Multi-stage optimized industrial park carbon emission reduction method and device

Abstract

The invention discloses a multi-stage optimized industrial park carbon emission reduction method and device, and the method comprises the steps: building a stepped carbon transaction cost model according to a carbon index and an actual carbon emission amount; integrating a total cost minimization objective function of investment, operation, maintenance and stepped carbon transaction cost; according to an XGBoost-PSO hybrid algorithm, optimizing a total cost minimization objective function; and analyzing a carbon emission reduction effect based on the stepped carbon transaction cost model and the optimized total cost minimization objective function to obtain an industrial park carbon emission reduction control strategy. The device comprises a processor and a memory. On the premise of ensuring economic development, the industrial structure and the energy structure are optimized, energy conservation and emission reduction of the industrial park are stimulated, and high-quality development of the park is promoted.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission reduction, and particularly to a multi-stage optimized carbon emission reduction method and device for industrial parks. Background Art

[0002] Industrial parks are key carriers for improving industrial energy efficiency. Promoting the concentration of industrial enterprises in parks can significantly improve production efficiency and energy efficiency. However, the energy consumption and carbon emissions of industrial parks also increase accordingly.

[0003] Currently, the construction timing and carbon trading mechanism are widely applied to the research of energy consumption and carbon emissions in parks. Domestic and foreign research shows that introducing construction timing into the planning of the park's integrated energy system, that is, optimizing the equipment configuration in each planning stage, can effectively avoid the problems of redundant and aging park equipment to meet the load growth demand. In the multi-stage planning of the park's integrated energy system based on the stepped carbon trading mechanism, a stepped carbon trading mechanism is established according to the relationship between actual carbon emissions and carbon quotas, which has a positive impact on the low-carbon and economic performance of the park system operation. The carbon emission reduction work in industrial parks should, on the basis of accurately grasping the carbon emission reduction effect, set carbon emission reduction paths according to the characteristics of the carbon emission reduction effect clustering type to provide a basis for carbon emission reduction in different stages.

[0004] However, the research on the carbon emission measurement system in the carbon emission industry and region is still in its infancy. At the same time, there are local optimum problems in the process algorithm, and there is also less research on construction timing and carbon indicators. In summary, the multi-stage comprehensive optimization method for carbon emission reduction for economic development is not yet perfect, and it is difficult to realize the analysis of the decarbonization trend. Summary of the Invention

[0005] The present invention provides a multi-stage optimized carbon emission reduction method and device for industrial parks. The present invention comprehensively considers construction timing and carbon indicators, and uses an optimization algorithm to solve the local optimum problem, improving the carbon emission analysis and prediction capabilities of high-carbon emission industries and regions, and improving the accuracy of the decarbonization trend analysis technology; on the premise of ensuring economic development, optimizing the industrial structure and energy structure, motivating industrial parks to save energy and reduce emissions, and promoting the high-quality development of parks, as described in detail below:

[0006] A multi-stage optimized carbon emission reduction method for industrial parks, the method includes:

[0007] Establish a stepped carbon trading cost model according to carbon indicators and actual carbon emissions;

[0008] Integrate the total cost minimization objective function of investment, operation, maintenance and stepped carbon trading costs; optimize the total cost minimization objective function according to the XGBoost-PSO hybrid algorithm;

[0009] Analyze the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function to obtain the carbon emission reduction control strategy for the industrial park.

[0010] Among them, the analysis of the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function is as follows:

[0011] Calculate the carbon emission reduction effect results at different stages based on the stepped carbon trading cost model, the optimized total cost minimization objective function, and the Tapio decoupling model;

[0012] Integrate DTW-K-means clustering to process the carbon emission reduction effect results, realize different carbon emission reduction paths, and finally obtain the carbon emission reduction control strategy for the industrial park.

[0013] Among them, the specific method for establishing the stepped carbon trading cost model according to the carbon index and the actual carbon emissions is as follows:

[0014] Carbon index:

[0015]

[0016] In the formula, E * , respectively represent the carbon indexes of the industrial park integrated energy system, system purchased electricity, cogeneration equipment, and gas boiler; is the carbon index per unit of electricity; P GRID,t is the purchased electricity power of the system from the superior power grid at time t; is the carbon index per unit of heat; P CHP,t and H CHP,t are the electric and heat output powers of cogeneration at time t respectively; H GB,t is the heat output of the gas boiler at time t; is the conversion coefficient of the power generation of the cogeneration equipment to the heat generation;

[0017] Actual carbon emissions:

[0018] E = E GRID + E CHP + E GB - E P2G

[0019]

[0020] In the formula, E, E GRID , E CHP , E GB respectively represent the actual carbon emissions of the industrial park integrated energy system, purchased electricity, cogeneration equipment, and gas boiler; E P2G is the CO captured by the power-to-gas equipment 2Quantity; β g is the carbon capture coefficient, which is the amount of CO required for the power - to - gas equipment to convert a unit of electricity 2 Quantity; P P2G,t is the electrical input power of the power - to - gas equipment at time t;

[0021] Step - type carbon trading cost model:

[0022] E IPIES = E - E *

[0023] Step - type carbon trading cost is:

[0024]

[0025] In the formula, μ is the basic carbon trading price; τ is the price growth rate; l is the length of the carbon emission interval. When E is less than E * , E IPIES is less than 0, indicating that the actual carbon emissions in the park are less than the carbon quota.

[0026] Among them, the total - cost minimization objective function integrating investment, operation, maintenance and step - type carbon trading cost is:

[0027] Investment: Determined by photovoltaic, energy conversion and energy storage equipment, the investment cost F EI,s is:

[0028]

[0029] In the formula, E PV is the unit - capacity construction cost of the photovoltaic equipment; E PV,s is the configured capacity of the photovoltaic equipment in the s - th planning stage; Π is the set of energy conversion equipment; F i is the unit - capacity construction cost of the i - th type of energy conversion equipment; E i,s is the configured capacity of the i - th type of energy conversion equipment in the s - th planning stage; F j is the unit - capacity construction cost of the j - th type of energy storage equipment; E j,s is the capacity configuration of the j - th type of energy storage equipment in the s - th planning stage;

[0030] Operation:

[0031]

[0032] In the formula, f e,t and f g,t are the electricity price and gas price at time t respectively; E GAS,n,t is the gas purchase power of the park's integrated energy system from the superior gas network at time t in the n - th year; is the penalty cost per unit of abandoned wind volume; and P PV,n,t are the maximum and actual power outputs of photovoltaics in the t period of the nth year, respectively;

[0033] Maintenance:

[0034]

[0035] The total cost objective function;

[0036]

[0037] In the formula, γ represents the leaf tree penalty coefficient; T is the number of tree leaf nodes; ω is the leaf weight; λ is the weight penalty coefficient.

[0038] Among them, the Tapio decoupling model:

[0039]

[0040] In the formula, T represents the decoupling index, △C and △F respectively represent the changes in carbon emissions and minimized cost at the end period relative to the base period, and C and F are the carbon emissions and minimized cost at the end period respectively.

[0041] Second aspect, a multi-stage optimized carbon emission reduction device for industrial parks, the device includes:

[0042] A building module, configured to establish a stepped carbon trading cost model according to carbon indicators and actual carbon emissions;

[0043] An optimization module, configured to integrate the total cost minimization objective function of investment, operation, maintenance, and stepped carbon trading costs; optimize the total cost minimization objective function according to the XGBoost-PSO hybrid algorithm;

[0044] An acquisition module, configured to analyze the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function, and obtain the carbon emission reduction control strategy for the industrial park.

[0045] Among them, the acquisition module is:

[0046] A calculation sub-module, configured to calculate the carbon emission reduction effect results in different stages based on the stepped carbon trading cost model, the optimized total cost minimization objective function, and the Tapio decoupling model;

[0047] A processing sub-module, configured to fuse DTW-K-means clustering, process the carbon emission reduction effect results, realize a differentiated carbon emission reduction path, and finally obtain the carbon emission reduction control strategy for the industrial park.

[0048] Third aspect: A multi-stage optimized carbon emission reduction device for industrial parks, the device comprising: a processor and a memory, wherein program instructions are stored in the memory, and the processor calls the program instructions stored in the memory to enable the device to execute the method described in any one of the first aspect.

[0049] Fourth aspect: A computer-readable storage medium storing a computer program, the computer program comprising program instructions, and when the program instructions are executed by a processor, the processor is caused to execute the method described in any one of the first aspect.

[0050] The beneficial effects of the technical solution provided by the present invention are as follows:

[0051] 1) The cost function provided by the present invention comprehensively considers the investment, operation, maintenance, and carbon trading costs during the entire life cycle of the industrial park, explores the impact of carbon indicators on carbon emission reduction in the construction time sequence, and improves economic benefits and environmental quality on the premise of ensuring the normal operation of the integrated energy system of the industrial park;

[0052] 2) Aiming at the local optimum problem existing in the cost-target function of the integrated energy system of the industrial park, the present invention adopts an algorithm based on the particle swarm optimization algorithm (PSO) combined with the extreme gradient boosting regression tree (XGBoost), and realizes the optimization of the objective function value through an efficient global parameter search mode, improving the accuracy of the carbon emission reduction analysis of the industrial park;

[0053] 3) Based on the Tapio decoupling model, the present invention uses the dynamic time warping (DTW) and K-means algorithms to cluster the carbon emission reduction effect results between carbon emission reduction and cost, and optimizes the carbon emission reduction path according to the characteristics of the carbon emission reduction effect clustering type, which can provide a basis for carbon emission reduction control in different stages.

[0054] Therefore, a multi-stage optimized carbon emission reduction method for industrial parks provided by the present invention can optimize multi-stage carbon emission reduction based on economic development and encourage energy conservation and emission reduction in industrial parks. Description of the Drawings

[0055] Figure 1 It is a flowchart of a multi-stage optimized carbon emission reduction method for industrial parks;

[0056] Figure 2 It is a multi-stage planning diagram of IPIES;

[0057] Figure 3 It is a flowchart of the XGBoost-PSO algorithm. Detailed Embodiments

[0058] To make the objectives, technical solutions, and advantages of the present invention clearer, the following further describes the embodiments of the present invention in detail.

[0059] Example 1

[0060] A multi-stage optimized carbon emission reduction method for industrial parks, see Figure 1 , the method includes the following steps:

[0061] 101: Initialization stage: Construct a multi-stage dynamic programming model;

[0062] (1) Define the multi-device coupling constraint;

[0063] (2) Divide the planning period into N stages, and optimize the capacity configuration of the devices in each stage.

[0064] Among them, the devices will be restricted by various factors during operation. The constraint conditions of multiple devices are to ensure the normal operation of the devices (the constraint condition formula is used as a prerequisite for setting the optimal device capacity). Further, on the basis of meeting the constraint conditions for the normal operation of the devices, the device capacity is optimally configured in the initialization stage. Similarly, under the optimal capacity configuration, subsequent calculations of data such as carbon indicators and carbon emissions are carried out.

[0065] 102: Carbon index quantification and ladder trading modeling;

[0066] (1) Calculate carbon indicators and actual carbon emissions such as IPIES, system purchased electricity, combined heat and power equipment, and gas boilers;

[0067] (2) Establish a ladder carbon trading cost model according to the carbon indicators and actual carbon emissions.

[0068] 103: Construct a total cost objective function, and use the XGBoost and PSO hybrid algorithm to optimize the total cost objective function;

[0069] (1) Integrate the total cost minimization objective function of investment, operation, maintenance, and ladder carbon trading costs;

[0070] (2) Optimize the total cost minimization objective function according to the XGBoost-PSO hybrid algorithm.

[0071] 104: Analyze the carbon emission reduction effect based on the ladder carbon trading cost model and the optimized total cost minimization objective function;

[0072] (1) Calculate the decoupling index based on the Tapio decoupling model (the decoupling index between the actual carbon emissions in step 102 and the optimal total cost in step 103);

[0073] (2) Integrate DTW-K-means clustering to achieve differentiated carbon emission reduction paths.

[0074] Among them, based on the calculated carbon emission reduction effect results at different stages, that is, the decoupling index. The DTW-K-means clustering algorithm is used to cluster the carbon emission reduction effects at different stages in the time dimension. Specifically, DTW is used to measure the similarity of the carbon emission reduction effect time series, and the K-means clustering algorithm is combined to achieve its clustering in the time dimension. Based on the DTW-K-means clustering algorithm, the carbon emission reduction effects at different stages are clustered from the time dimension.

[0075] In summary, through the above steps 101-104, the embodiments of the present invention comprehensively consider the construction time sequence, carbon indicators, and carbon emission reduction effects, not only optimizing the equipment configuration, but also motivating the park to reduce carbon emissions through the carbon trading mechanism, improving the economic benefits and environmental quality, and providing a basis for carbon emission reduction policies at different stages.

[0076] Embodiment 2

[0077] The following further introduces the solution in Embodiment 1 with specific formulas, as detailed in the following description:

[0078] 201: Based on the construction time sequence, construct a long-term multi-stage planning model for the integrated energy system of the industrial park;

[0079] 2011: Divide the planning period of the IPIES into N stages, and the multi-stage sequence S is expressed as follows:

[0080] S = [S 1 , S 2 ,..., S i ,..., S N

[0081] In the formula, S i refers to the i-th stage, i = 1, 2,..., N.

[0082] The corresponding equipment set sequence E set is as follows:

[0083] E set = [E set1 , E set2 ,..., E seti , E seti+1 ,..., E setN

[0084] In the formula, E set refers to the equipment set configured in the S i stage; E seti+1 is the equipment increment set based on E seti , E seti-1 ,..., E set1 . The supporting operation plan is as follows: ​​

[0085] P = [P 1 , P 2 ,..., P n ,..., P N

[0086] In the formula, represents the power value per hour of each type of equipment in the nth year, where n = 1, 2,..., n; represents the power value of the m-type equipment at the t-th moment in the nth year, where t = 1, 2,..., 8760.

[0087] 2012: Multi-stage planning equipment constraints

[0088] (1) Photovoltaic output constraint

[0089] The output power of the photovoltaic is determined by external irradiation intensity and operating temperature, etc. The output constraint of the photovoltaic system can be expressed as:

[0090]

[0091] In the formula, and E PV,nom respectively represent the actual photovoltaic output and the rated photovoltaic power generation capacity at the t-th time period of the d-th day in the nth year of the s-th planning period; η PV is the operating efficiency of the photovoltaic system; I s,n,d,t and I r respectively are the real-time solar irradiation intensity and the reference irradiation intensity at the t-th time period of the d-th day in the nth year of the s-th planning period; α is the temperature coefficient; T s,n,d,t and T r respectively represent the real-time operating temperature and the reference operating temperature of the photovoltaic cell at the t-th time period of the d-th day in the nth year of the s-th planning period; represents the ambient temperature at the t-th time period of the d-th day in the nth year of the s-th planning period; λ is the temperature radiation coefficient.

[0092] (2) Energy storage equipment constraints

[0093] The energy storage equipment includes electricity storage equipment and heat storage equipment. Taking the battery electricity storage equipment as an example, the energy stored in the energy storage equipment in a certain time period is expressed as follows:

[0094] And

[0095] In the formula, and respectively represent the energy of the battery energy storage equipment before and after charge and discharge at the t-th time period of the d-th day in the nth year of the s-th planning period; and ​respectively represent the charging and discharging powers of energy storage devices installed in the s-th planning period at the t-1 time period of the n-th year; ∈ is the self-discharge rate; η ESE,cha and η ESE,dis respectively represent the charging and discharging efficiencies.

[0096] In the long-term multi-stage planning, the energy stored in the energy storage device in the last 1 hour of each day is equal to the energy stored in the energy storage device in the first 1 hour of the next day, that is:

[0097]

[0098] In the formula, and respectively represent the energies of the energy storage device at the last 1 hour of the (d-1)-th day and the first 1 hour of the d-th day;

[0099] and respectively represent the energies of the energy storage device at the last 1 hour of the last day of the (n-1)-th year and the first 1 hour of the first day of the n-th year.

[0100] Energy storage device energy and charge-discharge limitations:

[0101]

[0102] In the formula, and respectively represent the upper and lower limits of the energy stored in the energy storage device in the s-th planning period; and respectively represent the maximum charging and discharging powers of the energy storage device in the s-th planning period; and respectively represent the charging and discharging states of the battery.

[0103] (3) Carbon capture equipment constraints

[0104] The energy consumption of the carbon capture unit is provided by the CHP, and the energy consumption of the carbon capture equipment is:

[0105]

[0106] In the formula, represents the total energy consumption of the carbon capture equipment at the t time period of the d-th day of the n-th year in the s-th planning period; E f is the fixed energy consumption of the carbon capture equipment; represents the operating energy consumption of the carbon capture equipment at the t time period of the d-th day of the n-th year in the s-th planning period.

[0107] (4) Other energy conversion equipment constraints

[0108] The operating constraints of each device within the system include the output range and energy conversion efficiency at each moment. The constraints for the power output device are as follows:

[0109]

[0110] In the formula: is the output of the power output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle; and are the upper and lower limits of the output of the power output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle, respectively; is the rated power of the power output device; is the input power of the power output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle; is the energy conversion efficiency of the power output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle.

[0111] The constraints for the heat output device are as follows:

[0112]

[0113] In the formula: is the output of the heat output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle; and are the upper and lower limits of the output of the heat output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle, respectively; is the rated power of the heat output device; is the input power of the heat output device e at the t-th time period on the d-th day of the n-th year in the s-th planning cycle.

[0114] (5) Equipment service life constraint

[0115] When the operating year n of the equipment e is greater than its total life N e , the equipment will no longer be used in the system:

[0116]

[0117] (6) Equipment power decay constraint

[0118] The annual decay of the equipment power:

[0119]

[0120] In the formula: is the maximum output of the equipment e in the n-th year of the s-th planning cycle during operation; is the maximum output of the equipment e in the first year of installation; is the annual decay rate of the equipment e output.

[0121] (7) Energy balance constraint

[0122] The electricity balance is expressed as follows:

[0123]

[0124] Where: and are the CHP output, grid power purchase, electricity consumption of the electric chiller, electricity consumption of the carbon capture equipment, and electricity load at the t-th time period on the d-th day of the n-th year in the s-th planning period, respectively; and are the electricity loads after price-based and incentive-based demand response at the t-th time period on the d-th day of the n-th year, respectively.

[0125] The heat balance is expressed as follows:

[0126]

[0127] Where: and are the waste heat recovered by the CHP and the heat generated by the boiler at the t-th time period on the d-th day of the n-th year in the s-th planning period, respectively; are the heat participating in demand response, input to the absorption chiller, and input to the heat exchanger at the t-th time period on the d-th day of the n-th year in the s-th planning period, respectively.

[0128] In long-term planning, multi-stage planning will increase the configuration scale of equipment such as photovoltaic, significantly reduce the configuration scale of energy storage equipment, and at the same time reduce the total cost and carbon emissions.

[0129] 202: Based on the optimal equipment capacity configuration, calculate the carbon indexes, actual carbon emissions, and E of IPIES, system-purchased electricity from outside the grid, combined heat and power equipment, and gas boilers P2G is the carbon emissions captured by the power-to-gas equipment, and according to the carbon emissions accounting results, establish a stepped carbon trading cost model;

[0130] Adopt the common free quota method and use the free carbon emission quota for determining IPIES. At the same time, consider that carbon emissions mainly come from system-purchased electricity from outside the grid, combined heat and power equipment, and gas boilers. The carbon emission quota, that is, the carbon index, is as follows:

[0131]

[0132] Where, E * , represent the carbon indexes of the industrial park integrated energy system, system-purchased electricity from outside the grid, combined heat and power equipment, and gas boilers, respectively; is the carbon index per unit of electricity; P GRID,t is the power purchase power of the system from the superior grid at the t-th time period; Carbon index per unit of heat; P CHP,t and H CHP,t are the electrical and heat output powers of the combined heat and power generation in period t, respectively; H GB,t is the heat output of the gas boiler in period t; is the conversion coefficient of the power generation of the combined heat and power generation equipment to the calorific value. In the embodiment of the present invention, the power generation of the combined heat and power generation equipment is converted into an equivalent heat supply, and the carbon index is determined according to the total equivalent calorific value.

[0133] The calculation formula for the actual carbon emissions of IPIES is as follows:

[0134] E = E GRID + E CHP + E GB - E P2G

[0135]

[0136] In the formula, E, E GRID , E CHP , E GB respectively represent the actual carbon emissions of the industrial park integrated energy system, purchased electricity, combined heat and power generation equipment, and gas boiler; E P2G is the amount of CO 2 captured by the power-to-gas equipment; β g is the carbon capture coefficient, which is the amount of CO 2 required for the power-to-gas equipment to convert a unit of electricity; P P2G,t is the electrical input power of the power-to-gas equipment in period t.

[0137] The stepped carbon trading cost model is as follows:

[0138] Based on the relationship between the actual carbon emissions and the carbon index, a stepped carbon trading model is established. The difference between the actual carbon emissions and the carbon emission quota of IPIES:

[0139] E IPIES = E - E *

[0140] The stepped carbon trading cost is:

[0141]

[0142] In the formula, μ is the basic carbon trading price; τ is the price growth rate; l is the length of the carbon emission interval. When E is less than E * , E IPIES is less than 0, indicating that the actual carbon emissions of the park are less than the carbon index, and the carbon index can be sold at the initial carbon trading price to obtain benefits.

[0143] 203: Construct the objective function of minimizing the total cost of IPIES for investment, operation, maintenance, and ladder carbon trading costs during the whole life cycle of the industrial park based on the optimal equipment capacity configuration. Use the XGBoost-PSO optimization algorithm to optimize the objective function of minimizing the total cost of IPIES through an efficient global parameter search mode;

[0144] 2031: The long-term multi-stage planning model of IPIES takes the minimum sum of investment, operation, maintenance, and ladder carbon trading costs during the whole life cycle as the objective function;

[0145]

[0146] In the formula, F is the present value of the total life cycle cost; n represents the nth year of the planning period; S is the number of planning stages; N is the planning cycle; n s represents that the sth planning stage is the nth s year of the planning cycle; δ is the discount rate; F EI,s is the equipment investment cost of the sth planning stage; F UO,n , F UM,n and F CO2,n are the operation, maintenance, and ladder carbon trading costs of IPIES in the nth year respectively.

[0147] (1) Investment cost: The investment cost is determined by photovoltaic, energy conversion, and energy storage equipment. The calculation formula for the investment cost F EI,s is:

[0148]

[0149] In the formula, F PV is the construction cost per unit capacity of photovoltaic equipment; E PV,s is the configured capacity of photovoltaic equipment in the sth planning stage; Π is the set of energy conversion equipment; F i is the construction cost per unit capacity of the ith type of energy conversion equipment; E i,s is the configured capacity of the ith type of energy conversion equipment in the sth planning stage; F j is the construction cost per unit capacity of the jth type of energy storage equipment; E j,s is the capacity configuration of the jth type of energy storage equipment in the sth planning stage.

[0150] (2) Operation cost: The operation cost includes the cost of purchasing electricity from the superior power grid, purchasing gas from the superior gas grid, and the penalty cost for abandoned wind of photovoltaic equipment. The calculation formula for the operation cost F UO,n is:

[0151]

[0152] In the formula, f e,t and f g,tare the electricity price and gas price in period t respectively; E GAS,n,t The gas power purchased by the park's integrated energy system from the superior gas grid during period t in year n; Penalty fee for unit wind abandonment; and P PV,n,t They are the maximum and actual photovoltaic outputs in period t of year n, respectively.

[0153] (3) Maintenance costs: Maintenance costs include the maintenance costs of photovoltaic, energy conversion, and energy storage equipment. Assume that the unit maintenance cost vector of each type of equipment is O = [O PV , O P2G , O EB , O CHP , O GB , O ES , O HS , O GS ], the vector composed of the output power of various types of equipment in the nth year and tth period is P n,t , then the maintenance cost of IPIES is F UM,n It is expressed as follows:

[0154]

[0155] 2032: XGBoost-PSO hybrid algorithm optimizes the total cost objective function;

[0156] The IPIES objective function is an operating cost function. It is a multivariable, nonlinear, and constrained combinatorial optimization problem. The PSO algorithm has a poor constraint effect on variables and is prone to fall into local optimality. According to the model requirements, the variables to be optimized in the particle matrix are determined, including photovoltaics, energy storage equipment, carbon capture equipment, other energy conversion equipment, equipment service life, equipment power attenuation, and energy balance constraints, a total of 7 variables to be optimized.

[0157] Assume that a dataset D = {(x i ,y i )|x i ∈R m ,y i ∈R,i=1,2,...,n} consists of m features and a total of n samples, where R m and R represent the m-dimensional real vector data set and real number set respectively.

[0158]

[0159] In the formula, f k is a regression tree, K is the total number of regression trees, and F is the regression tree space.

[0160] The objective function is expressed as:

[0161]

[0162] In the formula, l represents the loss function, which is used to measure the error between the classification prediction value and the true value; is the classification prediction value; i is the true value; Ω(f k ) is the regularization term.

[0163] The XGBoost algorithm uses gradient boosting iterative operations. After each iteration, a new regression tree will be added. The result of the tth iteration operation is:

[0164]

[0165] Combining the formulas, the objective function expression of the tth iteration is calculated as:

[0166]

[0167] Perform a second-order Taylor expansion and add the regularization term Ω(f k ) to prevent overfitting.

[0168]

[0169] In the formula, γ represents the leaf tree penalty coefficient; T is the number of leaf nodes; ω is the leaf weight; λ is the weight penalty coefficient. Based on the XGBoost principle and PSO algorithm theory, PSO is applied to the parameter optimization of XGBoost regression prediction. Taking the minimum sum of investment, operation, maintenance, and step-by-step carbon trading costs throughout the life cycle as the objective function, it can effectively control the system operation cost and reduce carbon emissions.

[0170] 204: Based on the optimization results of the IPIES total cost minimization objective function, the Tapio decoupling model is used, combined with dynamic time warping (DTW) and K-means algorithm to cluster the carbon reduction effect results between carbon emissions and IPIES minimized total cost, and the carbon reduction path is optimized according to the characteristics of the carbon reduction effect clustering type.

[0171] Based on the Tapio decoupling model, the dynamic time warping (DTW) and K-means algorithms are used to cluster the carbon emission reduction effect results between carbon emission reduction and minimization cost, and the carbon emission reduction path is optimized according to the clustering type characteristics of the carbon emission reduction effect.

[0172] Tapio decoupling is used to analyze the carbon reduction effect, as follows:

[0173]

[0174] Wherein, T represents the decoupling index, ΔC and ΔF respectively represent the changes in carbon emissions and minimized cost at the end stage relative to the base period, and C and F are respectively the carbon emissions and minimized cost at the end stage.

[0175] Based on the calculated carbon emission reduction effect results at different stages, that is, the decoupling index, the DTW-K-means clustering algorithm is used to cluster the carbon emission reduction effect in the time dimension. Specifically, DTW is used to measure the similarity of the carbon emission reduction effect time series, and the K-means clustering algorithm is combined to achieve its clustering in the time dimension. The Euclidean distance between the carbon emission reduction effect time series X = [x 1 , x 2 ,..., x m , Y = [y 1 , y 2 ,..., y m is as follows:

[0176]

[0177] Improve the distance algorithm of the K-means clustering model based on the DTW algorithm for application to the carbon emission reduction effect time series clustering.

[0178] (1) Use the elbow method to determine the optimal number of clusters. The sum of squared errors can intuitively display the clustering error:

[0179]

[0180] Wherein, H is the sum of squared errors, c i is the i-th class, d is the sample point of c i , and m i is the centroid of c i . When the number of clusters k is greater than k, the larger k is, the smoother the decline of the sum of squared errors, and the position corresponding to the elbow is the optimal number of clusters.

[0181] (2) For X = [x 1 , x 2 ,..., x m , cluster it into k clusters, and C = [c 1 , c 2 ,..., c k is the set of the centers of each cluster. The objective function of K-means is as follows:

[0182]

[0183] Wherein, μ i is the center of c i .

[0184] Based on the DTW-K-means clustering algorithm, the carbon emission reduction effects at different stages are clustered from the time dimension. The number of clustering centers K is 3, and the carbon emission reduction effects can be summarized as potential type, accumulation type, and pressure type. According to the characteristics of the carbon emission reduction effect clustering types, the carbon emission reduction trend is analyzed, and the carbon emission reduction path is optimized.

[0185] If the carbon emission reduction effect is classified as the potential type, it means that the overall trend is relatively stable, and carbon emissions and costs are maintained in a state of strong decoupling or weak decoupling. However, in order to achieve the goal of stable strong decoupling between future carbon emissions and economic growth, the carbon emission reduction work still needs to be continuously promoted. If the carbon emission reduction effect is of the accumulation type, it indicates that although the overall situation is not stable, it is developing in a positive direction. This reflects that at this stage, while economic development is guaranteed, the carbon emission reduction effect is also gradually improving, and is moving towards the goal of strong decoupling between carbon emissions and economic growth. If the carbon emission reduction effect in a certain stage is classified as the pressure type, it means that the overall performance is poor, and the carbon emission reduction work faces great challenges. This shows that there is still a large room for improvement in the carbon emission reduction effect in the process of pursuing economic growth, and more powerful measures need to be taken to optimize the carbon emission reduction work while ensuring the stable development of the economy.

[0186] Embodiment 3

[0187] A multi-stage optimized carbon emission reduction device for industrial parks, the device includes:

[0188] A building module, used to establish a stepped carbon trading cost model according to carbon indicators and actual carbon emissions;

[0189] An optimization module, used to integrate the total cost minimization objective function of investment, operation, maintenance, and stepped carbon trading costs; optimize the total cost minimization objective function according to the XGBoost-PSO hybrid algorithm;

[0190] An acquisition module, used to analyze the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function, and obtain the carbon emission reduction control strategy for the industrial park.

[0191] Among them, the acquisition module is:

[0192] A calculation sub-module, used to calculate the carbon emission reduction effect results at different stages based on the stepped carbon trading cost model, the optimized total cost minimization objective function, and the Tapio decoupling model;

[0193] A processing sub-module, used to fuse DTW-K-means clustering, process the carbon emission reduction effect results, realize a differentiated carbon emission reduction path, and finally obtain the carbon emission reduction control strategy for the industrial park.

[0194] In summary, the embodiments of the present invention comprehensively consider the construction sequence and carbon indicators, and use an optimization algorithm to solve the local optimum problem, improve the carbon emission analysis and prediction capabilities of high-carbon emission industries and regions, and enhance the accuracy of decarbonization trend analysis technology; on the premise of ensuring economic development, optimize the industrial structure and energy structure, encourage energy conservation and emission reduction in industrial parks, and promote the high-quality development of the parks.

[0195] Embodiment 4

[0196] A carbon emission reduction device for industrial parks with multi-stage optimization, the device includes: a processor and a memory, and program instructions are stored in the memory. The processor calls the program instructions stored in the memory to enable the device to execute the following method steps in Embodiment 1:

[0197] Establish a stepped carbon trading cost model according to carbon indicators and actual carbon emissions;

[0198] Integrate the total cost minimization objective function of investment, operation, maintenance and stepped carbon trading costs; optimize the total cost minimization objective function according to the XGBoost-PSO hybrid algorithm;

[0199] Analyze the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function, and obtain the carbon emission reduction control strategy for the industrial park.

[0200] Among them, the analysis of the carbon emission reduction effect based on the stepped carbon trading cost model and the optimized total cost minimization objective function is as follows:

[0201] Calculate the carbon emission reduction effect results of different stages based on the stepped carbon trading cost model, the optimized total cost minimization objective function, and the Tapio decoupling model;

[0202] Fuse DTW-K-means clustering to process the carbon emission reduction effect results and realize different carbon emission reduction paths.

[0203] Among them, establishing a stepped carbon trading cost model according to carbon indicators and actual carbon emissions is specifically:

[0204] Carbon indicators:

[0205]

[0206] In the formula, E * , respectively represent the carbon indicators of the integrated energy system of the industrial park, the system's externally purchased electricity, the cogeneration equipment, and the gas boiler; is the carbon indicator per unit of electricity; P GRID,t is the power purchase power of the system from the superior power grid at time t; is the carbon indicator per unit of heat; P CHP,t and HCHP,t are the electrical and heat output powers of the combined heat and power generation in period t; H GB,t is the heat output of the gas boiler in period t; is the conversion coefficient of the power generation of the combined heat and power generation equipment to the calorific value;

[0207] Actual carbon emissions:

[0208] E = E GRID + E CHP + E GB - E P2G

[0209]

[0210] In the formula, E, E GRID , E CHP , E GB respectively represent the actual carbon emissions of the integrated energy system of the industrial park, purchased electricity, combined heat and power generation equipment, and gas boiler; E P2G is the amount of CO 2 captured by the power-to-gas equipment; β g is the carbon capture coefficient, which is the amount of CO 2 required for the power-to-gas equipment to convert a unit of electricity; P P2G,t is the electrical input power of the power-to-gas equipment in period t;

[0211] Step carbon trading cost model:

[0212] E IPIES = E - E *

[0213] Step carbon trading cost is:

[0214]

[0215] In the formula, μ is the basic carbon trading price; τ is the price growth rate; l is the length of the carbon emission interval. When E is less than E * , E IPIES is less than 0, indicating that the actual carbon emissions of the park are less than the carbon quota.

[0216] Among them, the total cost minimization objective function integrating investment, operation, maintenance and step carbon trading cost is:

[0217] Investment: Determined by photovoltaic, energy conversion, and energy storage equipment. The investment cost F EI,s is:

[0218]

[0219] In the formula, F PV is the unit capacity construction cost of the photovoltaic equipment; EPV,s is the configured capacity of the photovoltaic equipment in the s-th planning stage; Π is the set of energy conversion equipment; F i is the construction cost per unit capacity of the i-th type of energy conversion equipment; E i,s is the configured capacity of the i-th type of energy conversion equipment in the s-th planning stage; F j is the construction cost per unit capacity of the j-th type of energy storage equipment; E j,s is the capacity configuration of the j-th type of energy storage equipment in the s-th planning stage;

[0220] Operation:

[0221]

[0222] In the formula, f e,t and f g,t are the electricity price and gas price at time t respectively; E GAS,n,t is the gas purchase power of the integrated energy system in the park from the superior gas network at time t in the n-th year; is the penalty cost per unit of abandoned wind volume; and P PV,n,t are the maximum and actual power outputs of the photovoltaic at time t in the n-th year respectively;

[0223] Maintenance:

[0224]

[0225] Total cost objective function;

[0226]

[0227] In the formula, γ represents the leaf tree penalty coefficient; T is the number of tree leaf nodes; ω is the leaf weight; λ is the weight penalty coefficient.

[0228] Among them, the Tapio decoupling model:

[0229]

[0230] In the formula, T represents the decoupling index, △C and △F respectively represent the changes in carbon emissions and minimized cost at the end period relative to the base period, and C and F are the carbon emissions and minimized cost at the end period respectively.

[0231] It should be noted here that the device description in the above embodiments corresponds to the method description in the embodiments, and the embodiments of the present invention will not be elaborated here.

[0232] The execution subjects of the above-mentioned processor and memory can be devices with computing functions such as a computer, a single-chip microcomputer, and a microcontroller. Specifically, in implementation, the embodiments of the present invention do not limit the execution subject and are selected according to the needs in actual applications.

[0233] Data signals are transmitted between the memory and the processor via a bus, which is not elaborated in the embodiments of the present invention.

[0234] In summary, the embodiments of the present invention comprehensively consider the construction timing and carbon indicators, and use an optimization algorithm to solve the local optimum problem, improve the carbon emission analysis and prediction capabilities of high-carbon emission industries and regions, and enhance the accuracy of decarbonization trend analysis technology; on the premise of ensuring economic development, optimize the industrial structure and energy structure, encourage energy conservation and emission reduction in industrial parks, and promote the high-quality development of the parks.

[0235] Based on the same inventive concept, the embodiments of the present invention also provide a computer-readable storage medium, which includes a stored program that controls the device where the storage medium is located to execute the method steps in the above embodiments when the program runs.

[0236] The computer-readable storage medium includes, but is not limited to, flash memory, hard disk, solid-state drive, etc.

[0237] It should be noted here that the description of the readable storage medium in the above embodiments corresponds to the description of the method in the embodiments, and the embodiments of the present invention will not elaborate on this here.

[0238] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part.

[0239] The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted through a computer-readable storage medium. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium can be a magnetic medium or a semiconductor medium, etc.

[0240] In the embodiments of the present invention, except for those with special descriptions, the models of each device are not limited, and any device that can perform the above functions can be used.

[0241] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred embodiment, and the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0242] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A multi-stage optimized industrial park carbon emission reduction method, characterized in that: The method comprises: Establish a tiered carbon trading cost model based on carbon indicators and actual carbon emissions; The total cost minimization objective function integrating investment, operation, maintenance and tiered carbon trading costs; the total cost minimization objective function is optimized using the XGBoost-PSO hybrid algorithm; Based on the step-by-step carbon trading cost model and the optimized total cost minimization objective function, the carbon emission reduction effect is analyzed to obtain the carbon emission reduction control strategy for the industrial park.

2. A multi-stage optimized industrial park carbon emission reduction method according to claim 1, characterized in that: The carbon emission reduction effect is analyzed based on the step-by-step carbon trading cost model and the optimized total cost minimization objective function as follows: Based on the step-by-step carbon trading cost model and the optimized total cost minimization objective function, the carbon emission reduction effect results at different stages are calculated based on the Tapio decoupling model; By integrating DTW-K-means clustering, the carbon emission reduction effect results are processed to obtain differentiated carbon emission reduction paths, and finally the carbon emission reduction control strategy for the industrial park is obtained.

3. The multi-stage optimized industrial park carbon emission reduction method according to claim 1 is characterized in that: The specific method of establishing a tiered carbon trading cost model based on carbon indicators and actual carbon emissions is as follows: Carbon indicators: In the formula, E * , They represent the carbon index of the industrial park’s integrated energy system, the system’s purchased electricity, cogeneration equipment, and gas boilers; is the carbon index per unit of electricity; P GRID,t is the power purchased by the system from the upper grid during period t; is the carbon index per unit heat; P CHP,t and H CHP,t are the electrical and thermal output power of the cogeneration during period t; H GB,t is the heat output of the gas boiler during period t; The conversion factor of the power generation of the cogeneration equipment to the heat generation; Actual carbon emissions: E=E GRID +E CHP +E GB -AND P2G In the formula, E, E GRID 、E CHP 、E GB They represent the actual carbon emissions of the industrial park’s comprehensive energy system, purchased electricity, cogeneration equipment, and gas boilers; E P2G The amount of CO2 captured by the power-to-gas plant; β g is the carbon capture coefficient, which is the amount of CO2 required for the power-to-gas device to convert unit electricity; P P2G,t is the electrical input power of the power-to-gas device during period t; Ladder carbon trading cost model: AND IPIES =EE * Tiered carbon trading costs for: Where μ is the carbon trading base price; τ is the price growth rate; l is the length of the carbon emission interval. * , E IPIES Less than 0 means the actual carbon emissions of the park are less than the carbon index.

4. The multi-stage optimized industrial park carbon emission reduction method according to claim 1 is characterized in that: The total cost minimization objective function integrating investment, operation, maintenance and tiered carbon trading costs is: Investment: Determined by photovoltaic, energy conversion, and energy storage equipment. Investment cost F EI,s for: In the formula, F PV E is the unit capacity investment and construction cost of photovoltaic equipment; PV,s is the configuration capacity of the photovoltaic equipment in the sth planning stage; Π is the set of energy conversion equipment; F i E is the unit capacity investment and construction cost of the i-th type of energy conversion equipment; i,s is the configuration capacity of the i-th type of energy conversion equipment in the s-th planning stage; F j E is the unit capacity investment and construction cost of the j-th type of energy storage equipment; j,s is the capacity configuration of the j-th type of energy storage equipment in the s-th planning stage; run: In the formula, f e,t and f g,t are the electricity price and gas price in period t respectively; E GAS,n,t The gas power purchased by the park's integrated energy system from the superior gas grid during period t in year n; Penalty fee for unit wind abandonment; and P PV,n,t are the maximum and actual output of photovoltaic power in period t of year n, respectively; maintain: Total cost objective function; In the formula, γ represents the leaf tree penalty coefficient; T is the number of leaf nodes; ω is the leaf weight; λ is the weight penalty coefficient.

5. The multi-stage optimized industrial park carbon emission reduction method according to claim 2 is characterized in that: The Tapio decoupling model: Where T represents the decoupling index, △C and △F represent the changes in carbon emissions and minimization costs at the end of the period relative to the base period, respectively, and C and F represent the carbon emissions and minimization costs at the end of the period, respectively.

6. A multi-stage optimized industrial park carbon emission reduction device, characterized in that: The device comprises: Establish a module to build a tiered carbon trading cost model based on carbon indicators and actual carbon emissions; The optimization module is used to integrate the total cost minimization objective function of investment, operation, maintenance and tiered carbon trading costs; the total cost minimization objective function is optimized according to the XGBoost-PSO hybrid algorithm; The acquisition module is used to analyze the carbon emission reduction effect based on the step-by-step carbon trading cost model and the optimized total cost minimization objective function, and to obtain the carbon emission reduction control strategy for the industrial park.

7. The multi-stage optimized industrial park carbon emission reduction device according to claim 6 is characterized in that: The acquisition module is: The calculation submodule is used to calculate the carbon emission reduction effect results at different stages based on the step-by-step carbon trading cost model, the optimized total cost minimization objective function, and the Tapio decoupling model; The processing submodule is used to integrate DTW-K-means clustering, process the carbon emission reduction effect results, realize differentiated carbon emission reduction paths, and finally obtain the carbon emission reduction control strategy of the industrial park.

8. A multi-stage optimized industrial park carbon emission reduction device, characterized in that: The device comprises: a processor and a memory, wherein program instructions are stored in the memory, and the processor calls the program instructions stored in the memory to enable the device to execute any one of the methods of claims 1-5.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 5.

Images