predicted;
a model building module configured to: the electric heating and cooling load prediction model comprises: and building a shared learning layer of the multi-task learning MTL by adopting the GRU neural network.
9. A computer readable storage medium, having stored thereon a computer program, which, when being executed by a processor, carries out the steps of the GRU-MTL based combined electrical cooling and heating load prediction method according to any one of claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps in the GRU-MTL based combined cooling and heating load prediction method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850013.8A CN113609762A (en) | 2021-07-27 | 2021-07-27 | Electric cooling and heating load combined prediction method and system based on GRU-MTL |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110850013.8A CN113609762A (en) | 2021-07-27 | 2021-07-27 | Electric cooling and heating load combined prediction method and system based on GRU-MTL |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113609762A true CN113609762A (en) | 2021-11-05 |
Family
ID=78338442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110850013.8A Pending CN113609762A (en) | 2021-07-27 | 2021-07-27 | Electric cooling and heating load combined prediction method and system based on GRU-MTL |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609762A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114676941A (en) * | 2022-05-30 | 2022-06-28 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN115034518A (en) * | 2022-07-29 | 2022-09-09 | 华北电力大学 | Method and system for predicting multi-element load of cooling, heating and power |
| CN115983448A (en) * | 2022-12-14 | 2023-04-18 | 南京信息工程大学 | Multi-energy load prediction method based on space-time diagram neural network |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619389A (en) * | 2019-09-23 | 2019-12-27 | 山东大学 | Load prediction method and system of combined cooling heating and power system based on LSTM-RNN |
| CN111950793A (en) * | 2020-08-17 | 2020-11-17 | 浙江工业大学 | Comprehensive energy system load prediction method considering multivariate load coupling characteristics |
| CN113095598A (en) * | 2021-05-07 | 2021-07-09 | 国网山东省电力公司经济技术研究院 | Multi-energy load prediction method, system, device and medium |
-
2021
- 2021-07-27 CN CN202110850013.8A patent/CN113609762A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619389A (en) * | 2019-09-23 | 2019-12-27 | 山东大学 | Load prediction method and system of combined cooling heating and power system based on LSTM-RNN |
| CN111950793A (en) * | 2020-08-17 | 2020-11-17 | 浙江工业大学 | Comprehensive energy system load prediction method considering multivariate load coupling characteristics |
| CN113095598A (en) * | 2021-05-07 | 2021-07-09 | 国网山东省电力公司经济技术研究院 | Multi-energy load prediction method, system, device and medium |
Non-Patent Citations (2)
| Title |
|---|
| KUNPENG ZHANG: "A multitask learning model for traffic flow and speed", IEEE, pages 9 * |
| 孙庆凯: "基于 LSTM 和多任务学习的综合能源系统多元负荷预测", 电力系统自动化, vol. 45, no. 5, pages 8 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114676941A (en) * | 2022-05-30 | 2022-06-28 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN114676941B (en) * | 2022-05-30 | 2022-09-30 | 国网天津市电力公司经济技术研究院 | Electric-thermal load combined self-adaptive prediction method and device for park comprehensive energy system |
| CN115034518A (en) * | 2022-07-29 | 2022-09-09 | 华北电力大学 | Method and system for predicting multi-element load of cooling, heating and power |
| CN115983448A (en) * | 2022-12-14 | 2023-04-18 | 南京信息工程大学 | Multi-energy load prediction method based on space-time diagram neural network |
| CN115983448B (en) * | 2022-12-14 | 2023-09-12 | 南京信息工程大学 | Multi-energy load prediction method based on space-time diagram neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | A multiple time series-based recurrent neural network for short-term load forecasting | |
| CN113609762A (en) | Electric cooling and heating load combined prediction method and system based on GRU-MTL | |
| CN109948845A (en) | A kind of distribution network load shot and long term Memory Neural Networks prediction technique | |
| Atef et al. | A comparative study using deep learning and support vector regression for electricity price forecasting in smart grids | |
| CN109359786A (en) | A kind of power station area short-term load forecasting method | |
| Zhao et al. | DGM (1, 1) model optimized by MVO (multi-verse optimizer) for annual peak load forecasting | |
| Islam et al. | LSTM-based electrical load forecasting for Chattogram city of Bangladesh | |
| CN108876001A (en) | A kind of Short-Term Load Forecasting Method based on twin support vector machines | |
| Zhang et al. | Modeling IoT equipment with graph neural networks | |
| Tan et al. | Support vector machine algorithm for artificial intelligence optimization | |
| CN113158572A (en) | Short-term load prediction method and device | |
| CN115481788B (en) | Phase change energy storage system load prediction method and system | |
| Yang | A novel short-term multi-input–multi-output prediction model of wind speed and wind power with LSSVM based on improved ant colony algorithm optimization | |
| CN116526473A (en) | Particle swarm optimization LSTM-based electrothermal load prediction method | |
| Barić et al. | Short-term forecasting of electricity consumption using artificial neural networks-an overview | |
| Stankovic et al. | Forecasting ethereum price by tuned long short-term memory model | |
| Guo et al. | Applying gated recurrent units pproaches for workload prediction | |
| CN110738363B (en) | Photovoltaic power generation power prediction method | |
| CN117709540A (en) | Short-term bus load prediction method and system for identifying abnormal weather | |
| Setiawan et al. | Indoor Climate Prediction Using Attention-Based Sequence-to-Sequence Neural Network | |
| Showkati et al. | Short term load forecasting using echo state networks | |
| Li et al. | Short-term Power Load Forecasting Based on Gate Recurrent Unit Network and Cloud Computing Platform | |
| CN112330044A (en) | Support vector regression model based on iterative aggregation grid search algorithm | |
| CN116722541A (en) | Power system load prediction method and device based on convolutional neural network | |
| CN116300755A (en) | Double-layer optimal scheduling method and device for heat storage-containing heating system based on MPC |
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 |