Energy data processing method, device, computer equipment and storage medium

Abstract

The present application relates to an energy data processing method, apparatus, computer equipment and storage medium. The method comprises: obtaining and obtaining the current population to be evolved based on the target mathematical model, target operation strategy and target constraint conditions; performing incremental processing on the current population to be evolved to generate a target population to be dominated; obtaining and determining the target population to be evolved based on multiple non-dominated solution sets and target congestion distances of different levels corresponding to each target model combination in the target population to be dominated, and using the target population to be evolved as the current population to be evolved; returning to the operation of performing incremental processing on the current population to be evolved until the population evolution end condition is met to obtain the target population; calculating the target relative approximation corresponding to each target model combination in the target population; and determining the final model combination based on the comparison results of the relative approximations of each target. The use of this method can improve the efficiency of the comprehensive energy multi-objective optimization operation.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an energy data processing method, apparatus, computer equipment, and storage medium. Background Art

[0002] With the development of computer technology, many methods have emerged to optimize the capacity configuration of various equipment in integrated energy systems. For example, in previous studies, some experts have developed multi-objective optimization methods that consider energy, environment, and economic performance by using genetic algorithms in MATLAB software. Some experts have also built multi-objective operation optimization models based on genetic algorithms with economic and environmental protection as dual goals.

[0003] However, most of the integrated energy system capacity configuration methods in traditional technologies do not take into account the complementary coupling relationship between multiple heterogeneous energy sources and the carbon emission limits of related energy equipment, and pay less attention to the operating logic between different energy sources and variable factors such as annual load and meteorological data, resulting in low efficiency in achieving economic and environmentally friendly centralized integrated energy operation optimization. Summary of the Invention

[0004] Based on this, it is necessary to provide an energy data processing method, device, computer equipment and storage medium that can achieve economic and environmentally friendly centralized integrated energy operation optimization to address the above technical problems, thereby improving the efficiency of achieving economic and environmentally friendly centralized integrated energy operation optimization.

[0005] A method for processing energy data, the method comprising:

[0006] Acquire a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set;

[0007] Obtaining a target operation strategy, and combining different types of models in the target mathematical model based on the target operation strategy to obtain a plurality of intermediate model combinations;

[0008] Obtaining a target constraint condition, screening an intermediate model combination that meets the target constraint condition from the multiple intermediate model combinations as a target model combination, and using the set of the target model combinations as the current population to be evolved;

[0009] Performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generating a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated;

[0010] Obtaining dominance relationships corresponding to each target model combination in the target population to be dominated, obtaining a plurality of non-dominated solution sets of different levels based on the respective dominance relationships, obtaining target crowding distances corresponding to each target model combination based on the plurality of non-dominated solution sets of different levels, determining a target population to be evolved based on the plurality of non-dominated solution sets of different levels and the respective target crowding distances, and using the target population to be evolved as the current population to be evolved;

[0011] Returning to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain the target population;

[0012] Normalizing the target population to obtain a target normalized matrix, and obtaining target relative approximations corresponding to each target model combination in the target population based on the target normalized matrix;

[0013] Based on the comparison results of the relative approximations of the various targets, a final model combination is screened from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0014] In one embodiment, obtaining a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set, includes:

[0015] Acquiring wind power generation data, and fitting a wind turbine generator model based on the wind power generation data;

[0016] Acquiring photovoltaic data, and fitting a photovoltaic model based on the photovoltaic data;

[0017] Acquiring cooling, heating, and electricity trigeneration data, and fitting a cooling, heating, and electricity trigeneration system model based on the cooling, heating, and electricity trigeneration data;

[0018] Acquire gas boiler data, and obtain a gas boiler model based on the gas boiler data;

[0019] Acquiring absorption refrigeration data, and fitting an absorption refrigeration machine model based on the absorption refrigeration data;

[0020] Acquiring organic Rankine cycle data, and fitting an organic Rankine cycle model based on the organic Rankine cycle data;

[0021] Acquiring energy storage battery data, and fitting an energy storage battery model based on the energy storage battery data;

[0022] Acquiring thermal storage tank data, and fitting a thermal storage tank model based on the thermal storage tank data;

[0023] The energy input model set includes the wind turbine model, photovoltaic model, trigeneration system model and gas boiler model; the energy conversion equipment model set includes the absorption refrigerator model and organic Rankine cycle model; and the energy storage model set includes the energy storage battery model and thermal storage tank model.

[0024] In one embodiment, performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated includes:

[0025] Obtaining the fitness value corresponding to each target model combination in the current population to be evolved;

[0026] Merging the fitness values ​​corresponding to the current population to be evolved to obtain a first total, obtaining the selection probabilities corresponding to the target model combinations based on the fitness values ​​and the first total, and screening out a first set to be dominated from the current population to be evolved, where the first set to be dominated is a set of target model combinations in the current population to be evolved whose selection probabilities satisfy a first condition;

[0027] Performing descending sorting based on the fitness values ​​corresponding to the current population to be evolved to obtain a descending sorting result, and based on the descending sorting result, screening a first target number of target model combinations from the current population to be evolved, and using the set of the screened target model combinations as a second set to be dominated;

[0028] Randomly selecting a second target number of target model combinations from the current population to be evolved as random target model combinations, and screening a third dominating set from the current evolving population based on a comparison result of fitness values ​​corresponding to the random target model combinations;

[0029] Obtaining a uniform distribution factor, repeatedly selecting two target model combinations from the current population to be evolved as target model combinations to be crossed, obtaining corresponding cross target model combinations based on the uniform distribution factor and the target model combinations to be crossed, until a third target number of cross target model combinations is obtained, and using the set of each cross target model combination as a fourth dominating set;

[0030] Obtain a mutation factor, repeatedly select a target model combination from the current population to be evolved as the target model combination to be mutated, obtain an upper bound target model combination and a lower bound target model combination from the current population to be evolved, and calculate corresponding mutated target model combinations based on the mutation factor, the target model combination to be mutated, the upper bound target model combination, and the lower bound target model combination, until a fourth target number of mutated target model combinations are obtained, and use the set of the mutated target model combinations as a fifth dominating set;

[0031] The intermediate population to be dominated is obtained based on the combination of the first dominating set, the second dominating set, the third dominating set, the fourth dominating set and the fifth dominating set.

[0032] In one embodiment, obtaining dominance relationships corresponding to respective target model combinations in the target to-be-dominated population, and obtaining a plurality of non-dominated solution sets at different levels based on the respective dominance relationships includes:

[0033] Obtaining an objective function, calculating, based on the objective function, function values ​​corresponding to each target model combination in the target population to be dominated, and obtaining, based on comparison results of the function values, dominance relationships corresponding to each target model combination in the target population to be dominated;

[0034] Based on the dominance relationship corresponding to each target model combination in the target to-be-dominated population, a set of non-dominated target model combinations is screened out from the target to-be-dominated population as a first-level non-dominated solution set, the first-level non-dominated solution set is used as a current-level non-dominated solution set, and the target to-be-dominated population is used as a current dominated population;

[0035] Deleting the target model combination consistent with the non-dominated solution set of the current level in the current dominated population to obtain a target dominated population;

[0036] Based on the dominance relationships corresponding to the target model combinations in the target dominating population, a set of non-dominated target model combinations is screened out from the target dominating population as a non-dominated solution set of the current level, and the target dominating population is used as the current dominating population;

[0037] Return to the operation of deleting the target model combination in the current dominated population that is consistent with the non-dominated solution set of the current level to obtain the target dominated population, until each target model combination in the target to-be-dominated population has a non-dominated solution set of the corresponding level, and obtain the non-dominated solution sets of the multiple different levels.

[0038] In one embodiment, obtaining the target crowding distance corresponding to each target model combination based on the multiple non-dominated solution sets at different levels includes:

[0039] Obtaining initial crowding distances corresponding to each target model combination in the non-dominated solution sets of the multiple different levels, and using the initial crowding distances as current crowding distances;

[0040] Selecting one non-dominated solution set from the non-dominated solution sets at different levels in turn as the current non-dominated solution set;

[0041] Selecting a function from the objective function as a current function, and obtaining current function values ​​corresponding to each objective model combination in the current non-dominated solution set based on the current function;

[0042] Sorting the current function values ​​corresponding to the target model combinations in the current non-dominated solution set to obtain a current sorting result, and obtaining a maximum current function value and a minimum current function value corresponding to the current non-dominated solution set based on the current sorting result;

[0043] Based on the current crowding distance, the maximum current function value, the minimum current function value corresponding to the current dominated solution set, and the current function values ​​corresponding to the current non-dominated solution set, calculate the intermediate crowding distance corresponding to each target model combination in the current non-dominated solution set, and use the intermediate crowding distance as the current crowding distance;

[0044] Repeating the operation of selecting a function from the objective function as the current function until all functions in the objective function are selected, and obtaining the target crowding distance corresponding to each target model combination in the current dominating solution set;

[0045] The operation of sequentially selecting one non-dominated solution set from the multiple non-dominated solution sets at different levels as the current non-dominated solution set is repeated until each non-dominated solution set in the multiple non-dominated solution sets at different levels is selected, thereby obtaining the target crowding distance corresponding to each target model combination.

[0046] In one embodiment, determining the target population to be evolved based on the multiple non-dominated solution sets at different levels and the target crowding distances includes:

[0047] Based on the hierarchical relationship corresponding to the non-dominated solution sets at different levels, sequentially selecting non-dominated solution sets that meet a selection condition from the non-dominated solution sets at different levels as intermediate non-dominated solution sets;

[0048] The total number of target model combinations in the intermediate non-dominated solution set is used as the first target total number;

[0049] Obtaining a target population size, and when the first target total number is equal to the target population size, using the intermediate non-dominated solution set as the target population to be evolved;

[0050] When the first target total number is less than the target population number, obtaining a next-level non-dominated solution set corresponding to the intermediate non-dominated solution set;

[0051] The target crowding distances corresponding to the target model combinations in the non-dominated solution set of the next level are sorted to obtain a target sorting result. Based on the target sorting result, the corresponding target model combinations are sequentially added to the intermediate non-dominated solution set until the number of target model combinations in the intermediate non-dominated solution set is equal to the number of the target population, thereby obtaining the target population to be evolved.

[0052] In one embodiment, obtaining the target relative similarity corresponding to each target model combination in the target population based on the target normalization matrix includes:

[0053] determining a positive ideal solution and a negative ideal solution based on the target normalized matrix;

[0054] Calculating first distances corresponding to each target model combination in the target population based on the target normalized matrix and the positive ideal solution;

[0055] Calculating a second distance corresponding to each target model combination in the target population based on the target normalized matrix and the negative ideal solution;

[0056] Fusing the first distances and the second distances corresponding to each target model combination in the target population to obtain a total target distance corresponding to each target model combination;

[0057] Based on the ratio of the first distance corresponding to each target model combination to the corresponding target total distance, the target relative proximity corresponding to each target model combination is obtained.

[0058] An energy data processing device, comprising:

[0059] A model acquisition module is used to acquire a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set;

[0060] A model combination module is used to obtain a target operation strategy, and to combine different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations;

[0061] an initialization module, configured to obtain target constraints, select an intermediate model combination that meets the target constraints from the plurality of intermediate model combinations as a target model combination, and use the set of target model combinations as a current population to be evolved;

[0062] an increment module, configured to perform increment processing on the current population to be evolved to obtain an intermediate population to be dominated, and generate a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated;

[0063] a population evolution module, configured to obtain dominance relationships corresponding to respective target model combinations in the target population to be dominated, obtain a plurality of non-dominated solution sets at different levels based on the respective dominance relationships, obtain target crowding distances corresponding to the respective target model combinations based on the plurality of non-dominated solution sets at different levels, determine a target population to be evolved based on the plurality of non-dominated solution sets at different levels and the respective target crowding distances, and use the target population to be evolved as the current population to be evolved;

[0064] A population determination module is used to return to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain a target population;

[0065] An evaluation module is used to perform normalization processing on the target population to obtain a target normalization matrix, and obtain target relative approximations corresponding to each target model combination in the target population based on the target normalization matrix;

[0066] The result determination module is used to screen the target model combinations based on the comparison results of the relative approximations of the targets to obtain a final model combination, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0067] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0068] Acquire a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set;

[0069] Obtaining a target operation strategy, and combining different types of models in the target mathematical model based on the target operation strategy to obtain a plurality of intermediate model combinations;

[0070] Obtaining a target constraint condition, screening an intermediate model combination that meets the target constraint condition from the multiple intermediate model combinations as a target model combination, and using the set of the target model combinations as the current population to be evolved;

[0071] Performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generating a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated;

[0072] Obtaining dominance relationships corresponding to each target model combination in the target population to be dominated, obtaining a plurality of non-dominated solution sets of different levels based on the respective dominance relationships, obtaining target crowding distances corresponding to each target model combination based on the plurality of non-dominated solution sets of different levels, determining a target population to be evolved based on the plurality of non-dominated solution sets of different levels and the respective target crowding distances, and using the target population to be evolved as the current population to be evolved;

[0073] Returning to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain the target population;

[0074] Normalizing the target population to obtain a target normalized matrix, and obtaining target relative approximations corresponding to each target model combination in the target population based on the target normalized matrix;

[0075] Based on the comparison results of the relative approximations of the various targets, a final model combination is screened from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0076] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0077] Acquire a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set;

[0078] Obtaining a target operation strategy, and combining different types of models in the target mathematical model based on the target operation strategy to obtain a plurality of intermediate model combinations;

[0079] Obtaining a target constraint condition, screening an intermediate model combination that meets the target constraint condition from the multiple intermediate model combinations as a target model combination, and using the set of the target model combinations as the current population to be evolved;

[0080] Performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generating a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated;

[0081] Obtaining dominance relationships corresponding to each target model combination in the target population to be dominated, obtaining a plurality of non-dominated solution sets of different levels based on the respective dominance relationships, obtaining target crowding distances corresponding to each target model combination based on the plurality of non-dominated solution sets of different levels, determining a target population to be evolved based on the plurality of non-dominated solution sets of different levels and the respective target crowding distances, and using the target population to be evolved as the current population to be evolved;

[0082] Returning to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain the target population;

[0083] Normalizing the target population to obtain a target normalized matrix, and obtaining target relative approximations corresponding to each target model combination in the target population based on the target normalized matrix;

[0084] Based on the comparison results of the relative approximations of the various targets, a final model combination is screened from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0085] The above-mentioned energy data processing method, device, computer equipment and storage medium obtain a target mathematical model, combine different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations; obtain target constraints, screen the intermediate model combinations that meet the target constraints from the multiple intermediate model combinations as the target model combination, and use the set of the target model combinations as the current population to be evolved; perform incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generate a target population to be dominated based on the current population to be evolved and the combination of the intermediate population to be dominated; obtain the dominance relationship corresponding to each target model combination in the target population to be dominated. , based on the respective dominating relationships, a plurality of non-dominated solution sets of different levels are obtained; based on the plurality of non-dominated solution sets of different levels, the target crowding distances corresponding to the respective target model combinations are obtained; based on the plurality of non-dominated solution sets of different levels and the respective target crowding distances, a target population to be evolved is determined, and the target population to be evolved is used as the current population to be evolved; returning to perform incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until a population evolution end condition is met, thereby obtaining a target population; performing normalization processing on the target population to obtain a target normalized matrix; based on the target normalized matrix, the target relative similarity corresponding to each target model combination in the target population is obtained;Based on the comparison results of the relative approximations of the various targets, a final model combination is obtained from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model. By obtaining the target mathematical model and the target operation strategy, various models in the target mathematical model are combined based on the target operation strategy to obtain multiple intermediate model combinations, and target constraints are obtained. Based on the target constraints, target model combinations that meet the conditions are screened out from the multiple intermediate model combinations. The set of each target model combination is used as the current population to be evolved, and the current population to be evolved is incrementally processed and the current evolved population is merged to obtain the target population to be dominated. Based on the dominance relationship corresponding to each target model combination in the target population to be dominated, multiple non-dominated solution sets of different levels are obtained, and then based on multiple different The target crowding distances corresponding to each target model combination are calculated from the non-dominated solution sets at different levels. The target population to be evolved is determined based on the non-dominated solution sets at multiple levels and the target crowding distances. This target population to be evolved is used as the current population to be evolved, and incremental processing is continued on the current population to be evolved until the population evolution end condition is met. This results in the target population being normalized to obtain a target normalized matrix. Based on this normalized matrix, the target relative approximations corresponding to each target model combination in the target population are calculated. The final model combination is determined based on the comparison of the target relative approximations. This approach considers the complementary coupling between the energy devices corresponding to the target mathematical models, effectively planning the optimal capacity of the energy devices corresponding to the target mathematical models, thereby improving the efficiency of achieving economical and environmentally friendly centralized integrated energy operation optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 A diagram of an application environment of an energy data processing method in one embodiment;

[0087] Figure 2 1 is a flow chart of an energy data processing method according to an embodiment;

[0088] Figure 3 A schematic diagram of a process for constructing a target mathematical model in one embodiment;

[0089] Figure 4 A schematic diagram of a population increment process in one embodiment;

[0090] Figure 5 A schematic diagram of a process for generating a non-dominated solution set in one embodiment;

[0091] Figure 6 1 is a flow chart of target congestion calculation in one embodiment;

[0092] Figure 7 A schematic diagram of a process for generating a target population to be evolved in one embodiment;

[0093] Figure 8 Schematic diagram of a process for calculating target relative similarity in one embodiment;

[0094] Figure 9 A structural diagram of an integrated energy system in one embodiment;

[0095] Figure 10 A schematic diagram of a short-term power load combination forecasting model in one embodiment;

[0096] Figure 11 A diagram of an integrated energy system operation strategy in one embodiment;

[0097] Figure 12 A schematic diagram of an algorithm flow in one embodiment;

[0098] Figure 13 is a structural block diagram of an energy data processing device in one embodiment;

[0099] Figure 14 is a diagram of the internal structure of a computer device in one embodiment;

[0100] Figure 15 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0101] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0102] The energy data processing method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store data that server 104 needs to process. The data storage system can be integrated with server 104 or placed on a cloud or other network server. Terminal 102 is used to input energy data. The server 104 is configured to obtain a target mathematical model and a target operation strategy, combine different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations, obtain target constraints, screen the multiple intermediate model combinations based on the target constraints to obtain a current population to be evolved, perform incremental processing on the current population to be evolved and merge the populations to obtain a target population to be dominated, obtain dominance relationships corresponding to each target model combination in the target population to be dominated, obtain multiple non-dominated solution sets of different levels based on the dominance relationships, and then calculate target crowding distances corresponding to each target model combination, determine a target population to be evolved, use the target population to be evolved as the current population to be evolved, continue to perform incremental processing on the current population to be evolved until a population evolution end condition is met, and obtain a target population, perform normalization processing on the target population to obtain a target normalized matrix, obtain target relative approximations corresponding to each target model combination in the target population based on the target normalized matrix, and determine a final model combination based on a comparison result of the target relative approximations. The final model combination is used to plan the optimal capacity of energy equipment corresponding to the target mathematical model. Terminal 102 may include, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices may include smart speakers, smart TVs, smart air conditioners, and smart car devices. Portable wearable devices may include smart watches, smart bracelets, and head-mounted devices. Server 104 may be implemented as a standalone server or a server cluster consisting of multiple servers.

[0103] In one embodiment, Figure 2 As shown, a method for processing energy data is provided, which is applied to Figure 1 The following steps are used as an example to illustrate the server in the example:

[0104] Step S200: obtaining a target mathematical model, wherein the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set.

[0105] The target mathematical model refers to the mathematical model corresponding to each energy device, which is obtained by fitting the various energy data corresponding to each energy device. It is not limited to the mathematical models corresponding to the energy input model set, the energy conversion device model set, and the energy storage model set. It can also include mathematical models fitted according to the energy data corresponding to devices such as "gas turbines", "refrigerators", and "electric boilers". The energy input model set includes but is not limited to wind power generation models, photovoltaic models, trigeneration system models, and gas boiler models, which are used to represent the types of mathematical models related to energy input. The energy conversion device model set includes but is not limited to absorption chiller models and organic Rankine cycle models (also called ORC cycle models, ORC stands for Organic Rankine Cycle), which are used to represent the types of mathematical models related to energy conversion. The energy storage model set includes but is not limited to energy storage battery models and thermal storage tank models, which are used to represent the types of mathematical models related to energy storage.

[0106] Specifically, to analyze the complementary coupling characteristics between the various energy sources corresponding to each type of energy device, corresponding mathematical models are fitted based on the energy data corresponding to each type of energy device. The set of these mathematical models is referred to as the target mathematical model. Furthermore, to better analyze the operational relationships between various energy devices and the complementary coupling characteristics between various energy sources, the various mathematical models are categorized based on factors such as the output of each energy source and placed in corresponding sets, thereby improving the efficiency of subsequent analysis.

[0107] Step S202 : obtaining a target operation strategy, and combining different types of models in the target mathematical model based on the target operation strategy to obtain a plurality of intermediate model combinations.

[0108] The target operation strategy refers to the strategy for operating various energy devices based on the complementary and coordinated characteristics of multiple energy sources and actual needs. It can be used to preliminarily plan the optimal capacity of energy devices corresponding to the target mathematical model. The intermediate model combination refers to the combination of mathematical models corresponding to the combination of various operating energy devices preliminarily formulated based on the target operation strategy. To plan the optimal capacity of each energy device, the actual method is to combine the mathematical models of each energy device to obtain the optimal capacity combination of each energy device.

[0109] Specifically, a target operation strategy formulated according to actual conditions is obtained, and according to the corresponding various operation rules in the operation strategy, mathematical models corresponding to the various types of operation rules are selected from the various types of mathematical models in the target mathematical model, and the combination of these mathematical models is used as an intermediate model combination. Among them, the target operation strategy includes but is not limited to the regional power supply strategy, the regional cooling strategy and the regional heating strategy. The regional power supply strategy mainly uses the photovoltaic model, the wind turbine model, the trigeneration system model and the ORC model in the energy input model set, but is not limited to the models described here. These models are all power generation models in the output model of the corresponding energy equipment. In addition, in addition to inputting the data corresponding to each mathematical model (such as wind speed, light intensity, etc.) in the regional power supply strategy, it is also necessary to input data such as power consumption, and then the mathematical model corresponding to each energy device can be used to understand the size of the output of the corresponding energy device and the power load to determine whether the supply and demand balance of the corresponding energy is achieved. The district cooling strategy primarily involves, but is not limited to, an absorption chiller model, which provides cooling through the absorption chiller model. The data used in this strategy includes, but is not limited to, the cooling power of the lithium bromide absorption chiller and data related to cooling demand. When the cooling supply is insufficient, the electric chiller begins to provide cooling to achieve a balance between the supply and demand of the corresponding energy. The district heating strategy primarily utilizes, but is not limited to, a trigeneration system model, a thermal storage tank model, and a gas boiler model. In this strategy, the input data includes, but is not limited to, the input data of the corresponding mathematical model and data related to heating demand. When the heating supply is insufficient, the thermal storage tank and electric boiler can be used to provide heating to achieve a balance between the supply and demand of the corresponding energy.

[0110] Step S204 , obtaining target constraints, screening an intermediate model combination that meets the target constraints from the multiple intermediate model combinations as a target model combination, and using the set of target model combinations as the current population to be evolved.

[0111] Target constraints refer to conditions that significantly impact the economic and environmental optimization of centralized, integrated energy operations. These conditions significantly influence the ultimate planning of the optimal capacity of each energy device and also constrain the effectiveness of various mathematical model combinations. A target model combination refers to a mathematical model combination that satisfies the target constraints. The current population to be evolved refers to a population composed of each target model combination, treated as an individual. This population is about to enter the individual evolution stage, which involves selecting, crossover, and mutating individuals within the population to obtain an incremental population. From this incremental population, outstanding individuals are continuously selected and further subjected to selection, crossover, and mutation to obtain a population that meets the corresponding conditions.

[0112] Specifically, in order to better achieve economic and environmentally friendly centralized integrated energy operation optimization, integrated energy constraints are set. The integrated energy constraints include but are not limited to three categories: network energy balance constraints, energy device design capacity constraints, and equipment operation constraints. The target constraints in this embodiment mainly involve the energy balance constraints and equipment output constraints in the network energy balance constraints and equipment operation constraints. Among them, the energy balance constraint mainly reflects the balance of electricity, cooling, and heat power in the integrated energy. The relevant formulas of the energy balance constraint are shown in equations (1) to (3). The left term of equations (1) to (3) is the sum of the energy output of each device at time t, and the right term is the sum of the energy load of each device at time t. In equation (1), C ac is the cooling capacity generated by the absorption chiller, C ec is the electric cooling capacity, C load is the cooling load; in formula (2), H cchp is the heat generated by CCHP, H gb is the heat generated by the gas boiler, H eb is the heat generated by the electric boiler, H hst The heat provided to the heat storage tank, H load is the heat load, H ac is the heat consumed by the absorption chiller; in formula (3), P wt is the power generation of the wind turbine, P pv is the photovoltaic power generation power, P cchp is the electric power generated by CCHP (CCHP stands for Combined Cooling Heating and Power, also known as combined cooling, heating and power system), P bat The electric power provided by the energy storage battery, P buy To buy electricity for Internet access, P orc is the power generation power of the ORC cycle system, P load is the electrical load, P eb is the electric load required by the electric boiler, P ec is the electrical load required by the electric refrigeration unit, and the subscripts of the formula correspond to the various currently running energy devices. The device output constraint is mainly used to constrain the output energy of each energy device to be within the specified range.

[0113] C ac (t)+C ec (t) = C load (t) (1)

[0114] H cchp (t)+H gb (t)+H eb (t)+H hst (t) = H load (t)+H ac (t) (2)

[0115] P wt (t)+P pv (t)+P cchp (t)+P bat (t)+P buy (t)+P orc (t) = P load (t)+P eb (t)+P ec (t) (3)

[0116] Its definition can be shown as formula (4), where Indicates the minimum output power of device K, Indicates the output power of device K, Indicates the maximum output power of device K. In addition, it should be noted that the weather conditions must be considered for equipment corresponding to wind power generation and photovoltaic power generation. The weather constraints of wind power generation equipment are shown in formula (5). In formula (5), is the power generated by the wind turbine, ε wt Represents the wind power experience factor, V wt represents the wind speed; the weather constraints of the power generation equipment are shown in formula (6), where is the power generated by photovoltaics, ε pv represents the solar energy experience factor, I pv Represents solar radiation intensity. When the intermediate model combination satisfies the target constraint, it is used as the target model combination. After each intermediate model combination is determined to meet the target constraint, the set of target model combinations is used as the initial population to be evolved. To ensure efficient execution of subsequent processes, the initial population to be evolved is used as the current population to be evolved, allowing subsequent processes to perform incremental processing and other operations on the current population to be evolved.

[0117]

[0118]

[0119]

[0120] Step S206 , performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generating a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated.

[0121] Incremental processing refers to the process of obtaining the intermediate population to be dominated after performing operations such as selection, crossover, and mutation on the current population to be evolved. The intermediate population to be dominated refers to the set of target model combinations obtained after performing incremental processing operations on the current population to be evolved. The target population to be dominated refers to the population for which the dominance relationship of each target model combination in the population needs to be analyzed and the crowding distance corresponding to each target model needs to be calculated. This population contains the target model combinations in the current population to be evolved and the target model combinations obtained after incremental processing.

[0122] Specifically, in order to know more possible ways to achieve economic and environmentally friendly centralized integrated energy operation optimization, the current population to be evolved can be subjected to incremental processing including operations such as selection, crossover and mutation to obtain more target model combinations. The set of target model combinations obtained by incremental processing is the intermediate population to be dominated. The process of obtaining the intermediate population to be dominated uses methods of selection, crossover and mutation on the current population to be evolved, among which the selection methods include but are not limited to roulette selection, sorting selection, optimal individual preservation, random league selection and other methods, the crossover method applies but is not limited to the method of simulating binary crossover operators, and the mutation method applies but is not limited to the method of polynomial mutation operators. Through the joint action of these methods, more target model combinations that meet the target constraints are obtained, which is conducive to finding the target model combination corresponding to the optimal operation of centralized integrated energy that can achieve economic and environmental protection.

[0123] Step S208, obtaining the dominance relationship corresponding to each target model combination in the target population to be dominated, obtaining multiple non-dominated solution sets of different levels based on the multiple non-dominated solution sets of different levels, obtaining the target crowding distance corresponding to each target model combination based on the multiple non-dominated solution sets of different levels and the each target crowding distance, determining the target population to be evolved, and taking the target population to be evolved as the current population to be evolved.

[0124] Among them, the dominance relationship refers to the relationship between whether a target model combination can lead to a better or worse final result relative to other target model combinations, and can be used to indicate the quality of a target model combination. The non-dominated solution set refers to a set of target model combinations in which there is no target model combination that makes the objective function better than the objective function corresponding to any target model combination. The objective function refers to the function corresponding to the optimization target value. The target crowding distance refers to the density of the surrounding target model combinations corresponding to each target model combination in the target population to be dominated. The target population to be evolved refers to a set of target model combinations that are screened out from the target population to be dominated based on multiple non-dominated solution sets of different levels in the target population to be dominated and each of the target crowding distances.

[0125] Specifically, the target population to be dominated contains target model combinations corresponding to before and after incremental processing. It is necessary to select a better target model combination from these target model combinations to continue incremental processing. Since there may be good and bad target model combinations before and after incremental processing, it is necessary to analyze the dominance relationship corresponding to each target model combination in the target population to be dominated, and based on the dominance relationship corresponding to each target model combination, each target model combination is divided into non-dominated solution sets of different levels, and the target crowding distance corresponding to each target model combination at different levels is calculated. Then, according to the hierarchical relationship corresponding to the non-dominated solution sets of different levels and the target crowding distance corresponding to each target model combination, the target model combination that meets the conditions is screened out from the target population to be dominated, and the set of these target model combinations that meet the conditions is used as the target population to be evolved. In order to realize the cyclical execution of this process, the target population to be evolved is also required to be used as the current population to be evolved, thereby improving the efficiency of the execution of each process and laying a good data foundation for the implementation of subsequent processes.

[0126] Step S210 , returning to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain the target population.

[0127] The target population refers to the population obtained after the evolutionary end conditions are met. The optimization target values ​​corresponding to each target model combination in this population are equal, but the combination methods are different. The optimization target value is only the objective function value corresponding to each target model combination.

[0128] Specifically, in order to obtain a better target model combination, it is necessary to continuously iterate the current population to be evolved in order to continuously optimize the combination of the target model combination, so that there are more optimized target model combinations in the obtained target population, thereby enhancing the accuracy of the target model combination corresponding to the operation optimization of centralized integrated energy that can better achieve economic and environmental protection in the subsequent process.

[0129] Step S212 , performing normalization processing on the target population to obtain a target normalization matrix, and obtaining target relative similarities corresponding to each target model combination in the target population based on the target normalization matrix.

[0130] Normalization refers to the process of processing the data and limiting it to a certain range. The target normalization matrix is ​​a matrix that represents the objective function values ​​corresponding to each target model combination in the target population. The target relative similarity refers to the degree of similarity between each target model combination in the target population and the ideal optimal target model combination.

[0131] Specifically, to better evaluate each target model combination in the target population, the data corresponding to each target model combination is normalized. This means that the data corresponding to each target model combination is converted into extremely large indicator data, where the larger the better. Furthermore, to eliminate the influence of the indicator dimensions corresponding to different data, the normalized data must be normalized. This means that each data point is assigned the same weight, thereby obtaining a target normalization matrix. Next, the target relative approximation corresponding to each target model combination is calculated based on the target normalization matrix, providing a basis for determining the optimal target model combination in subsequent processes.

[0132] Step S214 , based on the comparison results of the relative approximations of the various targets, a final model combination is screened from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0133] Among them, the final model combination refers to the target model combination that can better balance economic and environmental factors when operating the corresponding energy equipment in the final model combination, and can better plan the optimal capacity of each energy equipment.

[0134] Specifically, the greater the target relative approximation corresponding to the target model combination, the closer the target model combination is to the target model combination in the ideal state. The target model combination corresponding to the maximum target relative approximation is selected from the target population as the final model combination. The optimal capacity of the corresponding energy equipment is planned based on the combination, thereby achieving the optimal operation of economical and environmentally friendly centralized integrated energy.

[0135] The above-mentioned energy data processing method obtains a target mathematical model, combines different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations; obtains target constraints, screens intermediate model combinations that meet the target constraints from the multiple intermediate model combinations as target model combinations, and uses the set of target model combinations as the current population to be evolved; performs incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generates a target population to be dominated based on the current population to be evolved and the combination of the intermediate population to be dominated; obtains the dominance relationship corresponding to each target model combination in the target population to be dominated, obtains multiple non-dominated solution sets of different levels based on each dominance relationship, obtains the target congestion distance corresponding to each target model combination based on the multiple non-dominated solution sets of different levels, and obtains the target congestion distance corresponding to each target model combination based on the multiple non-dominated solution sets of different levels and each The target crowding distance is calculated, and the target population to be evolved is determined, and the target population to be evolved is used as the current population to be evolved; the operation of performing incremental processing on the current population to be evolved to obtain the intermediate population to be dominated is returned, until the population evolution end condition is met to obtain the target population; the target population is normalized to obtain a target normalized matrix, and based on the target normalized matrix, the target relative approximation corresponding to each target model combination in the target population is obtained; based on the comparison result of each target relative approximation, a final model combination is screened from each target model combination, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model, taking into account the complementary coupling between the energy equipment corresponding to each target mathematical model, and better planning the optimal capacity of the energy equipment corresponding to the target mathematical model, thereby improving the efficiency of achieving economic and environmentally friendly centralized integrated energy operation optimization.

[0136] In one embodiment, Figure 3 As shown, step S200 includes:

[0137] Step S300: Acquire wind power generation data, and obtain a wind turbine generator model based on the wind power generation data.

[0138] The wind power generation data refers to the data used to fit the wind turbine model, including but not limited to wind speed-power data provided by the wind turbine manufacturer and actual environmental impact data.

[0139] Specifically, polynomial fitting is performed based on the actual working scenario of the wind turbine and the wind speed-power data provided by relevant wind turbine manufacturers to obtain the wind turbine model, which is expressed as shown in formula (7). In formula (7), P WT is the actual output power of the wind turbine at a certain moment, v ci is the cut-in wind speed, v ris the rated wind speed, v ∞ is the cut-out wind speed, f(v) is the fitting polynomial, P r is the rated output power of the fan.

[0140]

[0141] Step S302: Acquire photovoltaic data, and obtain a photovoltaic model based on the photovoltaic data.

[0142] The photovoltaic data refers to the data used to fit the photovoltaic model, including but not limited to data such as solar radiation angle, incident angle, and ambient temperature.

[0143] Specifically, photovoltaic power generation needs to consider solar radiation intensity, incident angle, ambient temperature and other conditions. The photovoltaic model is obtained based on the collected photovoltaic data. The photovoltaic model can represent the output power of the corresponding photovoltaic cell. The calculation formula is shown in formula (8). In formula (8), P PV is the actual output power of photovoltaic at a certain moment, P STC is the maximum output power of photovoltaic cells under standard test conditions, G Ac is the actual sunlight intensity of the photovoltaic cell at a certain moment, G STC The solar radiation intensity under standard test conditions, in kw / m 2 ; k is the power temperature coefficient, which reflects the impact of temperature changes on the power output and other performance of photovoltaic cells. The value is -0.0047 / K; T c is the actual temperature of the environment where the photovoltaic cell is located, T STC The temperature under the standard test environment is 25℃.

[0144]

[0145] Step S304 , obtaining cooling, heating, and power trigeneration data, and fitting a cooling, heating, and power trigeneration system model based on the cooling, heating, and power trigeneration data.

[0146] Among them, trigeneration data refers to the data used to fit the trigeneration system model, which includes but is not limited to the harvested heat recovery rate, the heating coefficient of the harvested heat, the harvested heat thermal efficiency, the lower calorific value of natural gas, the electricity generated by the gas turbine, the gas turbine power generation efficiency and other data.

[0147] Specifically, the CCHP system model corresponds to the CCHP system, which is a typical integrated energy system energy supply equipment, mainly including a prime mover, a heating device, and a refrigeration device, of which the gas turbine is its core equipment. In order to analyze the relationship between the CCHP system and other energy equipment, the CCHP data are collected and the corresponding CCHP system model is obtained based on the CCHP data. The model is shown in formulas (9) to (11). In formulas (9) to (11), H gt The actual amount of heat absorbed by the system, Q gt To consume the heat generated by the fuel, η wh is the heat recovery rate, η wh，c is the cooling recovery rate of the gas turbine, δ wh is the heating coefficient of heat harvesting, δ wh，c is the cooling coefficient of the gas turbine, ε wh is the thermal efficiency of heat harvesting, COP is the coefficient of refrigeration, C gt is the cooling power generated by the gas turbine, V gt is the natural gas consumption, L NG The lower calorific value of natural gas is 9.78 (kW·h) / m 3 , Δt time interval is 1h, P gt is the electricity generated by the gas turbine, η e The efficiency of gas turbine power generation.

[0148] H gt (t) = Q gt (t)·η wh ·δ wh ·ε wh (9)

[0149]

[0150]

[0151] Step S306: Obtain gas boiler data, and obtain a gas boiler model based on the gas boiler data.

[0152] The gas boiler data refers to data used to fit a gas boiler model, including but not limited to data such as the heat conversion efficiency of the gas boiler and the natural gas consumed by the gas boiler.

[0153] Specifically, the gas boiler uses natural gas for heating, which can realize gas-heat coupling conversion. By collecting gas boiler data, the mathematical model corresponding to the gas boiler is obtained based on the gas boiler data fitting, that is, the gas boiler model. Its calculation formula is shown in formulas (12) to (13). In formulas (12) to (13), P gbis the thermal power of the gas boiler, P gas，gb is the natural gas consumption power of the gas boiler, η gb is the heat conversion efficiency of the gas boiler, V gb is the natural gas consumed by the gas boiler, L A is the lower calorific value of natural gas used in gas boilers, and Δt is the time interval.

[0154] P gb (t) = P gas，gb (t)·η gb (12)

[0155]

[0156] Step S308: Acquire absorption refrigeration data, and obtain an absorption refrigeration machine model by fitting based on the absorption refrigeration data.

[0157] The absorption refrigeration data refers to data used to fit the absorption refrigerator model, including but not limited to the refrigeration coefficient and heat input power of the lithium bromide absorption refrigerator.

[0158] Specifically, the absorption chiller is the main equipment for heat collection and utilization in the trigeneration system. By collecting absorption chiller data and fitting the absorption chiller data, the absorption chiller model corresponding to the lithium bromide absorption chiller is obtained. The calculation formula is shown in formula (14). In formula (14), P ac is the cooling power of the lithium bromide absorption refrigerator, cop ac is the refrigeration coefficient of the lithium bromide absorption refrigerator, Q ac is the heat input power of the lithium bromide absorption refrigerator.

[0159] P ac (t) = cop ac Q ac (14)

[0160] Step S310 : acquiring organic Rankine cycle data, and fitting an organic Rankine cycle model based on the organic Rankine cycle data.

[0161] The organic Rankine cycle data refers to data used to fit the organic Rankine cycle model, including but not limited to the power generated by the expander of the ORC device, the power consumed by the pump to pressurize the working medium, and the heat input power of the ORC device.

[0162] Specifically, the ORC system uses organic working fluid as the energy carrier and generates electricity through the energy conversion cycle of existing heat. It is an effective technical approach to recover low-grade thermal energy and apply it to distributed energy systems. The organic Rankine cycle model corresponding to the ORC system is obtained by collecting organic Rankine cycle data and fitting it according to the organic Rankine cycle data. The formula of the model is shown in formulas (15) to (16). In formulas (15) to (16), P orc is the net output power of the ORC device, P orc，t is the power generated by the ORC equipment expander, P orc，p is the power consumed by the pump to pressurize the working medium, η orc is the thermoelectric conversion efficiency of the ORC device, Q w is the heat input power of the ORC device.

[0163] P orc (t) = P orc，t (t)-P orc，p (t) (15)

[0164]

[0165] Step S312: acquiring energy storage battery data, and fitting an energy storage battery model based on the energy storage battery data.

[0166] Among them, the energy storage battery data refers to the data used to fit the energy storage battery model, which includes but is not limited to the battery's maximum charging power, battery charging state, upper limit of charge state, battery capacity, battery charging efficiency, battery self-discharge efficiency and other data.

[0167] Specifically, the corresponding energy storage battery model is obtained by collecting energy storage battery data and fitting. The formula of the model is shown in formulas (17) to (18). In formulas (17) to (18), P c Battery charging power, P c max is the maximum charging power of the battery, SOC is the charging state of the battery, SOC max is the upper limit of the charge state, U bat is the battery capacity, η c is the battery charging power, Δt is the unit time, δ is the battery self-discharge rate, the unit is % / h, P c is the battery charging power determined by the formula.

[0168]

[0169]

[0170] Step S314: Acquire thermal storage tank data, and obtain a thermal storage tank model based on the thermal storage tank data.

[0171] The heat storage tank data refers to the data used to fit the heat storage tank model, including but not limited to the heat release rate of the heat storage tank itself, input thermal power, input conversion efficiency and other data.

[0172] Specifically, the corresponding thermal storage tank model is obtained by collecting the thermal storage tank data and fitting. The formula of the model is shown in formula (19). In formula (19), HSS is the thermal storage state of the thermal storage tank, which is the only state variable describing the charging and discharging process of the thermal storage tank, x is the self-heat release rate, Q hst is the input thermal power of the heat storage tank, Δt is the time interval, η h is the input conversion efficiency of the thermal storage tank.

[0173] HSS(t)=(1-x)·HSS(t-1)+Q hst (t)·Δt·η h (19)

[0174] Step S316, the energy input model set includes the wind turbine model, photovoltaic model, trigeneration system model and gas boiler model, the energy conversion equipment model set includes the absorption chiller model and organic Rankine cycle model, and the energy storage model set includes the energy storage battery model and thermal storage tank model.

[0175] Specifically, in this embodiment, a wind turbine model, a photovoltaic model, a trigeneration system model, and a gas boiler model are used as mathematical models of energy input type; an absorption chiller model and an organic Rankine cycle model are used as mathematical models of energy conversion device type; and an energy storage battery model and a heat storage tank model are used as mathematical models of energy storage type. These models can be applied to corresponding actual scenarios to realize the corresponding energy conversion, thereby achieving a balance between energy supply and demand.

[0176] In this embodiment, by collecting data corresponding to each energy device and fitting a corresponding mathematical model, it is beneficial to better analyze the operating relationship between each energy device and the complementary coupling characteristics between various energy sources.

[0177] In one embodiment, Figure 4 As shown, step S206 includes:

[0178] Step S400: Obtain the fitness value corresponding to each target model combination in the current population to be evolved.

[0179] Among them, the fitness value refers to the degree of survival advantage of each target model combination in the population, which is used to distinguish the good and bad of each target model combination.

[0180] Step S402: fuse the fitness values ​​corresponding to the current population to be evolved to obtain a first total number; based on the fitness values ​​and the first total number, obtain the selection probability corresponding to the target model combination; and select a first set to be dominated from the current population to be evolved. The first set to be dominated is a set of target model combinations in the current population to be evolved whose selection probability meets the first condition.

[0181] The first total refers to the total obtained by adding up the fitness values ​​corresponding to the current population to be evolved. The selection probability refers to the probability that each target model combination will be selected. The first condition refers to the selection probability being within a specified range.

[0182] Specifically, this step uses the roulette wheel selection method to select better target model combinations from the current population to be evolved, and uses the set of selected target model combinations as the first set to be dominated for subsequent further incremental evolution of these target model combinations to get closer to the final goal.

[0183] Step S404: perform descending sorting based on the fitness values ​​corresponding to the current population to be evolved to obtain a descending sorting result; based on the descending sorting result, screen a first target number of target model combinations from the current population to be evolved; and use the set of screened target model combinations as the second set to be dominated.

[0184] The descending sorting result refers to the sorting result obtained after sorting the fitness values ​​corresponding to the current population to be evolved from large to small. The first target quantity refers to the number of relatively good target model combinations screened according to the method in this step. The second set to be dominated refers to the set of relatively good target model combinations in the current population to be evolved screened according to the method.

[0185] Specifically, this step uses a sorting selection method to screen out a target model combination that can better plan the optimal capacity of the energy equipment corresponding to the target mathematical model from the current population to be evolved, and uses the screened target model combination as the second set to be dominated to prepare for further population evolution.

[0186] Step S406 , randomly selecting a second target number of target model combinations from the current population to be evolved as random target model combinations, and screening a third dominating set from the current evolving population based on a comparison result of the fitness values ​​corresponding to the random target model combinations.

[0187] The second target number refers to a fixed value set by an artificial person. The random target model combination refers to a specified number of target model combinations randomly selected from the current population to be evolved. The third dominating set refers to a set of relatively good target model combinations selected from each random target model combination.

[0188] Specifically, this step adopts the random league selection method, sets a fixed value of the second target number, and randomly selects the second target number of target model combinations from the current population to be evolved as the random target model combination each time, and selects the random target model combination corresponding to the maximum fitness value from the random target model combination screened each time and adds it to the third dominating set.

[0189] Step S408, obtain a uniform distribution factor, repeatedly select two target model combinations from the current population to be evolved as the target model combinations to be crossed, obtain corresponding cross target model combinations based on the uniform distribution factor and the target model combinations to be crossed, until a third target number of cross target model combinations are obtained, and use the set of each cross target model combination as the fourth dominating set.

[0190] Among them, the uniform distribution factor refers to the factor that affects the distance between the cross target model combination and its corresponding target model combination in the current population to be evolved. The target model combination to be crossed refers to the target model combination that will be crossed to generate a new target model combination operation. The cross target model combination refers to the new target model combination generated after the cross operation is performed on the cross model. The third target quantity refers to the number of new target model combinations generated according to the cross operation. The fourth dominating set refers to the set of new target model combinations generated after the cross operation.

[0191] Specifically, this step adopts the method of simulating a binary crossover operator to cross the target model combination in the current population to be evolved to generate a new target model combination, which to a certain extent improves the possibility of generating more and better target combinations.

[0192] Step S410, obtain the mutation factor, repeatedly select a target model combination from the current population to be evolved as the target model combination to be mutated, obtain the upper bound target model combination and the lower bound target model combination from the current population to be evolved, and calculate the corresponding mutated target model combination based on the mutation factor, the target model combination to be mutated, the upper bound target model combination and the lower bound target model combination, until the fourth target number of mutated target model combinations are obtained, and the set of the various mutated target model combinations is used as the fifth dominating set.

[0193] Among them, the mutation factor refers to the factor that determines the differential step size of each target model combination in the current population to be evolved, thereby affecting the search for the optimal target model combination. The target model combination to be mutated refers to the target model combination that will be mutated to generate a new target model combination. The mutated target model combination refers to the new target model combination generated after the mutation operation is performed on the mutated model. The upper bound target model combination refers to the optimal target model combination in the current population to be evolved. The lower bound target model combination refers to the worst target model combination in the current population to be evolved. The fifth dominating set refers to the set of new target model combinations generated after the mutation operation.

[0194] Specifically, this step adopts the polynomial mutation operator method to obtain the mutated target model combination, so that more possible target model combinations are obtained in the generated fifth dominating set, which is conducive to improving the possibility of obtaining the corresponding target model combination when obtaining the optimal capacity of the energy equipment corresponding to the planning target mathematical model.

[0195] Step S412: obtaining the intermediate to-be-dominated population based on the combination of the first dominating set, the second dominating set, the third dominating set, the fourth dominating set, and the fifth dominating set.

[0196] Specifically, according to the corresponding selection, crossover and mutation methods in the above steps, more new target model combinations are obtained. The collection of these new target model combinations is used as the intermediate population to be dominated and used for the subsequent further evolution of the population, which improves the efficiency of obtaining the optimal target model combination to a certain extent.

[0197] In this embodiment, based on the current population to be evolved, a variety of selection, crossover and mutation methods are used to generate a new generation of population corresponding to the current population to be evolved, and more target model combinations are obtained, thereby improving the possibility of obtaining the corresponding target model combination when obtaining the optimal capacity of the energy equipment corresponding to the planning target mathematical model, which is conducive to improving the efficiency of achieving economic and environmentally friendly centralized integrated energy operation optimization.

[0198] In one embodiment, Figure 5 As shown, step S208 includes:

[0199] Step S500: Obtain an objective function, calculate function values ​​corresponding to each target model combination in the target population to be dominated based on the objective function, and obtain dominance relationships corresponding to each target model combination in the target population to be dominated based on comparison results of the function values.

[0200] The objective function refers to the optimization indicator function set to achieve the optimal capacity configuration of the integrated energy system, including but not limited to the economic objective function and the environmental objective function. The economic objective function refers to the function used to calculate the total annual cost of the integrated energy system. The environmental objective function refers to the function used to calculate the total carbon emissions.

[0201] Specifically, the determination of the dominance relationship corresponding to each target model combination in the target population to be dominated depends on the determination of the objective function. In this embodiment, the objective function includes an economic objective function and an environmental objective function, and its independent variables are shown in formula (20). In formula (20), X is the combination of the capacity of the corresponding energy equipment in each target model combination, which includes but is not limited to the corresponding equipment shown in the formula, U wt 、U PV 、U bat 、U gt 、U ac 、U hst 、U gb They correspond to the capacities of the wind turbine model, photovoltaic model, energy storage battery model, combined heating and cooling system model, absorption chiller model, heat storage tank model and gas boiler model respectively.

[0202] X=[U wt , U PV , U bat , U gt , U ac , U hst , U gb ] (20)

[0203] Among them, the economic objective function is shown in formulas (21) to (27). In formula (21), minATC refers to the minimum annual total cost of the integrated energy system, C inv is the initial investment cost, which mainly includes the purchase cost of related equipment, C o&m Refers to the operation and maintenance cost, which mainly includes the maintenance cost of the operating equipment, C re is the replacement cost, which is determined by the life cycle of the equipment and the initial investment of the equipment. e is the energy cost, which is the purchase cost of the gas consumed by the gas turbine and gas boiler and the cost of purchasing electricity from the grid when the system's internal power is insufficient; in formula (22), k is the energy device, F cr is the capital recovery coefficient, C k is the investment cost of the energy device, U k is the design capacity of equipment K; in formula (23), i is the discount rate, y is the life cycle of the integrated energy system; in formula (24), k is the unit maintenance cost of equipment K, is the output energy; in formula (25), C r，kis the initial investment cost of equipment K, F rk is the capital recovery coefficient of equipment K; in formula (26), lk is the service life of equipment K, i is the discount rate, and y is the life cycle of the integrated energy system; in formula (27), c gas is the natural gas price, is the natural gas consumption, c grid，p The price of electricity purchased from the grid, For electricity purchase.

[0204] minATC=C inv +C o&m +C re +C e (twenty one)

[0205] C inv =∑C k ·U k ·F cr (twenty two)

[0206]

[0207]

[0208] C re =∑F rk ·C r，k (25)

[0209]

[0210]

[0211] The environmental objective function is shown in formula (28), where ACE is the total carbon emissions, μ c，g is the carbon emission from natural gas combustion, in kg / kWh, is the amount of natural gas consumed by the gas turbine, μ c，e Indirect carbon emissions from purchasing from the grid, in kg / kWh. The power purchased for the power grid. In this step, the dominance relationship corresponding to each target model combination in the target population to be dominated is obtained using a genetic algorithm. To briefly describe, arbitrarily select target model combination A and target model combination B. When the function values ​​of each target function corresponding to A are all less than the function values ​​of each target function corresponding to B, A is considered to dominate B; when the function values ​​of each target function corresponding to A are all less than or equal to the function values ​​of each target function corresponding to B, and there is a target function corresponding to A whose function value is less than the function value of a target function corresponding to B, A is considered to weakly dominate B; when the function values ​​of each target function corresponding to A are all less than or equal to the function value of the target function corresponding to B, and there is a target function corresponding to A whose function value is greater than the function value of a target function corresponding to B, A and B are considered to have no dominance relationship.

[0212]

[0213] Step S502: Based on the dominance relationship corresponding to each target model combination in the target population to be dominated, a set of non-dominated target model combinations is screened out from the target population to be dominated as a first-level non-dominated solution set, the first-level non-dominated solution set is used as a current-level non-dominated solution set, and the target population to be dominated is used as the current dominated population.

[0214] The first-level non-dominated solution set refers to the set of all target model combinations in the target to-be-dominated population that are not dominated by any other target model combination. The current-level non-dominated solution set refers to the non-dominated solution set corresponding to the next level of non-dominated solution set. The current dominated population refers to the population corresponding to the deletion of each target model combination corresponding to the previous level relative to the current level.

[0215] Specifically, all target model combinations that are not dominated by any other target model combinations are screened out from the target to-be-dominated population as the first-level non-dominated solution set. In order to obtain subsequent non-dominated solution sets of different levels, the first-level non-dominated solution set is used as the current-level non-dominated solution set and the target to-be-dominated population is used as the current dominated population to ensure the execution of subsequent processes.

[0216] Step S504 : Delete the target model combination in the current dominated population that is consistent with the non-dominated solution set of the current level to obtain a target dominated population.

[0217] The target dominated population refers to the set of target model combinations that have not yet been divided into corresponding levels of non-dominated solution sets after the target models that already have corresponding levels of non-dominated solution sets are deleted.

[0218] Specifically, in order to repeatedly divide the target model combination that has been divided into the corresponding level of the non-dominated solution set, it is necessary to eliminate the target model combination that has been divided.

[0219] Step S506 , based on the dominance relationships corresponding to the target model combinations in the target dominating population, a set of non-dominated target model combinations is screened out from the target dominating population as a current level non-dominated solution set, and the target dominating population is used as the current dominating population.

[0220] Among them, the current dominant population refers to the population corresponding to the next level of division stage.

[0221] Specifically, the corresponding target model combinations are divided into the current level according to the dominance relationship division corresponding to each target model combination in the target dominating population, and the target dominating population processed in the current step is used as the current dominating population to realize the cyclic division of levels.

[0222] Step S508, returning to the operation of deleting the target model combination in the current dominated population that is consistent with the non-dominated solution set of the current level to obtain the target dominated population, until each target model combination in the target to-be-dominated population has a non-dominated solution set of the corresponding level, and obtaining the non-dominated solution sets of the multiple different levels.

[0223] Specifically, each target model combination needs to be added to the non-dominated solution set of the corresponding level. According to the hierarchical relationship between the non-dominated solution sets, a better target model combination can be effectively selected, thereby improving the processing efficiency of each process.

[0224] In this embodiment, by judging the dominance relationship corresponding to each target model combination in the target to-be-dominated population based on the objective function, and dividing each target model combination into corresponding non-dominated solution sets of different levels according to the dominance relationship, the efficiency of selecting better target model combinations that meet the corresponding conditions is improved, thereby improving the efficiency of achieving economical and environmentally friendly centralized integrated energy operation optimization to a certain extent.

[0225] In one embodiment, Figure 6 As shown, step S208 further includes:

[0226] Step S600 : obtaining initial crowding distances corresponding to respective target model combinations in the non-dominated solution sets of the multiple different levels, and using the initial crowding distances as current crowding distances.

[0227] The initial crowding distance refers to the value corresponding to the initialization of each target model, and the current crowding distance refers to the crowding distance corresponding to each target model combination in the current step.

[0228] Specifically, before calculating each target model combination, it is necessary to perform an initialization operation on the crowding distance of each target model combination in the same layer to ensure the accuracy of subsequent calculation results.

[0229] Step S602 : Selecting one non-dominated solution set from the non-dominated solution sets at different levels in sequence as a current non-dominated solution set.

[0230] The current non-dominated solution set refers to the non-dominated solution set corresponding to the current level calculated in the current step.

[0231] Specifically, the target model combinations corresponding to non-dominated solution sets at different levels are different, so the corresponding sorting results when sorting the target model combinations at the same level are also different. It is necessary to calculate the corresponding target crowding distance for the target model combinations at each level in turn.

[0232] Step S604 : selecting a function from the objective functions as a current function, and obtaining current function values ​​corresponding to each objective model combination in the current non-dominated solution set based on the current function.

[0233] The current function refers to the objective function corresponding to the current operation. The current function value refers to the function value under the current function corresponding to the target model combination.

[0234] Specifically, in order for each objective function to be processed efficiently, one function needs to be selected in turn as the current function to ensure that each objective function can be executed in an orderly manner.

[0235] Step S606, sorting the current function values ​​corresponding to each target model combination in the current non-dominated solution set to obtain a current sorting result, and based on the current sorting result, obtaining the maximum current function value and the minimum current function value corresponding to the current non-dominated solution set.

[0236] The maximum current function value refers to the maximum value among all the results obtained by using the target model combination as the variable of the current function. The minimum current function value refers to the minimum value among all the results obtained by using the target model combination as the variable of the current function.

[0237] Specifically, according to the current function value, each target model combination is sorted in ascending order, the target model combination on the edge of the sorting is marked with a large value, the crowding distance corresponding to the first target model combination is set to 0, and the maximum function value and the minimum function value corresponding to the current function are extracted for the execution of subsequent processes.

[0238] Step S608: Based on the current crowding distance, the maximum current function value, the minimum current function value corresponding to the current dominated solution set, and the current function values ​​corresponding to the current non-dominated solution set, the intermediate crowding distance corresponding to each target model combination in the current non-dominated solution set is calculated, and the intermediate crowding distance is used as the current crowding distance.

[0239] Among them, the middle crowding distance refers to the crowding distance corresponding to each target model combination calculated under the current function. The acquisition of the final result requires that each target function participates in the calculation. Therefore, in order to carry out the process in a loop, the middle distance needs to be used as the current crowding distance.

[0240] Step S610 , repeatedly selecting a function from the objective function as the current function until all functions in the objective function are selected, and obtaining the target crowding distance corresponding to each target model combination in the current dominating solution set.

[0241] The target crowding distance refers to the crowding degree corresponding to each target model combination in the same layer.

[0242] Specifically, the target crowding distance corresponding to each target model combination requires the participation of each objective function. The larger the target crowding distance, the better the corresponding target model combination. The larger the target crowding distance, the better the diversity between the target model combinations.

[0243] Step S612: Repeat the operation of selecting one non-dominated solution set from the multiple non-dominated solution sets at different levels as the current non-dominated solution set until each non-dominated solution set in the multiple non-dominated solution sets at different levels is selected, thereby obtaining the target crowding distance corresponding to each target model combination.

[0244] Specifically, to calculate the target congestion degree corresponding to each target model, it is necessary to calculate each target model combination in the non-dominated solution set corresponding to each level until each target model combination has a corresponding target congestion distance.

[0245] In this embodiment, by calculating the target congestion distance corresponding to each target model combination, it is beneficial to ensure the diversity between the target model combinations, and improve the possibility of obtaining a better target model combination, which is beneficial to improve the efficiency of obtaining the target model combination with the optimal capacity of the energy equipment corresponding to the plannable target mathematical model.

[0246] In one embodiment, Figure 7 As shown, step S208 further includes:

[0247] Step S700 : Based on the hierarchical relationship corresponding to the non-dominated solution sets at different levels, non-dominated solution sets that meet a selection condition are sequentially selected from the non-dominated solution sets at different levels as intermediate non-dominated solution sets.

[0248] Among them, the intermediate non-dominated solution set refers to multiple non-dominated solution sets corresponding to the target model combination selected and added to the target population to be evolved.

[0249] Specifically, the closer the level corresponding to the non-dominated solution set is, the better the target model combination in the non-dominated solution set is. The target model combination in the non-dominated solution set is selected layer by layer according to the level until a non-dominated solution set of a certain level is encountered so that the number of selected target model combinations meets the expected population size, and then the selection is stopped.

[0250] Step S702: taking the total number of target model combinations in the intermediate non-dominated solution set as the first target total number.

[0251] The total number of first objectives refers to the number of all objective model combinations in the selected intermediate non-dominated solution set.

[0252] Step S704: obtaining the target population size. When the first target total number is equal to the target population size, taking the intermediate non-dominated solution set as the target population to be evolved.

[0253] Among them, the target population size refers to the number of populations in each specified generation, which is a fixed value.

[0254] Specifically, when the sum of the number of target model combinations in the non-dominated solution set selected according to the hierarchy is exactly equal to the number of the target population, the selected intermediate non-dominated solution set is just used as the target population to be evolved.

[0255] Step S706 : When the first target total number is less than the target population number, obtain the next level non-dominated solution set corresponding to the intermediate non-dominated solution set.

[0256] Specifically, when the number of target model combinations corresponding to the selected intermediate non-dominated solution set does not reach the target population size, but the target model combinations of the non-dominated solution set of the next level of the last level corresponding to the intermediate dominant solution set exceed the target population size, screening is performed based on the crowding distance of each target model combination of the non-dominated solution set of the next level corresponding to the intermediate non-dominated solution set.

[0257] Step S708, sort the target crowding distances corresponding to the target model combinations in the non-dominated solution set of the next level to obtain a target sorting result, and add the corresponding target model combinations to the intermediate non-dominated solution set in turn based on the target sorting result until the number of target model combinations in the intermediate non-dominated solution set is equal to the number of the target population, thereby obtaining the target population to be evolved.

[0258] The target ranking result refers to the result of sorting the corresponding target model combinations from large to small according to the crowding distances of each target in the non-dominated solution set of the next level.

[0259] Specifically, according to the target crowding distance corresponding to each target model combination in the non-dominated solution set of the next level, they are selected in descending order and added to the intermediate non-dominated solution set. When the number of target model combinations in the intermediate non-dominated solution set is equal to the number of target populations, the screening is stopped and the intermediate non-dominated solution set is used as the target population to be evolved.

[0260] In this embodiment, the target population to be evolved is determined based on the hierarchical relationship of non-dominated solution sets at multiple different levels and the target crowding distance corresponding to each target model combination, which not only ensures the diversity of the screened target model combinations but also improves the efficiency of screening better target model combinations.

[0261] In one embodiment, Figure 8 As shown, step S212 includes:

[0262] Step S800: determining a positive ideal solution and a negative ideal solution based on the target normalized matrix.

[0263] The positive ideal solution refers to the target model combination corresponding to the optimal or most ideal situation, while the negative ideal solution refers to the target model combination corresponding to the worst or least ideal situation.

[0264] Specifically, the maximum objective function value and the minimum objective function value corresponding to each objective function can be determined according to the target normalization matrix, and the positive ideal solution and the negative ideal solution can be determined based on these maximum objective function values ​​and minimum objective function values.

[0265] Step S802 : Calculate first distances corresponding to each target model combination in the target population based on the target normalized matrix and the positive ideal solution.

[0266] Among them, the first distance refers to the gap between the target model combination and the positive ideal solution.

[0267] Specifically, to calculate the degree of similarity between each target model combination and the optimal target model combination, that is, the degree of similarity to the positive ideal solution, it is necessary to calculate the first distance to provide a data basis for the calculation of the relative similarity of the targets in the subsequent process.

[0268] Step S804 : calculating the second distance corresponding to each target model combination in the target population based on the target normalized matrix and the negative ideal solution.

[0269] Among them, the second distance refers to the gap between the target model combination and the negative ideal solution.

[0270] Specifically, to calculate the degree of similarity between each target model combination and the optimal target model combination, that is, the degree of similarity to the positive ideal solution, it is necessary to calculate the second distance to prepare data for the calculation of the target relative proximity in the subsequent process.

[0271] Step S806 , fusing the first distance and the second distance corresponding to each target model combination in the target population to obtain the total target distance corresponding to each target model combination.

[0272] Fusion refers to the operation of adding the first distance and the second distance. The total target distance refers to the sum of the distances between the target model combination and the positive ideal solution and the negative ideal solution.

[0273] Step S808 : obtaining the target relative proximity corresponding to each target model combination based on the ratio of the first distance corresponding to each target model combination to the corresponding total target distance.

[0274] Specifically, the target relative approximation is used to indicate the degree of similarity between the target model combination and the optimal target model combination (which can refer to the ideal solution here). By calculating the target relative approximation corresponding to each target model combination in the target population, a judgment basis is provided for subsequently finding the optimal target model combination in the target population.

[0275] In this embodiment, the target relative approximation corresponding to each target model combination is calculated to judge the degree of similarity between each target model combination and the optimal model combination, which provides a judgment basis for obtaining the final model combination, thereby facilitating better acquisition of the final model combination for planning the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0276] In one embodiment, the mathematical model corresponding to each energy device obtained by fitting the relevant data of energy devices in a certain industrial park is used as the target mathematical model, where the integrated energy system corresponding to the industrial park is as follows: Figure 9 As shown, Figure 9 The CCHP in the system is called Combined Cooling Heating and Power system, also known as combined cooling, heating and power system. Figure 9The integrated energy system shown in the paper consists of four parts: energy input, energy conversion, energy storage, and energy output. It uses power grid electricity, wind power, solar energy, and natural gas as energy sources to provide users with electricity, district cooling, and district heating. The corresponding device types in the integrated energy system and the mathematical models corresponding to the devices based on the operation strategy are optimized according to the execution model. The process of finally obtaining the optimal capacity solution for each device can be roughly as follows: Figure 10 As shown, Figure 11 The corresponding flow chart (actual example) of the integrated energy system operating the corresponding energy equipment according to the operation strategy is shown. Figure 11 T is the number of devices. Figure 12 The process of selecting the target model combination from the initial population to the target population and the final model combination from the target population in one embodiment is shown. Figure 12 Where Gen is the number of population evolutions, m is used to count the number of evolutions currently, and N is the original number of populations. In the embodiment, the corresponding current population to be evolved is obtained by the target mathematical model, target operation strategy and target constraint conditions corresponding to a certain industrial park, and then according to Figure 12 The heritage algorithm (also called genetic algorithm) shown in the above example obtains the Pareto solution set, which is the corresponding target population. After obtaining the target population, we can then use the Pareto solution set to obtain the target population. Figure 12 The corresponding Topsis method is used to obtain the final model combination. The optimal model combination can best plan the optimal capacity of the corresponding energy equipment, which is conducive to enhancing the diversity of the obtained target model combination, thereby increasing the possibility of achieving multi-objective operation of the integrated energy system, and to a certain extent improving the efficiency of the integrated energy system in achieving economic and environmentally friendly centralized integrated energy operation optimization.

[0277] Based on the same inventive concept, embodiments of the present application also provide an energy data processing device for implementing the aforementioned energy data processing method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more energy data processing device embodiments provided below can be found in the above-described limitations of the energy data processing method and will not be further elaborated here.

[0278] In one embodiment, Figure 13 As shown, an energy data processing device is provided, including: a model acquisition module 1300, a model combination module 1302, an initialization module 1304, an increment module 1306, a population evolution module 1308, a population determination module 1310, an evaluation module 1312 and a result determination module 1314, wherein:

[0279] The model acquisition module 1300 is used to acquire a target mathematical model, where the target mathematical model includes an energy input model set, an energy conversion device model set, and an energy storage model set.

[0280] The model combination module 1302 is used to obtain a target operation strategy, and combine different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations.

[0281] The initialization module 1304 is used to obtain target constraints, select an intermediate model combination that meets the target constraints from the multiple intermediate model combinations as a target model combination, and use the set of target model combinations as the current population to be evolved.

[0282] The increment module 1306 is configured to perform increment processing on the current population to be evolved to obtain an intermediate population to be dominated, and generate a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated.

[0283] The population evolution module 1308 is used to obtain the dominance relationship corresponding to each target model combination in the target population to be dominated, obtain multiple non-dominated solution sets of different levels based on the multiple non-dominated solution sets of different levels, obtain the target crowding distance corresponding to each target model combination based on the multiple non-dominated solution sets of different levels, determine the target population to be evolved based on the multiple non-dominated solution sets of different levels and the each target crowding distance, and use the target population to be evolved as the current population to be evolved.

[0284] The population determination module 1310 is configured to return to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until a population evolution end condition is met to obtain a target population.

[0285] The evaluation module 1312 is configured to perform normalization processing on the target population to obtain a target normalization matrix, and obtain target relative similarities corresponding to each target model combination in the target population based on the target normalization matrix.

[0286] The result determination module 1314 is used to screen the target model combinations based on the comparison results of the relative approximations of the targets to obtain a final model combination, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

[0287] In one embodiment, the model acquisition module 1300 is further used to acquire wind power generation data, and fit a wind turbine model based on the wind power generation data; acquire photovoltaic data, and fit a photovoltaic model based on the photovoltaic data; acquire trigeneration data, and fit a trigeneration system model based on the trigeneration data; acquire gas boiler data, and fit a gas boiler model based on the gas boiler data; acquire absorption refrigeration data, and fit an absorption chiller model based on the absorption refrigeration data; acquire organic Rankine cycle data, and fit an organic Rankine cycle model based on the organic Rankine cycle data; acquire energy storage battery data, and fit an energy storage battery model based on the energy storage battery data; acquire heat storage tank data, and fit a heat storage tank model based on the heat storage tank data; the energy input model set includes the wind turbine model, photovoltaic model, trigeneration system model and gas boiler model, the energy conversion device model set includes the absorption chiller model and organic Rankine cycle model, and the energy storage model set includes the energy storage battery model and heat storage tank model.

[0288] In one embodiment, the incremental module 1306 is also used to obtain the fitness value corresponding to each target model combination in the current population to be evolved; the fitness values ​​corresponding to the current population to be evolved are merged to obtain a first total number, and based on the fitness values ​​and the first total number, the selection probability corresponding to each target model combination is obtained, and a first set to be dominated is screened out from the current population to be evolved, and the first set to be dominated is a set of target model combinations in the current population to be evolved whose selection probability meets the first condition; based on the fitness values ​​corresponding to the current population to be evolved, descending sorting is obtained, and based on the descending sorting result, a first target number of target model combinations is screened from the current population to be evolved, and the set of the screened target model combinations is used as a second set to be dominated; a second target number of target model combinations is randomly selected from the current population to be evolved as a random target model combination, and based on the comparison result of the fitness values ​​corresponding to the random target model combinations, a set of target model combinations to be dominated is screened out from the current population to be evolved. A third dominating set is selected; a uniform distribution factor is obtained, and two target model combinations are repeatedly selected from the current population to be evolved as the target model combination to be crossed, and a corresponding cross target model combination is obtained based on the uniform distribution factor and the target model combination to be crossed, until a third target number of cross target model combinations are obtained, and the set of the cross target model combinations is used as the fourth dominating set; a mutation factor is obtained, and a target model combination is repeatedly selected from the current population to be evolved as the target model combination to be mutated, and an upper bound target model combination and a lower bound target model combination are obtained from the current population to be evolved, and a corresponding mutated target model combination is calculated based on the mutation factor, the target model combination to be mutated, the upper bound target model combination and the lower bound target model combination, until a fourth target number of mutated target model combinations are obtained, and the set of the mutated target model combinations is used as the fifth dominating set; the intermediate population to be dominated is obtained based on the combination of the first dominating set, the second dominating set, the third dominating set, the fourth dominating set and the fifth dominating set.

[0289] In one embodiment, the population evolution module 1308 is also used to obtain an objective function, and based on the objective function, calculate the function value corresponding to each target model combination in the target to-be-dominated population, and based on the comparison results of the various function values, obtain the dominance relationship corresponding to each target model combination in the target to-be-dominated population; based on the dominance relationship corresponding to each target model combination in the target to-be-dominated population, screen out a set of each non-dominated target model combination from the target to-be-dominated population as a first-level non-dominated solution set, use the first-level non-dominated solution set as a current-level non-dominated solution set, and use the target to-be-dominated population as a current dominating population; and use the current dominance relationship as a set of each non-dominated target model combination. The target model combinations in the breeding population that are consistent with the non-dominated solution set of the current level are deleted to obtain a target dominated population; based on the dominance relationships corresponding to the target model combinations in the target dominated population, a set of non-dominated target model combinations is screened out from the target dominated population as the non-dominated solution set of the current level, and the target dominated population is used as the current dominated population; the operation of deleting the target model combinations in the current dominated population that are consistent with the non-dominated solution set of the current level is returned to obtain the target dominated population, until each target model combination in the target to-be-dominated population has a non-dominated solution set of the corresponding level, and the non-dominated solution sets of the multiple different levels are obtained.

[0290] In one embodiment, the population evolution module 1308 is further configured to obtain an initial crowding distance corresponding to each target model combination in the non-dominated solution sets of the multiple different levels, and use the initial crowding distance as the current crowding distance; sequentially select a non-dominated solution set from the non-dominated solution sets of the multiple different levels as the current non-dominated solution set; select a function from the target function as the current function, and based on the current function, obtain the current function value corresponding to each target model combination in the current non-dominated solution set; sort the current function values ​​corresponding to each target model combination in the current non-dominated solution set to obtain a current sorting result, and based on the current sorting result, obtain the maximum current function value and the minimum current function value corresponding to the current non-dominated solution set; and sort the current function values ​​corresponding to each target model combination in the current non-dominated solution set based on the current dominating solution set. The intermediate crowding distance corresponding to each target model combination in the current non-dominated solution set is calculated based on the current crowding distance, the maximum current function value, the minimum current function value and the current function values ​​corresponding to the current non-dominated solution set, and the intermediate crowding distance is used as the current crowding distance; the operation of selecting a function from the objective function as the current function is repeated until all functions in the objective function are selected, and the target crowding distance corresponding to each target model combination in the current dominated solution set is obtained; the operation of selecting one non-dominated solution set from the non-dominated solution sets of the multiple different levels as the current non-dominated solution set is repeated until each non-dominated solution set in the non-dominated solution sets of the multiple different levels is selected, and the target crowding distance corresponding to each target model combination is obtained.

[0291] In one embodiment, the population evolution module 1308 is also used to select non-dominated solution sets that meet the selection conditions from the non-dominated solution sets of the multiple different levels in turn as intermediate non-dominated solution sets based on the hierarchical relationship corresponding to the non-dominated solution sets of the multiple different levels; use the total number of target model combinations in the intermediate non-dominated solution set as the first target total number; obtain the target population number, and when the first target total number is equal to the target population number, use the intermediate non-dominated solution set as the target population to be evolved; when the first target total number is less than the target population number, obtain the non-dominated solution set of the next level corresponding to the intermediate non-dominated solution set; sort the target crowding distances corresponding to each target model combination in the non-dominated solution set of the next level to obtain a target sorting result, and add the corresponding target model combination to the intermediate non-dominated solution set in turn based on the target sorting result until the number of target model combinations in the intermediate non-dominated solution set is equal to the target population number, thereby obtaining the target population to be evolved.

[0292] In one embodiment, the evaluation module 1312 is also used to determine a positive ideal solution and a negative ideal solution based on the target normalized matrix; calculate the first distance corresponding to each target model combination in the target population based on the target normalized matrix and the positive ideal solution; calculate the second distance corresponding to each target model combination in the target population based on the target normalized matrix and the negative ideal solution; fuse the first distance and the corresponding second distance corresponding to each target model combination in the target population to obtain the target total distance corresponding to each target model combination; based on the ratio of the first distance corresponding to each target model combination and the corresponding target total distance, obtain the target relative approximation corresponding to each target model combination.

[0293] Each module in the energy data processing device described above may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0294] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 14 As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store energy data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, an energy data processing method is implemented.

[0295] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 15As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication), or other technologies. When the computer program is executed by the processor, an energy data processing method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0296] Those skilled in the art will understand that Figure 14 and Figure 15 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0297] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method examples when executing the computer program.

[0298] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0299] In one embodiment, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of each of the above-described method embodiments.

[0300] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0301] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0302] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0303] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for processing energy data, characterized in that: The method comprises: Obtain a target mathematical model, the target mathematical model includes an energy input model set, an energy conversion device model set and an energy storage model set, including: obtaining wind power generation data, fitting a wind turbine model based on the wind power generation data; obtaining photovoltaic data, fitting a photovoltaic model based on the photovoltaic data; obtaining trigeneration data, fitting a trigeneration system model based on the trigeneration data; obtaining gas boiler data, fitting a gas boiler model based on the gas boiler data; obtaining absorption refrigeration data, fitting an absorption refrigeration machine model based on the absorption refrigeration data; obtaining organic Rankine cycle data, fitting a trigeneration system model based on the organic Rankine cycle data Fitting to obtain an organic Rankine cycle model; acquiring energy storage battery data, and fitting to obtain an energy storage battery model based on the energy storage battery data; acquiring heat storage tank data, and fitting to obtain a heat storage tank model based on the heat storage tank data; the energy input model set includes the wind turbine model, photovoltaic model, trigeneration system model, and gas boiler model, the energy conversion device model set includes the absorption chiller model and the organic Rankine cycle model, and the energy storage model set includes the energy storage battery model and the heat storage tank model; obtaining a target operation strategy, and combining different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations; Obtaining target constraints, screening an intermediate model combination that meets the target constraints from the multiple intermediate model combinations as a target model combination, and using the set of target model combinations as the current population to be evolved; the target constraints mainly involve energy balance constraints and device output constraints in network energy balance constraints and device operation constraints; Performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, and generating a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated; Obtaining dominance relationships corresponding to each target model combination in the target population to be dominated, obtaining a plurality of non-dominated solution sets of different levels based on the respective dominance relationships, obtaining target crowding distances corresponding to each target model combination based on the plurality of non-dominated solution sets of different levels, determining a target population to be evolved based on the plurality of non-dominated solution sets of different levels and the respective target crowding distances, and using the target population to be evolved as the current population to be evolved; Returning to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain the target population; Normalizing the target population to obtain a target normalized matrix, and obtaining target relative approximations corresponding to each target model combination in the target population based on the target normalized matrix; Based on the comparison results of the relative approximations of the various targets, a final model combination is screened from the various target model combinations, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

2. The method according to claim 1, characterized in that The incremental processing of the current population to be evolved to obtain the intermediate population to be dominated comprises: Obtaining the fitness value corresponding to each target model combination in the current population to be evolved; Merging the fitness values ​​corresponding to the current population to be evolved to obtain a first total, obtaining the selection probabilities corresponding to the target model combinations based on the fitness values ​​and the first total, and screening out a first set to be dominated from the current population to be evolved, where the first set to be dominated is a set of target model combinations in the current population to be evolved whose selection probabilities satisfy a first condition; Performing descending sorting based on the fitness values ​​corresponding to the current population to be evolved to obtain a descending sorting result, and based on the descending sorting result, screening a first target number of target model combinations from the current population to be evolved, and using the set of the screened target model combinations as a second set to be dominated; Randomly selecting a second target number of target model combinations from the current population to be evolved as random target model combinations, and screening a third dominating set from the current evolving population based on a comparison result of fitness values ​​corresponding to the random target model combinations; Obtaining a uniform distribution factor, repeatedly selecting two target model combinations from the current population to be evolved as target model combinations to be crossed, obtaining corresponding cross target model combinations based on the uniform distribution factor and the target model combinations to be crossed, until a third target number of cross target model combinations is obtained, and using the set of each cross target model combination as a fourth dominating set; Obtain a mutation factor, repeatedly select a target model combination from the current population to be evolved as the target model combination to be mutated, obtain an upper bound target model combination and a lower bound target model combination from the current population to be evolved, and calculate corresponding mutated target model combinations based on the mutation factor, the target model combination to be mutated, the upper bound target model combination, and the lower bound target model combination, until a fourth target number of mutated target model combinations are obtained, and use the set of the mutated target model combinations as a fifth dominating set; The intermediate population to be dominated is obtained based on the combination of the first dominating set, the second dominating set, the third dominating set, the fourth dominating set and the fifth dominating set.

3. The method according to claim 2, characterized in that The randomly selecting a second target number of target model combinations from the current population to be evolved as random target model combinations, and screening a third dominating set from the current evolving population based on a comparison result of fitness values ​​corresponding to the respective random target model combinations, includes: The random league selection method is adopted, and a fixed second target number is set. Each time, the second target number of target model combinations are randomly selected from the current population to be evolved as the random target model combination. The random target model combination corresponding to the maximum fitness value is selected from the random target model combination screened each time and added to the third dominating set.

4. The method according to claim 1, wherein The obtaining of the dominance relationships corresponding to the respective target model combinations in the target to-be-dominated population, and obtaining a plurality of non-dominated solution sets at different levels based on the respective dominance relationships, comprises: Obtaining an objective function, calculating, based on the objective function, function values ​​corresponding to each target model combination in the target population to be dominated, and obtaining, based on comparison results of the function values, dominance relationships corresponding to each target model combination in the target population to be dominated; Based on the dominance relationship corresponding to each target model combination in the target to-be-dominated population, a set of non-dominated target model combinations is screened out from the target to-be-dominated population as a first-level non-dominated solution set, the first-level non-dominated solution set is used as a current-level non-dominated solution set, and the target to-be-dominated population is used as a current dominated population; Deleting the target model combination consistent with the non-dominated solution set of the current level in the current dominated population to obtain a target dominated population; Based on the dominance relationships corresponding to the target model combinations in the target dominating population, a set of non-dominated target model combinations is screened out from the target dominating population as a non-dominated solution set of the current level, and the target dominating population is used as the current dominating population; Return to the operation of deleting the target model combination in the current dominated population that is consistent with the non-dominated solution set of the current level to obtain the target dominated population, until each target model combination in the target to-be-dominated population has a non-dominated solution set of the corresponding level, and obtain the non-dominated solution sets of the multiple different levels.

5. The method according to claim 1, characterized in that The obtaining of the target crowding distance corresponding to each target model combination based on the multiple non-dominated solution sets at different levels includes: Obtaining initial crowding distances corresponding to each target model combination in the non-dominated solution sets of the multiple different levels, and using the initial crowding distances as current crowding distances; Selecting one non-dominated solution set from the non-dominated solution sets at different levels in turn as the current non-dominated solution set; Selecting a function from the objective function as a current function, and obtaining current function values ​​corresponding to each objective model combination in the current non-dominated solution set based on the current function; Sorting the current function values ​​corresponding to the target model combinations in the current non-dominated solution set to obtain a current sorting result, and obtaining a maximum current function value and a minimum current function value corresponding to the current non-dominated solution set based on the current sorting result; Based on the current crowding distance, the maximum current function value, the minimum current function value corresponding to the current dominated solution set, and the current function values ​​corresponding to the current non-dominated solution set, calculate the intermediate crowding distance corresponding to each target model combination in the current non-dominated solution set, and use the intermediate crowding distance as the current crowding distance; Repeating the operation of selecting a function from the objective function as the current function until all functions in the objective function are selected, and obtaining the target crowding distance corresponding to each target model combination in the current dominating solution set; The operation of sequentially selecting one non-dominated solution set from the multiple non-dominated solution sets at different levels as the current non-dominated solution set is repeated until each non-dominated solution set in the multiple non-dominated solution sets at different levels is selected, thereby obtaining the target crowding distance corresponding to each target model combination.

6. The method according to claim 1, characterized in that The determining of the target population to be evolved based on the multiple non-dominated solution sets at different levels and the target crowding distances includes: Based on the hierarchical relationship corresponding to the non-dominated solution sets at different levels, sequentially selecting non-dominated solution sets that meet a selection condition from the non-dominated solution sets at different levels as intermediate non-dominated solution sets; The total number of target model combinations in the intermediate non-dominated solution set is used as the first target total number; Obtaining a target population size, and when the first target total number is equal to the target population size, using the intermediate non-dominated solution set as the target population to be evolved; When the first target total number is less than the target population number, obtaining a next-level non-dominated solution set corresponding to the intermediate non-dominated solution set; The target crowding distances corresponding to the target model combinations in the non-dominated solution set of the next level are sorted to obtain a target sorting result. Based on the target sorting result, the corresponding target model combinations are sequentially added to the intermediate non-dominated solution set until the number of target model combinations in the intermediate non-dominated solution set is equal to the number of the target population, thereby obtaining the target population to be evolved.

7. The method according to claim 1, characterized in that The step of obtaining the target relative approximation corresponding to each target model combination in the target population based on the target normalization matrix includes: determining a positive ideal solution and a negative ideal solution based on the target normalized matrix; Calculating first distances corresponding to each target model combination in the target population based on the target normalized matrix and the positive ideal solution; Calculating a second distance corresponding to each target model combination in the target population based on the target normalized matrix and the negative ideal solution; Fusing the first distances and the second distances corresponding to each target model combination in the target population to obtain a total target distance corresponding to each target model combination; Based on the ratio of the first distance corresponding to each target model combination to the corresponding target total distance, the target relative proximity corresponding to each target model combination is obtained.

8. An energy data processing device, characterized in that: The device comprises: The model acquisition module is used to acquire a target mathematical model, which includes an energy input model set, an energy conversion device model set, and an energy storage model set, including: acquiring wind power generation data, and fitting a wind turbine model based on the wind power generation data; acquiring photovoltaic data, and fitting a photovoltaic model based on the photovoltaic data; acquiring trigeneration data, and fitting a trigeneration system model based on the trigeneration data; acquiring gas boiler data, and fitting a gas boiler model based on the gas boiler data; acquiring absorption refrigeration data, and fitting an absorption refrigeration machine model based on the absorption refrigeration data; acquiring organic Rankine cycle data, and fitting an organic Rankine cycle model based on the organic Rankine cycle data. The organic Rankine cycle model is obtained by fitting the cycle data; energy storage battery data is obtained, and an energy storage battery model is obtained by fitting the energy storage battery data; heat storage tank data is obtained, and a heat storage tank model is obtained by fitting the heat storage tank data; the energy input model set includes the wind turbine model, photovoltaic model, trigeneration system model and gas boiler model, the energy conversion device model set includes the absorption chiller model and the organic Rankine cycle model, and the energy storage model set includes the energy storage battery model and the heat storage tank model; a target operation strategy is obtained, and different types of models in the target mathematical model are combined based on the target operation strategy to obtain multiple intermediate model combinations; A model combination module is used to obtain a target operation strategy, and to combine different types of models in the target mathematical model based on the target operation strategy to obtain multiple intermediate model combinations; An initialization module is used to obtain target constraints, select an intermediate model combination that meets the target constraints from the multiple intermediate model combinations as a target model combination, and use the set of target model combinations as the current population to be evolved; the target constraints mainly include energy balance constraints and device output constraints in network energy balance constraints and device operation constraints; an increment module, configured to perform increment processing on the current population to be evolved to obtain an intermediate population to be dominated, and generate a target population to be dominated based on a combination of the current population to be evolved and the intermediate population to be dominated; a population evolution module, configured to obtain dominance relationships corresponding to respective target model combinations in the target population to be dominated, obtain a plurality of non-dominated solution sets at different levels based on the respective dominance relationships, obtain target crowding distances corresponding to the respective target model combinations based on the plurality of non-dominated solution sets at different levels, determine a target population to be evolved based on the plurality of non-dominated solution sets at different levels and the respective target crowding distances, and use the target population to be evolved as the current population to be evolved; A population determination module is used to return to the operation of performing incremental processing on the current population to be evolved to obtain an intermediate population to be dominated, until the population evolution end condition is met to obtain a target population; An evaluation module is used to perform normalization processing on the target population to obtain a target normalization matrix, and obtain target relative approximations corresponding to each target model combination in the target population based on the target normalization matrix; The result determination module is used to screen the target model combinations based on the comparison results of the relative approximations of the targets to obtain a final model combination, and the final model combination is used to plan the optimal capacity of the energy equipment corresponding to the target mathematical model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

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

Images