Multi-energy-flow planning method, device and equipment for integrated energy system and storage medium

Features are extracted through the LSTM and CNN-LSTM models and combined with the XGBoost algorithm to generate and operate the strategy, the low-carbon economic problem of multi-energy flow planning of integrated energy system in a multi-market environment is solved, and the operation effect of lower carbon economy and energy saving and consumption reduction is achieved. .

CN118917579APending Publication Date: 2024-11-08ELECTRIC POWER RES INST OF STATE GRID ZHEJIANG ELECTRIC POWER COMAPNY
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410929102.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The existing multi-energy flow planning methods of integrated energy systems are difficult to effectively match electricity prices, carbon prices and natural gas prices in a multi-market environment, making it difficult to achieve low-carbon economic system operation.

Method used

The LSTM and CNN-LSTM models are used to extract timing and spatiotemporal features, combined with the XGBoost algorithm to predict the multi-market environment, generate operation strategies, and determine the multi-energy flow output of the integrated energy system through optimization algorithms to reduce operating costs and carbon emissions.

Benefits of technology

A lower-carbon and economical system operation model has been achieved, the operating cost control effect of the comprehensive energy system has been improved, the carbon emission reduction potential has been tapped, and the operation of energy conservation and consumption reduction has been ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917579A_ABST
    Figure CN118917579A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-energy-flow planning method, device and equipment for an integrated energy system and a storage medium. According to the invention, on the basis of the preprocessed historical data, the time sequence features are extracted according to the long-short-term memory recurrent neural network framework, the spatial-temporal features are extracted according to the spatial-temporal network, and the obtained features are input into the extreme value gradient boosting model, so that the prediction of the carbon price, the electricity price and the natural gas price is realized. And determining a comprehensive energy system operation strategy adapted to the multi-market environment, establishing a comprehensive energy system multi-energy flow optimization planning model containing the operation cost and the carbon emission cost by combining prediction results of the carbon price, the electricity price and the natural gas price, and solving by adopting an optimization algorithm. The method fully considers the influence of the multi-market environment on the formulation of the operation strategy of the integrated energy system, can effectively achieve the multi-energy-flow planning adapting to the market change, effectively improves the reasonability of the multi-energy-flow planning of the integrated energy system, and enhances the collaborative operation level of the integrated energy system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of integrated energy system planning, and in particular to a method, device, equipment and storage medium for multi-energy flow planning of an integrated energy system taking into account multiple market environments. Background Art

[0002] The integrated energy system built on the basis of multi-energy flow coupling utilizes the complementary characteristics of different energy types such as wind, light, water, and natural gas to coordinate planning to meet the multiple load demands such as electricity, heat, cooling, and steam. The multi-market environment such as the electricity, carbon, and natural gas markets is an important factor affecting the coordinated planning of the integrated energy system. In order to more accurately determine the electricity price, carbon price, and natural gas price in the multi-market environment and propose an optimized operation strategy for the low-carbon economy, it is necessary to predict the multi-market environment and clarify the supply and demand interaction mechanism between the multi-market environment and the multi-energy flow of the integrated energy system, so as to form a multi-energy flow low-carbon economic operation strategy for the integrated energy system that adapts to the multi-market environment.

[0003] The existing multi-energy flow planning methods of integrated energy systems mainly take carbon emission reduction, operating cost reduction, and system efficiency improvement as optimization goals to meet the diversified energy needs of regional users. However, in reality, the multi-energy flow planning model is limited by multi-market environmental factors such as electricity prices, carbon prices, and natural gas prices that are related to the optimization goals, making it difficult to match the energy needs of regional users in a multi-market environment in a low-carbon and economical manner. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects in the above-mentioned prior art and to provide a method, device, equipment and storage medium for multi-energy flow planning of an integrated energy system taking into account multiple market environments. It comprehensively considers the evolution trend of electricity procurement, natural gas procurement and carbon emission costs, predicts the multi-energy flow demands of the integrated energy system such as cold, heat, electricity, steam, etc., and plans the multi-energy flow output of the integrated energy system to improve the low-carbon economic management and control effect of the operating cost of the integrated energy system.

[0005] In a first aspect, the present invention provides a method for multi-energy flow planning of an integrated energy system considering a multi-market environment, comprising:

[0006] Capture the impact of meteorological characteristics in historical data on current electricity prices, carbon prices, and natural gas prices to obtain time series characteristics;

[0007] Extract the spatial dependencies of multivariate load demand and carbon trading quota data in the target area, and input the results obtained by the CNN framework into the LSTM architecture to obtain spatiotemporal features;

[0008] Based on the time series features, spatiotemporal feature extraction results and other external environment features, the XGBoost algorithm is used to generate an overall forecast of the multi-market environment, and an operation strategy is selected based on the results;

[0009] Based on the forecast results of carbon price, electricity price and natural gas price and the selected operation strategy, the multi-energy flow output of the integrated energy system is determined by an objective function including operation cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

[0010] The present invention is based on preprocessed historical data, extracts time series features according to the long short-term memory recursive neural network (LSTM) framework, extracts spatiotemporal features according to the spatiotemporal network (CNN-LSTM) framework, and inputs the acquired features into the extreme value gradient boosting (XGBoost) model to realize the prediction of carbon price, electricity price and natural gas price, determine the operation strategy of the integrated energy system that adapts to the multi-market environment, and combines the prediction results of carbon price, electricity price and natural gas price to establish a multi-energy flow optimization planning model of the integrated energy system including operating cost and carbon emission cost, and adopts the optimization algorithm to solve it.

[0011] As a preferred scheme of the multi-energy flow planning method of the integrated energy system described in the present invention, the historical data includes at least one of the meteorological information of the target area, multiple load demands, carbon trading quota data, electricity price data, carbon price data and natural gas price data; the meteorological information data includes at least one of solar radiation, temperature, humidity, air pressure and wind speed; the multiple loads include at least one of heat, cold, electricity and steam loads.

[0012] As a preferred solution of the multi-energy flow planning method of the integrated energy system of the present invention, the LSTM model includes a cell state c, a forget gate f, an input gate i and an output gate o. The specific formula of the LSTM model is as follows:

[0013] i t =σ(W i [x t ,h t-1 ]+b i ),

[0014] f t =σ(W f [x t ,h t-1 ]+b f ),

[0015] c t =f t *c t-1 +i t tanh(W c [x t ,h t-1 ]+b c ),

[0016] ot =σ(W o [x t ,h t-1 ]+b o ),

[0017] h t =o t *tanh(c t ),

[0018] Among them, W i , W f , W c , W o Represents the weight matrices of the input gate, forget gate, cell state, and output gate respectively; b i 、b f 、b c 、b o They represent the bias terms of the input gate, forget gate, cell state, and output gate respectively; σ represents the ReLU function; tanh represents the hyperbolic tangent function; h t Represents the hidden layer state.

[0019] As a preferred solution of the multi-energy flow planning method of the integrated energy system described in the present invention, according to the meteorological information at time t to t-ph, the LSTM model is used to capture the impact of meteorological information on electricity price, carbon price and natural gas price. The meteorological information of the region is expressed as:

[0020] I w =[W t-ph ,W t-(p-1)h ,...,W t ],

[0021] W=[w temperature ,w windspeed ,...,w visibility ],

[0022] Where p represents the input sequence I w ∈R p×q The length of ; q represents W∈R q×1 length; W represents the meteorological information of temperature, humidity, air pressure, wind speed and solar radiation in the area.

[0023] As a preferred solution of the multi-energy flow planning method for the integrated energy system of the present invention, the CNN model has four layers: an input layer, a convolutional layer, a maximum pooling layer and an output layer;

[0024] Divide a region M into H×W regions based on geographical location, that is Therefore, the input of the CNN model is expressed as:

[0025]

[0026]

[0027] Among them, D and E represent the multi-load demand and carbon trading quota of the region, respectively; d and e represent the multi-load demand and carbon trading quota of the sub-region;

[0028] The input of the CNN-LSTM model is represented as:

[0029] I s =[[D t-ph ,E t-ph ],[D t-(p-1)h ,E t-(p-1)h ],...,[D t ,E t ]],

[0030] Among them, I s Represents the CNN-LSTM model input.

[0031] As a preferred solution of the multi-energy flow planning method of the integrated energy system described in the present invention, the XGBoost algorithm is a wig model based on the boosting integration idea. The forward distribution algorithm is used for greedy learning during training. The objective function of XGBoost consists of a loss function and a regularization term, which is expressed as:

[0032]

[0033]

[0034] Among them, L(Φ) is the objective function; represents the loss function; represents the regularization term; represents the i-th sample x i The predicted value; T = {(x1,y1),(x2,y2),...,(x n ,y n )} represents the training set;

[0035] The objective function is optimized by second-order Taylor expansion, regularization term expansion and removal of constant terms to obtain the final objective function:

[0036]

[0037] Among them, g i represents the first-order derivative of the loss function; h i Represents the second derivative of the loss function.

[0038] As a preferred solution of the multi-energy flow planning method of the integrated energy system of the present invention, the overall prediction of the multi-market environment is: extracting the time series features based on the LSTM framework Results of spatiotemporal feature extraction based on CNN-LSTM framework and other external environmental characteristics (refers to the external environmental characteristics that affect the multi-energy flow of the integrated energy system in addition to the time series characteristics and spatiotemporal characteristics, such as the local electricity market transaction characteristics, etc.) input into the XGBoost model, x i Defined as input fusion feature The predicted value It is expressed as:

[0039]

[0040] As a preferred scheme of the multi-energy flow planning method of the integrated energy system described in the present invention, the operation strategies include the strategy of determining heat by electricity, the strategy of determining electricity by heat, the operation strategy of electricity-heat load following and the operation strategy based on energy storage device.

[0041] As a preferred solution of the multi-energy flow planning method for the integrated energy system of the present invention, the objective function including the operating cost and the carbon emission cost is expressed as:

[0042]

[0043] Among them, COST in Indicates the cost of purchasing electricity, heat, and natural gas energy; COST om It represents the operation and maintenance cost; Represents the cost of carbon dioxide emissions; COST loss Represents the cost of additional losses.

[0044] In a second aspect, the present invention provides a multi-energy flow planning device for an integrated energy system considering multiple market environments, comprising:

[0045] Time series feature acquisition unit: captures the impact of meteorological features in historical data on current electricity prices, carbon prices, and natural gas prices, and obtains time series features;

[0046] Spatiotemporal feature acquisition unit: extracts the spatial dependencies of multivariate load demand and carbon trading quota data in the target area, and inputs the results obtained by the CNN framework into the LSTM architecture to obtain spatiotemporal features;

[0047] Operation strategy selection unit: based on the time series characteristics, spatiotemporal characteristics and other external environment characteristics, the XGBoost algorithm is used to generate an overall forecast of the multi-market environment, and an operation strategy is selected based on the result;

[0048] Multi-energy flow planning determination unit: Based on the prediction results of carbon price, electricity price, and natural gas price and the selected operation strategy, the multi-energy flow output of the integrated energy system is determined through an objective function that includes operating cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

[0049] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method when executing the computer program.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the described method when executed by a processor.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] Based on historical data such as meteorological information, multi-load demand, carbon trading quota data, electricity price data, carbon price data, and natural gas price data, the present invention extracts time series features and spatiotemporal features through LSTM models and CNN-LSTM models, and inputs them into the XGBoost model to predict the multi-market environment, and proposes an operation strategy for an integrated energy system that adapts to the multi-market environment, so as to guide the system operation mode of lower carbon economy, tap the carbon emission reduction potential of the integrated energy system, and ensure the energy-saving and consumption-reducing operation of the integrated energy system. The present invention fully considers the impact of climate change, the evolution of the electricity-carbon market, and the multi-load demand on the multi-market environment, realizes the evolution and update of the operation strategy of the integrated energy system, and improves the coordinated operation level of the integrated energy system that adapts to market changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 It is a flow chart of the multi-energy flow planning method of the integrated energy system of the present invention;

[0055] Figure 2 This is an architecture diagram of the multi-energy flow planning system for the integrated energy system of the present invention;

[0056] Figure 3 A schematic diagram of a logical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] Example 1

[0059] In the existing technology, integrated energy systems mostly adopt multi-energy flow planning models with optimization goals such as reducing carbon emissions, reducing operating costs, and improving system efficiency. However, traditional planning models cannot effectively meet the diverse load demands in a multi-market environment, and the relevant integrated energy system operation strategies are still imperfect, which may lead to reduced operating efficiency of the integrated energy system, reduced carbon reduction effects, and decreased economic benefits.

[0060] As a preferred embodiment of the present invention, Figure 1 It can be seen that this embodiment is based on historical data such as meteorological information, multi-load demand, carbon trading quota data, electricity price data, carbon price data, natural gas price data, etc., and extracts time series features and spatiotemporal features through LSTM model and CNN-LSTM model, and inputs them into the XGBoost model to predict the multi-market environment, optimize the operation strategy of the integrated energy system, and combine the multi-market environment prediction results and system operation strategy to establish a multi-energy flow planning model for the integrated energy system.

[0061] This embodiment provides a multi-energy flow planning method for an integrated energy system considering a multi-market environment. The specific steps of the multi-energy flow planning method for an integrated energy system considering a multi-market environment are as follows:

[0062] S1: Historical data acquisition and processing.

[0063] Specifically, the historical data includes meteorological information of the target area, multi-load demand, carbon trading quota data, electricity price data, carbon price data, and natural gas price data. The meteorological information data includes but is not limited to solar radiation, temperature, humidity, air pressure, wind speed, etc. The multi-loads include heat, cold, electricity, steam loads, etc.

[0064] Furthermore, historical data needs to be preprocessed to construct a dataset, and a supervised dataset is produced according to the prediction task, including the following steps:

[0065] S1.1 performs linear interpolation to fill in the missing values ​​in the historical data. The specific formula is:

[0066]

[0067] Among them, xτ is the value of the unknown data point; is the value of the previous data point; is the value of the next data point; τ is the time value of the unknown data point; τ i is the time value of the next data point; τ j The time value of the next data point.

[0068] S1.2 performs normalization on the historical data after linear interpolation processing, scaling their values ​​to the interval [0,1]. The normalization expression is as follows:

[0069]

[0070] Among them, x' is the normalized data; x is the original data; x max is the maximum value of the sample data; x min is the minimum value of the sample data.

[0071] S1.3 Each consecutive (M+N) time nodes t1,...,t M+N The data is divided into two parts, using the first M time nodes t1~t M The historical data is used as input data, and then N time nodes t M+1 ~t M+N The historical data of is used as the corresponding supervision information to construct a training data set. M and N are hyperparameters determined by specific task requirements.

[0072] S2: Time series feature extraction based on LSTM framework.

[0073] Specifically, the time series feature extraction based on the LSTM framework captures the impact of meteorological characteristics on current electricity prices, carbon prices, and natural gas prices through the LSTM model; the LSTM model structure includes a cell state (c), a forget gate (f), an input gate (i), and an output gate (o). The specific formula of the LSTM model is as follows:

[0074] i t =σ(W i [x t ,h t-1 ]+b i ),

[0075] f t =σ(W f [x t ,h t-1 ]+b f ),

[0076] c t =f t *c t-1 +it tanh(W c [x t ,h t-1 ]+b c ),

[0077] o t =σ(W o [x t ,h t-1 ]+b o ),

[0078] h t =o t *tanh(c t ).

[0079] Furthermore, based on the meteorological information at time t~t-ph, the LSTM model is used to capture the impact of meteorological information on electricity prices, carbon prices, and natural gas prices. The meteorological information in the region can be expressed as:

[0080] I w =[W t-ph ,W t-(p-1)h ,...,W t ],

[0081] W=[w temperature ,w windspeed ,...,w visibility ].

[0082] S3: Spatiotemporal feature extraction based on CNN-LSTM framework.

[0083] Specifically, the spatiotemporal feature extraction method based on the CNN-LSTM framework is to extract the spatial dependencies of the multi-load demand and carbon trading quota data in the target area through the CNN architecture, and input the results obtained by the CNN framework into the LSTM architecture to obtain the spatiotemporal features.

[0084] It should be noted that there are four layers in the CNN model: input layer, convolution layer, maximum pooling layer and output layer.

[0085] Furthermore, a region M is divided into H×W regions based on geographical location, that is, Therefore, the input of the CNN model can be expressed as:

[0086]

[0087]

[0088] Furthermore, the input of the CNN-LSTM model can be expressed as:

[0089] I s=[[D t-ph ,E t-ph ],[D t-(p-1)h ,E t-(p-1)h ],...,[D t ,E t ]].

[0090] S4: Multi-market environment prediction based on XGBoost.

[0091] Specifically, the multi-market environment prediction based on XGBoost uses the XGBoost algorithm to generate the overall prediction of the multi-market environment and selects the operation strategy based on the results. In addition, the XGBoost algorithm is a wig model based on the idea of ​​boosting integration. The forward distribution algorithm is used for greedy learning during training. The objective function of XGBoost consists of two parts: the loss function and the regularization function, which can be expressed as:

[0092]

[0093]

[0094] Furthermore, the objective function is optimized through the steps of second-order Taylor expansion, regularization term expansion, and removal of constant terms to obtain the final objective function:

[0095]

[0096] It should be noted that the overall forecast of the multi-market environment is the result of extracting time series features based on the LSTM framework. Spatial feature extraction results based on CNN framework and other external environmental characteristics Input to the XGBoost model, x i Defined as input fusion feature The predicted value It can be expressed as:

[0097]

[0098] S5: Development of integrated energy system operation strategies adapted to multi-market environments.

[0099] Specifically, the operation strategies include but are not limited to the electricity-to-heat strategy, the heat-to-electricity strategy, the electricity-to-heat load following operation strategy, and the energy storage device-based operation strategy.

[0100] S6: Multi-energy flow planning of integrated energy systems under different boundary conditions.

[0101] Specifically, the multi-energy flow planning of the integrated energy system under different boundary conditions is based on the prediction results of carbon price, electricity price, and natural gas price and the selection results of operation strategy. The multi-energy flow output of the integrated energy system is determined by an objective function that includes operating cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

[0102] Furthermore, the objective function including operating cost and carbon emission cost can be expressed as:

[0103]

[0104] Among them, COST in Indicates the cost of purchasing electricity, heat, and natural gas energy; COST om It represents the operation and maintenance cost; Represents the cost of carbon dioxide emissions; COST loss Represents the cost of additional losses.

[0105] Specifically, the optimization algorithm for solving the objective function includes but is not limited to a Pareto optimal solution search algorithm, a trade-off strategy method, a Pareto dominance method, and a multi-objective particle swarm optimization algorithm.

[0106] Example 2

[0107] This embodiment provides a multi-energy flow planning device for an integrated energy system considering multiple market environments, such as Figure 2 As shown, it consists of a data preprocessing unit, a timing feature acquisition unit, an operation strategy selection unit and a multi-energy flow planning determination unit.

[0108] Data preprocessing unit: used to obtain and process historical data.

[0109] Time series feature acquisition unit: The LSTM model is used to capture the impact of meteorological features in historical data on current electricity prices, carbon prices, and natural gas prices to obtain time series features.

[0110] Spatiotemporal feature acquisition unit: Based on the CNN architecture, the spatial dependencies of the multivariate load demand and carbon trading quota data in the target area are extracted, and the results obtained by the CNN framework are input into the LSTM architecture to obtain spatiotemporal features.

[0111] Operation strategy selection unit: Based on the time series characteristics, spatiotemporal characteristics and other external environment characteristics, the XGBoost algorithm is used to generate an overall forecast of the multi-market environment, and an operation strategy is selected based on the results.

[0112] Multi-energy flow planning determination unit: Based on the prediction results of carbon price, electricity price, and natural gas price and the selected operation strategy, the multi-energy flow output of the integrated energy system is determined through an objective function that includes operating cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

[0113] The historical data includes at least one of meteorological information of the target area, multiple load demands, carbon trading quota data, electricity price data, carbon price data and natural gas price data; wherein the meteorological information data includes at least one of solar radiation, temperature, humidity, air pressure and wind speed; the multiple loads include at least one of heat, cold, electricity and steam loads.

[0114] The LSTM model includes cell state c, forget gate f, input gate i and output gate o. The specific formula of the LSTM model is as follows:

[0115] i t =σ(W i [x t ,h t-1 ]+b i ),

[0116] f t =σ(W f [x t ,h t-1 ]+b f ),

[0117] c t =f t *c t-1 +i t tanh(W c [x t ,h t-1 ]+b c ),

[0118] o t =σ(W o [x t ,h t-1 ]+b o ),

[0119] h t =o t *tanh(c t ),

[0120] Among them, W i , W f , W c , W o Represents the weight matrices of the input gate, forget gate, cell state, and output gate respectively; b i 、bf 、b c 、b o They represent the bias terms of the input gate, forget gate, cell state, and output gate respectively; σ represents the ReLU function; tanh represents the hyperbolic tangent function; h t Represents the hidden layer state.

[0121] According to the meteorological information at time t~t-ph, the LSTM model is used to capture the impact of meteorological information on electricity price, carbon price and natural gas price. The meteorological information of the region is expressed as:

[0122] I w =[W t-ph ,W t-(p-1)h ,...,W t ],

[0123] W=[w temperature ,w windspeed ,...,w visibility ],

[0124] Where p represents the input sequence I w ∈R p×q The length of ; q represents W∈R q×1 length; W represents the meteorological information of temperature, humidity, air pressure, wind speed and solar radiation in the area.

[0125] There are four layers in the CNN model: input layer, convolution layer, maximum pooling layer and output layer;

[0126] Divide a region M into H×W regions based on geographical location, that is Therefore, the input of the CNN model is expressed as:

[0127]

[0128]

[0129] Among them, D and E represent the multi-load demand and carbon trading quota of the region, respectively; d and e represent the multi-load demand and carbon trading quota of the sub-region;

[0130] The input of the CNN-LSTM model is represented as:

[0131] I s =[[D t-ph ,E t-ph ],[D t-(p-1)h ,E t-(p-1)h ],...,[D t ,E t ]],

[0132] Among them, I sRepresents the CNN-LSTM model input.

[0133] The XGBoost algorithm is a wig model based on the idea of ​​boosting integration. The forward distribution algorithm is used for greedy learning during training. The objective function of XGBoost consists of two parts: the loss function and the regularization term, which are expressed as:

[0134]

[0135]

[0136] Among them, L(Φ) is the objective function; represents the loss function; represents the regularization term; represents the i-th sample x i The predicted value; T = {(x1,y1),(x2,y2),...,(x n ,y n )} represents the training set;

[0137] The objective function is optimized by second-order Taylor expansion, regularization term expansion and removal of constant terms to obtain the final objective function:

[0138]

[0139] Among them, g i represents the first-order derivative of the loss function; h i represents the second derivative of the loss function;

[0140] The overall prediction of the multi-market environment is: extract the time series features based on the LSTM framework Results of spatiotemporal feature extraction based on CNN-LSTM framework and other external environmental characteristics Input to the XGBoost model, x i Defined as input fusion feature The predicted value It is expressed as:

[0141]

[0142] The objective function including operating cost and carbon emission cost is expressed as:

[0143]

[0144] Among them, COST in Indicates the cost of purchasing electricity, heat, and natural gas energy; COST om It represents the operation and maintenance cost; Represents the cost of carbon dioxide emissions; COST loss Represents the cost of additional losses.

[0145] Example 3

[0146] This embodiment provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the electronic device executes the method according to Embodiment 1 of the present invention.

[0147] Example 4

[0148] This embodiment provides a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method according to Embodiment 1 of the present invention.

[0149] refer to Figure 3 , a block diagram of an electronic device 400 that can be used as a server or client of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0150] like Figure 3 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0151] A plurality of components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, an output unit 407, a storage unit 408, and a communication unit 409. The input unit 406 may be any type of device capable of inputting information to the electronic device 400, and the input unit 406 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 407 may be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 408 may include, but is not limited to, a disk, an optical disk. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks, and may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0152] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above. For example, in some embodiments, the aforementioned metal corrosion prediction method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. In some embodiments, the computing unit 401 may be configured to perform the aforementioned metal corrosion prediction method in any other appropriate manner (e.g., by means of firmware).

[0153] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0154] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0155] As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0156] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0157] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0158] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0159] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A multi-energy flow planning method for an integrated energy system. It is characterized in that include: Capture the impact of meteorological characteristics in historical data on current electricity prices, carbon prices, and natural gas prices to obtain time series characteristics; Extract the spatial dependencies of multivariate load demand and carbon trading quota data in the target area, and input the results obtained by the CNN framework into the LSTM architecture to obtain spatiotemporal features; Based on the time series characteristics, spatiotemporal characteristics and other external environment characteristics, the XGBoost algorithm is used to generate an overall forecast of the multi-market environment, and an operation strategy is selected based on the results; Based on the forecast results of carbon price, electricity price and natural gas price and the selected operation strategy, the multi-energy flow output of the integrated energy system is determined by an objective function including operation cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

2. The multi-energy flow planning method for an integrated energy system according to claim 1, It is characterized in that The historical data includes at least one of meteorological information of the target area, multiple load demands, carbon trading quota data, electricity price data, carbon price data and natural gas price data; wherein the meteorological information data includes at least one of solar radiation, temperature, humidity, air pressure and wind speed; the multiple loads include at least one of heat, cold, electricity and steam loads.

3. The multi-energy flow planning method for an integrated energy system according to claim 1, It is characterized in that The LSTM model includes cell state c, forget gate f, input gate i and output gate o. The specific formula of the LSTM model is as follows: i t =σ(W i [x t ,h t-1 ]+b i ), f t =σ(W f [x t ,h t-1 ]+b f ), c t =f t *c t-1 +i t tanh(W c [x t ,h t-1 ]+b c ), the t =σ(W o [x] t ,h t-1 ]+b o ), h t =o t *tanh(c t ), Among them, W i , W f , W c , W o Represents the weight matrices of the input gate, forget gate, cell state, and output gate respectively; b i 、b f 、b c 、b o They represent the bias terms of the input gate, forget gate, cell state, and output gate respectively; σ represents the ReLU function; tanh represents the hyperbolic tangent function; h t Represents the hidden layer state.

4. The multi-energy flow planning method for an integrated energy system according to claim 3, It is characterized in that According to the meteorological information at time t~t-ph, the LSTM model is used to capture the impact of meteorological information on electricity price, carbon price and natural gas price. The meteorological information of the region is expressed as: I w =[W t-ph ,W t-(p-1)h ,...,W t ], In=[in temperature ,In windspeed ,...,In visibility ], Where p represents the input sequence I w ∈R p×q The length of ; q represents W∈R q×1 length; W represents the meteorological information of temperature, humidity, air pressure, wind speed and solar radiation in the area.

5. The multi-energy flow planning method for an integrated energy system according to claim 1, It is characterized in that There are four layers in the CNN model: input layer, convolution layer, maximum pooling layer and output layer; Divide a region M into H×W regions based on geographical location, that is, M = {r ij } H×M , so the input of the CNN model is expressed as: Among them, D and E represent the multi-load demand and carbon trading quota of the region, respectively; d and e represent the multi-load demand and carbon trading quota of the sub-region; The input of the CNN-LSTM model is represented as: I s =[[D t-ph ,E t-ph ],[D t-(p-1)h ,E t-(p-1)h ],...,[D t ,E t ]], Among them, I s Represents the CNN-LSTM model input.

6. The multi-energy flow planning method for an integrated energy system according to claim 1, It is characterized in that The XGBoost algorithm is a wig model based on the idea of ​​boosting integration. The forward distribution algorithm is used for greedy learning during training. The objective function of XGBoost consists of two parts: the loss function and the regularization term, which are expressed as: Among them, L(Φ) is the objective function; represents the loss function; represents the regularization term; represents the i-th sample x i The predicted value of 1 ,y 1 ),(x 2 ,y 2 ),...,(x n ,y n )} represents the training set; The objective function is optimized by second-order Taylor expansion, regularization term expansion and removal of constant terms to obtain the final objective function: Among them, g i represents the first-order derivative of the loss function; h i represents the second derivative of the loss function; The overall prediction of the multi-market environment is: extract the time series features based on the LSTM framework Results of spatiotemporal feature extraction based on CNN-LSTM framework and other external environmental characteristics Input to the XGBoost model, x i Defined as input fusion feature The predicted value It is expressed as:

7. The multi-energy flow planning method for an integrated energy system according to claim 1, It is characterized in that The objective function including operating cost and carbon emission cost is expressed as: Among them, COST in Indicates the cost of purchasing electricity, heat, and natural gas energy; COST om It represents the operation and maintenance cost; Represents the cost of carbon dioxide emissions; COST loss Represents the cost of additional losses.

8. A multi-energy flow planning device for an integrated energy system, It is characterized in that include: Time series feature acquisition unit: captures the impact of meteorological features in historical data on current electricity prices, carbon prices, and natural gas prices, and obtains time series features; Spatiotemporal feature acquisition unit: extracts the spatial dependencies of multivariate load demand and carbon trading quota data in the target area, and inputs the results obtained by the CNN framework into the LSTM architecture to obtain spatiotemporal features; Operation strategy selection unit: based on the time series characteristics, spatiotemporal characteristics and other external environment characteristics, the XGBoost algorithm is used to generate an overall forecast of the multi-market environment, and an operation strategy is selected based on the result; Multi-energy flow planning determination unit: Based on the prediction results of carbon price, electricity price, and natural gas price and the selected operation strategy, the multi-energy flow output of the integrated energy system is determined through an objective function that includes operating cost and carbon emission cost, and the optimization algorithm is used to solve it. According to the solution results, the multi-energy flow planning of the integrated energy system is determined.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Smart park comprehensive energy optimization management and control system and management and control method

    CN120069179A