Carbon emission trajectory evaluation and optimization method and system based on ARIMA and GA-BP

Abstract

The invention belongs to the technical field of carbon emission trajectory evaluation and optimization, and discloses a carbon emission trajectory evaluation and optimization method and system based on ARIMA and GA-BP. The two evaluation models are an energy internet carbon emission trajectory evaluation model based on an extended incidence matrix and a whole-process carbon emission trajectory evaluation model based on a comprehensive energy system in the high-carbon emission industry. And an energy internet node carbon emission intensity prediction model based on ARIMA and GA-BP is provided, the model reveals influence factors of node carbon emission intensity, and the carbon emission intensity is accurately predicted. By taking minimization of the whole-process carbon emission trajectory of the high-carbon emission industry as an optimization target and combining the energy consumption characteristics of the flexible load of the high-carbon emission industry, a high-carbon emission industry comprehensive energy system optimization model based on low-carbon demand response is proposed, and response can be made according to the operation state of an external energy internet; and the carbon emission track is minimized by using the flexibility brought by multi-energy complementation, so that deep emission reduction is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon emission trajectory evaluation and optimization, and in particular relates to a carbon emission trajectory evaluation and optimization method and system based on ARIMA and GA-BP. Background Art

[0002] As a typical representative of traditional economic development and energy-dependent production models, high-carbon emission industries are an important target of my country's carbon emission governance. Developing comprehensive transformation plans for high-carbon emission industries that save energy, reduce emissions, and are low-carbon and environmentally friendly is an important breakthrough in achieving the "dual carbon" goals.

[0003] However, at present, high-carbon emission industries, such as electricity, steel, cement, chemicals, transportation, construction and agriculture, have many deficiencies in technology, process and operation mode, which not only lead to high carbon emissions, but also restrict the process of low-carbon transformation. In the past development process, high-carbon emission industries are often accompanied by large amounts of energy consumption and greenhouse gas emissions, which puts great pressure on the environment. By formulating a comprehensive and targeted comprehensive transformation plan, starting from the adjustment of energy utilization structure, upgrading of production processes and optimization of waste treatment, high-carbon emission industries are gradually transformed towards green and low-carbon. Summary of the invention

[0004] The purpose of the present invention is to provide a carbon emission trajectory evaluation and optimization method and system based on ARIMA and GA-BP to address the above-mentioned problems, hoping to improve the problem that the existing high-carbon emission industries have high carbon emissions and put great pressure on the environment.

[0005] The technical solution adopted by the present invention is as follows: A carbon emission trajectory assessment and optimization method and system based on ARIMA and GA-BP, in order to improve the accuracy of carbon emission assessment in high-carbon emission industries, two assessment models are proposed for the quantification of exogenous and endogenous carbon emission trajectories in high-carbon emission industries: one is an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix, and the other is a full-process carbon emission trajectory assessment model based on a comprehensive energy system in high-carbon emission industries. A carbon emission intensity prediction model for energy interconnection network nodes based on ARIMA and GA-BP is proposed, which can reveal the influencing factors of node carbon emission intensity and accurately predict carbon emission intensity. Finally, with the minimization of the carbon emission trajectory of the whole process of high-carbon emission industries as the optimization goal, combined with the energy consumption characteristics of flexible loads in high-carbon emission industries, an optimization model for the comprehensive energy system of high-carbon emission industries based on low-carbon demand response is proposed, which can respond according to the operating status of the external energy interconnection network, and minimize its own carbon emission trajectory by utilizing the flexibility brought by multi-energy complementarity, so as to achieve deep emission reduction.

[0006] Among them, the ARIMA model is an autoregressive integrated moving average model, which is a statistical model used for time series analysis and prediction; GA-BP is a genetic algorithm optimized neural network regression prediction algorithm.

[0007] A carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP, the carbon emission trajectory evaluation and optimization method comprises the following steps: Carbon emission trajectory assessment: construct an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix to assess exogenous carbon emission trajectories; and construct a full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries to assess endogenous carbon emission trajectories; Carbon emission intensity prediction: building an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP to predict the intensity of node carbon emissions; Optimize carbon emission trajectory, build a comprehensive energy system optimization model for high-carbon emission industries based on low-carbon demand response, respond according to the operating status of the external energy interconnection network, and use the flexibility of multi-energy complementarity to minimize its own carbon emission trajectory and carry out deep emission reduction; System integration and effect verification.

[0008] Furthermore, the construction of the energy interconnection network carbon emission trajectory assessment model based on the extended correlation matrix includes the following steps: S1: Constructing an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix: ; S2: Construct an energy internet node carbon emission intensity assessment model based on a unified model of multi-energy flows in the energy internet network, calculate the carbon emissions corresponding to the energy flows through network branches, nodes or energy converter ports, and calculate the node carbon emission intensity; S3: Perform energy flow tracking.

[0009] Furthermore, the step S2 specifically includes the following steps: Expand the association matrix and record the expanded association matrix between nodes as ; ; in, is a node set, Injection branch The energy flow, For branch Xiangzhilu Passing the bypass The energy flow transmitted, that is, the energy flow from the branch Passing the bypass The outgoing energy flow, is the expanded matrix Ai OK j Column elements; Calculate the carbon emissions corresponding to the energy flow through a network branch, node or energy converter port: ; in, For branch roads, For branch of carbon emissions, For branch The carbon emission intensity of For branch of carbon emission energy; Calculate the node carbon emission intensity, which is the weighted average of the carbon intensity of each inflow branch: ; in, For nodes, For Node The carbon emission intensity of Inflow node The index of the branch, For branch roads, For flow through branch The energy flow, for Carbon emission intensity of branch inflows; ; in, For node set To Node The energy injected, For the node To Node Passing the bypass The energy flow injected, Energy injection node Infused energy flow; The carbon emission intensity of each node is equal to the carbon emission intensity of its outflow branch: ; in, Outflow Node The index of the branch, For branch roads, for Carbon emission intensity of branch inflows.

[0010] Furthermore, the construction of a full-process carbon emission trajectory assessment model based on an integrated energy system of a high-carbon emission industry specifically includes the following steps: T1: Construct carbon emission flow models based on energy conversion equipment, including single-input-single-output energy conversion equipment, single-input-multiple-output energy conversion equipment and energy conversion process carbon emission flow models; T2: The carbon emission trajectory assessment process for the entire integrated energy system of high-carbon emission industries first evaluates the carbon emission intensity of energy interconnection network nodes to quantify the exogenous carbon emission trajectory; then quantifies the endogenous carbon emission trajectory, establishes a carbon emission model for energy conversion equipment, and then constructs a carbon emission flow model for the energy conversion process to evaluate the endogenous carbon emission trajectory.

[0011] Furthermore, the construction of the energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP specifically includes the following steps: A method for generating operation data based on MCMC algorithm simulation. Through the MCMC algorithm, simulation is performed based on the system load curve, the output curve of conventional units and the output curve of new energy units to obtain system operation data. The simulated data is substituted into the energy Internet node carbon emission intensity assessment model based on the energy Internet network multi-energy flow unified model to assess the carbon emission intensity of the node. The operation data is obtained by continuous sampling simulation to assess the carbon emission intensity of the node, thereby obtaining a large amount of sample data to provide a training data set for the prediction model. Construct a carbon emission intensity prediction model based on ARIMA and GA-BP; Optimize BP neural network through GA algorithm; Construct a tandem combination model of ARIMA and GA-BP; The carbon emission intensity prediction model based on ARIMA and GA-BP is used to train the samples and output the results of predicting carbon emission intensity.

[0012] Furthermore, the construction of the high-carbon emission industry comprehensive energy system optimization model based on low-carbon demand response specifically includes the following steps: E1: Set the objective function: ; in, For the accounting cycle, For the time period, It is a collection of energy supply units within high carbon emission industries. is the set of energy flow types of the Energy Internet, is the carbon emission trajectory of the internal energy supply unit of the high carbon emission industry in period t, is the carbon emission trajectory generated by the energy flow of the Energy Internet during period t; E2: Minimize the carbon emission trajectory of the entire high carbon emission process during the accounting cycle: ; in, , and for Cooling, electricity and heating loads of the time period, , and is the carbon emission intensity corresponding to the cooling, electric and heating loads in period t, For the accounting cycle, is the time period; E3: Build a low-carbon demand response model and set constraints to find the input and output energy balance relationship of equipment within high-carbon emission industries.

[0013] Furthermore, the system integration and effect verification includes the following steps: Embed assessment, prediction and optimization models into integrated energy systems for high carbon emission industries; Verify the emission reduction effect through historical data backtesting and real-time simulation, and compare the differences in carbon emission trajectories before and after optimization; Dynamically adjust model parameters to adapt to external environment updates.

[0014] Furthermore, a carbon emission trajectory evaluation and optimization system based on ARIMA and GA-BP includes: The first building module is used to construct an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix; The second building block is used to construct a full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries; The third building module is used to build a carbon emission intensity prediction model for energy interconnection network nodes based on ARIMA and GA-BP; The fourth building block is used to construct an integrated energy system optimization model for high-carbon emission industries based on low-carbon demand response; The first building block is connected to the second building block, the third building block and the fourth building block in sequence.

[0015] Furthermore, a processor is provided, wherein the processor is used to run a program, wherein when the program is run, the carbon emission trajectory assessment and optimization method based on ARIMA and GA-BP described in any one of claims 1 to 7 is executed.

[0016] Furthermore, a computer-readable storage medium includes a stored program, wherein the program controls the device where the computer-readable storage medium is located to execute the carbon emission trajectory assessment and optimization method based on ARIMA and GA-BP as described in any one of claims 1-7 when running.

[0017] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention is based on the carbon emission trajectory evaluation and optimization method of ARIMA and GA-BP. On the basis of the unified modeling of multi-energy flows in the energy interconnection network, the energy interconnection network carbon emission trajectory evaluation model based on the extended correlation matrix is ​​adopted to effectively evaluate its carbon emission trajectory and realize the quantification of the exogenous carbon emission trajectory. A carbon emission flow model based on energy conversion equipment is constructed to reveal the distribution law of carbon emission trajectories among energy conversion equipment, thereby realizing the quantification of endogenous carbon emission trajectories. Finally, the evaluation model of exogenous and endogenous carbon emission trajectories is comprehensively considered to quantify the carbon emissions of the entire energy consumption process of the comprehensive energy system in high-carbon emission industries, thereby improving the accuracy of carbon emission assessment in high-carbon emission industries.

[0018] At the same time, combined with the energy interconnection network carbon emission trajectory assessment model based on the extended correlation matrix, the full-process carbon emission trajectory assessment model of the integrated energy system of high-carbon emission industries, and the energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP, the influencing factors of node carbon emission intensity are revealed, thereby accurately predicting the carbon emission intensity of energy interconnection network nodes.

[0019] The present invention can also take into account the predicted node carbon emission intensity to a certain extent, respond to the comprehensive energy system optimization model of high-carbon emission industries based on low-carbon demand, respond according to the operating status of the external energy Internet network, adjust internal flexibility resources, and optimize the endogenous carbon emission trajectory, ultimately achieving controllable carbon emission trajectory of the entire process of high-carbon emission industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 It is a block diagram of the energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP of the present invention; Figure 3 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0021] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0022] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0023] like Figure 1 As shown, the carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP includes the following steps: Carbon emission trajectory assessment: construct an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix to assess exogenous carbon emission trajectories; and construct a full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries to assess endogenous carbon emission trajectories; Carbon emission intensity prediction: building an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP to predict the intensity of node carbon emissions; Optimize carbon emission trajectory, build a comprehensive energy system optimization model for high-carbon emission industries based on low-carbon demand response, respond according to the operating status of the external energy interconnection network, and use the flexibility of multi-energy complementarity to minimize its own carbon emission trajectory and carry out deep emission reduction; System integration and effect verification.

[0024] This method constructs an energy internet carbon emission trajectory assessment model based on an extended correlation matrix and a full-process carbon emission trajectory assessment model based on an integrated energy system for high-carbon emission industries, taking into account the quantification of exogenous and endogenous carbon emission trajectories, and further improves the prediction accuracy of carbon emission intensity through the combination of ARIMA and GA-BP, minimizing the carbon trajectory through multi-energy complementarity and demand response, and promoting the low-carbon transformation of high-carbon industries.

[0025] Example 1 like Figure 1 As shown, one embodiment of the present invention is a carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP, the method comprising: constructing an energy interconnection network carbon emission trajectory evaluation model based on an extended correlation matrix; constructing a full-process carbon emission trajectory evaluation model based on a comprehensive energy system of a high-carbon emission industry; constructing an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP; constructing a high-carbon emission industry comprehensive energy system optimization model based on low-carbon demand response; including, constructing a mathematical model based on high-carbon emission industry comprehensive energy system equipment; constructing a carbon emission trajectory optimization model based on a high-carbon emission industry.

[0026] The construction of the energy interconnection network carbon emission trajectory assessment model based on the extended correlation matrix includes: the exogenous carbon emission trajectory is the carbon emission trajectory of the high carbon emission industry drawing energy flow from the energy interconnection network (including the distribution network, regional heating network and natural gas system). The construction of a unified model of multi-energy flow based on the energy interconnection network; the construction of a network node carbon emission intensity assessment model based on the unified model of multi-energy flow of the energy interconnection network.

[0027] The whole process carbon emission trajectory assessment model based on the integrated energy system of high carbon emission industries is constructed, including: the endogenous carbon emission trajectory is the carbon emission trajectory of the virtual carbon flow transmission of the energy conversion equipment in high carbon emission industries (divided into two categories: single input-single output, single input-multiple output). Construct a carbon emission flow model based on energy conversion equipment. Combining exogenous and endogenous carbon emission trajectories, the whole process carbon emission trajectory of the integrated energy system of high carbon industries can be observed.

[0028] The method comprises: using Markov Chain Monte Carlo (MCMC) simulation to generate operating data (system load, conventional unit output, and new energy unit output curve) to mine factors affecting node carbon emission intensity, using ARIMA and GA-BP neural networks to map the nonlinear relationship between carbon emission intensity and energy supply unit output and energy demand data, and constructing a carbon emission intensity prediction model based on ARIMA and GA-BP.

[0029] The construction of an integrated energy system optimization model for high-carbon emission industries based on low-carbon demand response includes: constructing a mathematical model based on integrated energy system equipment for high-carbon emission industries; and constructing an optimization model for carbon emission trajectories for high-carbon emission industries.

[0030] The mathematical model of the comprehensive energy system equipment in high-carbon emission industries is constructed; the carbon emission trajectory optimization model in high-carbon emission industries is constructed, including: the mathematical model of the comprehensive energy system equipment in high-carbon emission industries includes the power supply system (photovoltaic, wind turbines, energy interconnection network injection power model), the heating system (cogeneration unit, heating equipment model), and the energy storage equipment model. The carbon emission trajectory optimization model takes minimizing the carbon emission trajectory as the objective function, and the low-carbon demand response model includes two types of loads that can be reduced and transferred, and the input and output of each energy are constraints.

[0031] Example 2 Another embodiment of the present invention is that the construction of the energy interconnection network carbon emission trajectory assessment model based on the extended correlation matrix includes the following steps: S1: Constructing an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix: Due to the different characteristics of electricity, heat and gas energy networks and the complex energy coupling relationship, the unified model of multi-energy flow based on energy interconnection network is unified into ; S2: Construct an energy internet node carbon emission intensity assessment model based on a unified model of multi-energy flows in the energy internet network, calculate the carbon emissions corresponding to the energy flows through network branches, nodes or energy converter ports, and calculate the node carbon emission intensity; The step S2 specifically includes the following steps: Expand the association matrix and record the expanded association matrix between nodes as ; ; in, is a node set, Injection branch The energy flow, For branch Xiangzhilu Passing the bypass The energy flow transmitted, is the expanded matrix A i OK j Column elements; matrix It has the following properties: (1) Node The load power is equal to the matrix No. The sum of the row elements; (2) Node The injected energy flow is equal to the matrix No. The sum of the column elements; (3) Matrix is a full-rank matrix, that is, a reversible matrix.

[0032] Calculate the carbon emissions corresponding to the energy flow through a network branch, node or energy converter port: ; in, For branch roads, For branch of carbon emissions, For branch The carbon emission intensity of For branch of carbon emission energy; Calculate the node carbon emission intensity, which is the weighted average of the carbon intensity of each inflow branch: ; in, For nodes, For Node The carbon emission intensity of Inflow node The index of the branch, For branch roads, For flow through branch The energy flow, for Carbon emission intensity of branch inflows; ; in, For node set To Node The energy injected, For the node To Node Passing the bypass Infused energy flow; Energy injection node The injected energy flow, if the node is not connected to an energy injection device, its value is 0; The carbon emission intensity of each node is equal to the carbon emission intensity of its outflow branch: ; in, Outflow Node The index of the branch, For branch roads, for Carbon emission intensity of branch inflows.

[0033] S3: Perform energy flow tracking.

[0034] Assume that the symmetric matrix of the energy flow injected by the energy injection node is , and the corresponding energy injection node carbon emission intensity is , so we have: ; ; in, Carbon emissions from energy injection nodes, Infusing equipment with energy The energy flow, for n -dimensional unit column vector, ; Combining the two equations, we can get: ; ; in, is the load power column vector; the above formula is used to reveal the relationship between the energy flow injected by the energy injection node and the load power, which can realize energy flow tracking. Similarly, combining the two formulas, we can get: ; in, is the node carbon emission intensity; the above formula is used to reveal the relationship between the carbon emissions of the energy injection node and the carbon emissions of the load node, and can realize the network carbon emission flow tracking.

[0035] definition is the carbon emission distribution coefficient matrix, denoted as , is the carbon emission distribution coefficient matrix D i OK j The column elements are calculated as: ; In summary, the carbon emissions on the power generation side can be allocated to the load side. The carbon emission calculation formula for the carbon emission flow corresponding to the energy flow during energy transfer is: ; In the formula, Represents the energy injection node In the transmission network Carbon emissions allocated to is the carbon emission distribution coefficient matrix D i OK j Column elements, For transmission network jk The energy flow, For Node i carbon emission intensity.

[0036] By establishing the topological structure of the energy interconnection network, clarifying the correlation between each node, designing an extended correlation matrix, quantifying the exogenous carbon emission trajectory, analyzing the carbon emission contribution of each node in the network through matrix operations, and identifying key carbon emission paths.

[0037] Example 3 Another embodiment of the present invention is that the construction of a full-process carbon emission trajectory assessment model based on an integrated energy system of a high-carbon emission industry specifically includes the following steps: T1: Construct carbon emission flow models based on energy conversion equipment, including single-input-single-output energy conversion equipment, single-input-multiple-output energy conversion equipment and energy conversion process carbon emission flow models; (1) Single-input-single-output energy conversion equipment; According to the law of conservation of carbon emissions, all carbon emissions related to input energy should be allocated to output energy, so: ; in, is the index of the single-input-single-output energy conversion device. The carbon emission intensity of the input and output ports of the single-input-single-output energy conversion device is expressed as and The energy of the input and output ports is expressed as and express; The conversion efficiency of a single-input-single-output energy conversion device is ,Right now ; The carbon emission intensity of the input and output ports of a single-input-single-output type energy conversion device can be expressed as: ; (2) Single-input-multiple-output energy conversion equipment; ; in, is the carbon emission intensity of the input port of the cogeneration unit, and the electrical power and thermal power of the output port of the cogeneration unit are respectively and express; and They represent the carbon emission intensity corresponding to the electrical power and thermal power output ports of the cogeneration unit, The carbon emission intensity of the input port of the cogeneration unit; (3) Carbon emission flow model of energy conversion process; The energy conversion process of high carbon emission industries is modeled as ports with multiple inputs and outputs, using the energy coupling matrix Describes the steady-state relationship between energy input-output vectors: ; The input and output port energy vectors are respectively given by and express; Defining the carbon emission intensity coupling matrix To describe the carbon emission relationship between input and output ports: ; in, and is the carbon emission intensity vector of input and output ports;

[0038] in, is the carbon emission intensity coupling matrix E Elements of describe the carbon emission relationship between input and output ports of each node; The carbon emissions of each output port are: ; T2: The carbon emission trajectory assessment process for the entire integrated energy system of high-carbon emission industries first evaluates the carbon emission intensity of energy interconnection network nodes to quantify the exogenous carbon emission trajectory; then quantifies the endogenous carbon emission trajectory, establishes a carbon emission model for energy conversion equipment, and then constructs a carbon emission flow model for the energy conversion process to evaluate the endogenous carbon emission trajectory.

[0039] Example 4 like Figure 2As shown, another embodiment of the present invention is to construct an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP; (1) Method for generating operation data based on MCMC algorithm simulation; The MCMC algorithm is used to simulate the system load curve, the output curve of conventional units and the output curve of new energy units to obtain the system operation data. The simulated data is substituted into the energy Internet node carbon emission intensity assessment model based on the energy Internet network multi-energy flow unified model to assess the carbon emission intensity of the node. The operation data is obtained by continuous sampling simulation to assess the carbon emission intensity of the node, thereby obtaining a large amount of sample data to provide a training data set for the prediction model. (2) Construct a carbon emission intensity prediction model based on ARIMA and GA-BP; The ARIMA model is a model that uses a linear combination of historical values ​​of variables and past errors to represent future values. The ARIMA model is also written as : ; ; in, Unstable, is stable process; The ARMA model considered as an integrated ), whose three parameters are: is the number of differences, , are the orders of autoregression and moving average, respectively.

[0040] for , that is, lag 1 to p The original sequence observation value of order; For stable process( go through d sub-difference); is the autoregression coefficient (AR coefficient), which represents the lagged observation value For the current value The weight of influence; Lag 1 to q White noise error term of order; It is the moving average coefficient (MA coefficient), which indicates the past 1 to q The error term of For the current value The weight of influence; is an autoregressive operator polynomial, defined as ,in Bis a backshift operator (e.g. ); is the moving average operator polynomial, defined as , which is used to describe the lag structure of the error term.

[0041] GA algorithm optimizes BP neural network: ① First, we need to determine the topological structure of the network. The number of neurons in the input layer and output layer of the BP neural network is consistent with the number of elements in the input and output vectors, respectively, while the number of neurons in the hidden layer is obtained by manual trial calculation.

[0042] ②First, based on the network structure determined by ①, the total number of all the network's computational neuron parameters (weights and thresholds) can be obtained. Then, all the weights and thresholds are arranged and combined and encoded into a binary string, so that the weight and threshold of each node correspond to the corresponding fragment on the binary string. Finally, an initialization population is randomly generated through the binary string.

[0043] ③ Several subsets containing all node weights and threshold information can be obtained by decoding the generated initial population. After inputting these subset information data into the neural network, the neural network is trained and tested using training samples and test samples respectively, and the training error and test error of the model can be obtained.

[0044] A. For the initial BP neural network, first use the training set to train to get each true value The corresponding predicted value , and then define the network training error in the form of mean square error, and the calculation formula is as follows:

[0045] in, is the network training error, is the number of samples in the training set; B. Define the neural network test error. For the BP neural network that has been trained in ①, use the test set to test and get each true value The corresponding predicted value , and finally define the test error in the form of mean square error, which is calculated as follows: ; in, is the number of test set samples, Test error for the neural network; ④ First, calculate the fitness of the population based on the test error. The individuals with larger fitness values ​​have higher probability of being retained. Then perform GA genetic operation to obtain a new generation of population. After decoding the new population, substitute the information into the neural network, and the network enters a new round of training and testing to obtain new test errors and population fitness. This cycle repeats until the test error is lower than the expected value or the system cycle reaches the preset iteration limit.

[0046] ⑤ Output the optimal individual and decode it to get the ideal network initial weights and thresholds.

[0047] (3) Serial combination model of ARIMA and GA-BP The fitted value of the ARIMA model is used as the input of the GA-BP neural network, and the corresponding measured data is used as its output. Then the GA-BP neural network is trained, so that a network that reflects the deviation between the ARIMA fitted value and the actual value can be obtained. Finally, the predicted value of the ARIMA model at the next moment is input into the trained GA-BP neural network, and the network output is the final prediction result of the combined model.

[0048] (4) Carbon emission intensity prediction model training process based on ARIMA and GA-BP ① Taking the system operation data obtained by MCMC simulation method as input parameters and the carbon emission intensity obtained by the evaluation model as output parameters, the training samples of ARIMA and GA-BP neural networks were constructed.

[0049] ② Adjust the learning parameters and input the training samples into the ARIMA and GA-BP neural network model to obtain the mapping relationship between input and output.

[0050] ③The test process is as follows: Input the predicted system operation data into the ARIMA and GA-BP neural network models to obtain the predicted carbon emission intensity ; At the same time, the predicted system operation data is brought into the evaluation model to obtain the carbon emission intensity evaluation value . The root mean square error (RMSE) and mean absolute error (MAE) were selected as evaluation indicators.

[0051] ; ; In the formula, and represent the root mean square error and mean absolute error, respectively.

[0052] Example 5 Another embodiment of the present invention is to construct an integrated energy system optimization model for high carbon emission industries based on low carbon demand response; (1) Constructing a mathematical model for integrated energy system equipment in high-carbon emission industries 1. Mathematical model of main equipment of power supply system The power supply of the power supply system is divided into: electric power drawn from the energy Internet, distributed new energy units and cogeneration units.

[0053] ①Energy Internet Injection Power Model The electric power obtained from the upper power grid can be expressed as: ; In the formula, is the electric power drawn from the energy internet by high carbon emission industries, and its value is .

[0054] The natural gas obtained from the upper gas grid can be expressed as: ; In the formula, The natural gas drawn from the energy internet for high carbon emission industries is worth .

[0055] ② Wind turbine model ; In the formula, , , , They are real-time, cut-in, cut-out and rated wind speed; , are the real-time output and rated output of the wind turbine respectively; a, b, c represent the fixed parameters of the wind turbine.

[0056] ③ Photovoltaic generator model ; In the formula, , They respectively represent the real-time output and rated output of the photovoltaic generator set.

[0057] 2. Mathematical model of heating system The heat sources of the heating system are divided into: cogeneration units, electric boilers, auxiliary boilers and heat pumps.

[0058] ① Mathematical model of cogeneration unit The electrical efficiency of a combined heat and power unit is Combined heat and power generation, thermal efficiency is Combined heat and power. Indicates; the electrical power and thermal power output of the cogeneration unit at time t are , ; The maximum ramp rate of the combined heat and power unit is The upper and lower limits of electric power are , .

[0059] ; It means that the electrical and thermal power output of the CHP unit are coupled through a fixed heat-to-electricity ratio; ; They are the ramping constraint and electric power output limit constraint of the cogeneration unit respectively.

[0060] ②Heating equipment model ; In the formula, the gas consumption power and heating power of the auxiliary boiler are expressed as , ; The gas-to-heat efficiency of the auxiliary boiler is expressed as .

[0061] 3. Mathematical model of energy storage devices Since high-carbon emission industries are equipped with both heat storage and gas storage energy storage devices and the models are similar, a unified model can be used for modeling.

[0062] ; ; ; ; ; In the formula, is the energy stored at time t; the power charged and released at time t is respectively , The charging and discharging efficiencies are , ; is a binary variable of charging and discharging status, Represents charge, represents release; , are the maximum values ​​of charging and releasing power respectively; , are the minimum and maximum values ​​of the energy stored in the energy storage device.

[0063] (2) Construct a carbon emission trajectory optimization model based on high carbon emission industries; E1: Set the objective function: ; in, For the accounting cycle, For the time period, It is a collection of energy supply units within high carbon emission industries. is the set of energy flow types of the Energy Internet, is the carbon emission trajectory of the internal energy supply unit of the high carbon emission industry in period t, is the carbon emission trajectory generated by the energy flow of the Energy Internet during period t; The carbon emission trajectory optimization model for high-carbon emission industries aims to minimize the carbon emission trajectory of high-carbon emission industries. The carbon emission trajectory here includes the carbon emission trajectory from electricity and natural gas from the energy Internet, as well as the carbon emission trajectory generated by energy production units within the high-carbon emission industries.

[0064] E2: Energy storage equipment in high-carbon emission industries will store some energy, and the responsibility for the carbon emission trajectory corresponding to this unconsumed energy should not be borne by high-carbon emission industries. Therefore, the essence of minimizing the carbon emission trajectory of high-carbon emission industries is to minimize the carbon emission trajectory of the entire energy consumption process within the accounting cycle, which can be equivalently converted into: ; in, , and for Cooling, electricity and heating loads of the time period, , and is the carbon emission intensity corresponding to the cooling, electric and heating loads in period t, For the accounting cycle, is the time period; set up for Equipment The predicted value of carbon emission intensity at each moment, Its energy conversion efficiency is is the carbon emission intensity of the output port of the device at time t. Then, the relationship between the carbon emission intensity of the ports of single-input-single-output and single-input-multiple-output energy conversion devices is as shown in the formula: ; E3: Build a low-carbon demand response model and set constraints to find the input and output energy balance relationship of equipment within high-carbon emission industries.

[0065] Demand response models are divided into two categories: load curtailment (LC) and load shifting (LS).

[0066] ; In the formula, For the LC type demand response The load reduction at the time, Indicates LC type demand response The maximum value of load reduction at any moment; Indicates LS type demand response The amount of transfer at a time, For the LS type demand response The maximum value of the transfer amount at any time.

[0067] The constraints are as follows: There is a coupling relationship between the various devices in the high-carbon emission industry, based on which the input-output energy balance relationship can be obtained: ; In the formula, is the energy output matrix; is the energy coupling matrix; is the energy input matrix. for: ; Total electricity consumption of high carbon emission industries Indicates that the total natural gas volume Indicates that the energy Internet grid, photovoltaic generators within high-carbon emission industries, and wind turbines provide electricity to high-carbon emission industries respectively. , and ; The amount of natural gas provided by the Energy Internet natural gas network and the internal gas storage equipment of the high-carbon emission industry is and .

[0068] Combining the conversion parameters of various energy conversion equipment within the high-carbon emission industry, it can be written in the form of: ; In the formula, , and They represent the cooling, electricity and heating loads of high carbon emission industries respectively; and are the proportions of the electricity allocated to the compression refrigeration unit and the electrical load in the total electricity; is the share of natural gas allocated to CHP units in the total natural gas volume; is the proportion of the power allocated to the compression refrigeration unit in the total power output of the cogeneration unit, and , , , All are in the range [0,1]; is the conversion efficiency of the compression refrigeration unit; Combined heat and power Combined heat and power are the electricity and heat conversion efficiency of the combined heat and power unit respectively; Heating efficiency for HP units; is the heat conversion efficiency of AB unit; and They are LC and The load reduction and transfer amount of LS type demand response.

[0069] ; The above formula represents the upper and lower limit constraints of the electricity load and heat load in high-carbon emission industries. , , and They are the minimum and maximum values ​​of electricity and heat loads in high carbon emission industries, respectively.

[0070] Example 6 like Figure 3 As shown, another embodiment of the present invention is a carbon emission trajectory assessment and optimization system based on ARIMA and GA-BP, including: a first building module for constructing an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix; a second building module for constructing a full-process carbon emission trajectory assessment model based on a high-carbon emission industry integrated energy system; a third building module for constructing an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP; a fourth building module for constructing a high-carbon emission industry integrated energy system optimization model based on low-carbon demand response; including constructing a mathematical model based on high-carbon emission industry integrated energy system equipment; and constructing a carbon emission trajectory optimization model based on high-carbon emission industries. Each of the above modules can be implemented by software or hardware.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP, characterized by: The carbon emission trajectory assessment and optimization method comprises the following steps: Carbon emission trajectory assessment: construct an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix to assess exogenous carbon emission trajectories; and construct a full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries to assess endogenous carbon emission trajectories; Carbon emission intensity prediction: building an energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP to predict the intensity of node carbon emissions; Optimize carbon emission trajectory, build a comprehensive energy system optimization model for high-carbon emission industries based on low-carbon demand response, respond according to the operating status of the external energy interconnection network, and use the flexibility of multi-energy complementarity to minimize its own carbon emission trajectory and carry out deep emission reduction; System integration and effect verification.

2. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 1, characterized in that: The construction of the energy interconnection network carbon emission trajectory assessment model based on the extended correlation matrix includes the following steps: S1: Constructing an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix: ; S2: Construct an energy internet node carbon emission intensity assessment model based on a unified model of multi-energy flows in the energy internet network, calculate the carbon emissions corresponding to the energy flows through network branches, nodes or energy converter ports, and calculate the node carbon emission intensity; S3: Perform energy flow tracking.

3. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 2 is characterized in that: The step S2 specifically includes the following steps: Expand the association matrix and record the expanded association matrix between nodes as ; ； in, is a node set, Injection branch The energy flow, For branch Xiangzhilu Passing the bypass The energy flow transmitted, is the expanded matrix A i OK j Column elements; Calculate the carbon emissions corresponding to the energy flow through a network branch, node or energy converter port: ； in, For branch roads, For branch of carbon emissions, For branch The carbon emission intensity of For branch of carbon emission energy; Calculate the node carbon emission intensity, which is the weighted average of the carbon intensity of each inflow branch: ； in, For nodes, For Node The carbon emission intensity of Inflow node The index of the branch, For branch roads, For flow through branch The energy flow, for Carbon emission intensity of branch inflows; ； in, For node set To Node The energy injected, For the node To Node Passing the bypass The energy flow injected, Energy injection node Infused energy flow; The carbon emission intensity of each node is equal to the carbon emission intensity of its outflow branch: ； in, Outflow Node The index of the branch, For branch roads, for Carbon emission intensity of branch inflows.

4. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 1, characterized in that: The construction of the full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries specifically includes the following steps: T1: Construct carbon emission flow models based on energy conversion equipment, including single-input-single-output energy conversion equipment, single-input-multiple-output energy conversion equipment and energy conversion process carbon emission flow models; T2: The carbon emission trajectory assessment process for the entire integrated energy system of high-carbon emission industries first evaluates the carbon emission intensity of energy interconnection network nodes to quantify the exogenous carbon emission trajectory; then quantifies the endogenous carbon emission trajectory, establishes a carbon emission model for energy conversion equipment, and then constructs a carbon emission flow model for the energy conversion process to evaluate the endogenous carbon emission trajectory.

5. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 1, characterized in that: The construction of the energy interconnection network node carbon emission intensity prediction model based on ARIMA and GA-BP specifically includes the following steps: A method for generating operation data based on MCMC algorithm simulation. Through the MCMC algorithm, simulation is performed based on the system load curve, the output curve of conventional units and the output curve of new energy units to obtain system operation data. The simulated data is substituted into the energy Internet node carbon emission intensity assessment model based on the energy Internet network multi-energy flow unified model to assess the carbon emission intensity of the node. The operation data is obtained by continuous sampling simulation to assess the carbon emission intensity of the node, thereby obtaining a large amount of sample data to provide a training data set for the prediction model. Construct a carbon emission intensity prediction model based on ARIMA and GA-BP; Optimize BP neural network through GA algorithm; Construct a tandem combination model of ARIMA and GA-BP; The carbon emission intensity prediction model based on ARIMA and GA-BP is used to train the samples and output the results of predicting carbon emission intensity.

6. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 1, characterized in that: The construction of the high-carbon emission industry comprehensive energy system optimization model based on low-carbon demand response specifically includes the following steps: E1: Set the objective function: ； in, For the accounting cycle, For the time period, It is a collection of energy supply units within high carbon emission industries. is the set of energy flow types of the Energy Internet, is the carbon emission trajectory of the internal energy supply unit of the high carbon emission industry in period t, is the carbon emission trajectory generated by the energy flow of the Energy Internet during period t; E2: Minimize the carbon emission trajectory of the entire high carbon emission process during the accounting cycle: ； in, , and for Cooling, electricity and heating loads of the time period, , and is the carbon emission intensity corresponding to the cooling, electric and heating loads in period t, For the accounting cycle, is the time period; E3: Build a low-carbon demand response model and set constraints to find the input and output energy balance relationship of equipment within high-carbon emission industries.

7. The carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to claim 1, characterized in that: The system integration and effect verification includes the following steps: Embed assessment, prediction and optimization models into integrated energy systems for high carbon emission industries; Verify the emission reduction effect through historical data backtesting and real-time simulation, and compare the differences in carbon emission trajectories before and after optimization; Dynamically adjust model parameters to adapt to external environment updates.

8. A carbon emission trajectory evaluation and optimization system based on ARIMA and GA-BP, using the carbon emission trajectory evaluation and optimization method based on ARIMA and GA-BP according to any one of claims 1 to 7, characterized in that: The system comprises: The first building module is used to construct an energy interconnection network carbon emission trajectory assessment model based on an extended correlation matrix; The second building block is used to construct a full-process carbon emission trajectory assessment model based on the integrated energy system of high-carbon emission industries; The third building module is used to build a carbon emission intensity prediction model for energy interconnection network nodes based on ARIMA and GA-BP; The fourth building block is used to construct an integrated energy system optimization model for high-carbon emission industries based on low-carbon demand response; The first building block is connected to the second building block, the third building block and the fourth building block in sequence.

9. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the carbon emission trajectory assessment and optimization method based on ARIMA and GA-BP according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program, when running, controls the device where the computer-readable storage medium is located to execute the carbon emission trajectory assessment and optimization method based on ARIMA and GA-BP described in any one of claims 1 to 7.

Images