Smart park management method and system based on multi-source data

Abstract

The invention discloses a smart park management method and system based on multi-source data, and the method comprises the steps: carrying out the denoising of the collected multi-source data, carrying out the format conversion and normalization processing, and carrying out the integrated reconstruction of the multi-source data based on the normalization processing; carbon data is extracted from multi-source data based on integrated reconstruction for checking, and a carbon database is constructed; predicting the total carbon emission amount of the park by using the carbon emission prediction model, and obtaining the rated total carbon emission data amount of the park to determine a park carbon transaction mode; and optimizing the carbon transaction of the park by taking the optimal economic benefit as a target function in different carbon transaction modes, and guiding each industry in the park to carry out upgrading management according to optimized carbon transaction data. According to the method, the optimal distribution of the carbon emission is realized, so that the emission reduction efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart park management, and specifically to a smart park management method and system based on multi-source data. Background Art

[0002] With the rapid development of information technology, the construction of smart parks has become an important direction for urban development. Smart parks refer to advanced park development concepts that integrate new-generation information and communication technologies, have the capabilities of rapid information collection, high-speed information transmission, highly centralized computing, intelligent transaction processing, and ubiquitous service provision, and achieve timely, interactive, and integrated information perception, transmission, and processing within the park, with the goal of improving the park's industrial agglomeration capabilities, corporate economic competitiveness, and sustainable development of the park. And with the increasing attention paid to the environment, smart carbon management of parks has gradually become a trend.

[0003] The existing technologies in the field of carbon management in the park are mainly intelligent monitoring systems based on technologies such as the Internet of Things, big data and artificial intelligence, which can achieve real-time monitoring and data analysis of various carbon emission sources in the park, providing a scientific basis for emission reduction decisions. These technologies help park managers fully understand the carbon emission status and identify emission reduction potential by collecting and analyzing key data such as energy consumption, carbon emissions, and environmental quality.

[0004] However, the information integrated by the existing smart park management system is diverse, and the format of the information is not necessarily unified, which means that there are still many deficiencies in data processing, analysis and application, such as serious data island phenomenon, insufficient data analysis, insufficient decision-making support capabilities, and many parks have unclear data in carbon emission management. Data relies on periodic or task-based reporting, and a complete energy carbon monitoring system has not yet been established. This has led to the lack of unification of pollution data and carbon emission data, and it is impossible to provide accurate and comprehensive data support for carbon trading and management. Summary of the invention

[0005] The purpose of the present invention is to provide a smart park management method and system based on multi-source data, and to carry out carbon trading and management taking into account the actual situation of carbon demand in the park.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a smart park management method based on multi-source data, the steps of the smart park management method include: After denoising, the collected multi-source data is converted into a format and normalized, and the multi-source data based on normalization is integrated and reconstructed; specifically, the collected data is screened, irrelevant values ​​and erroneous values ​​are removed, and the screened data is converted into a format; the integrated reconstruction of multi-source data refers to integrating data from different sources into one database; Extract carbon data from the integrated and reconstructed multi-source data, verify the extracted carbon data, and build a carbon database based on the verified carbon data; The total carbon emissions of the park are predicted using the carbon emission prediction model, and the predicted total carbon emissions of the park are compared with the total carbon emissions data of the park rating to determine the carbon trading model of the park; the carbon trading model of the park includes the internal carbon trading model and the master-slave carbon trading model; The carbon emission prediction model is used to optimize the carbon trading in the park with the optimal economic benefit as the objective function in the park's carbon trading model, and the various industries in the park are guided to upgrade their management based on the optimized carbon trading data.

[0007] According to the above technical solution, the carbon data includes carbon emission data, carbon capture data and carbon sequestration data.

[0008] According to the above technical solution, the carbon database construction step includes: Extracting carbon emission data from the integrated and reconstructed multi-source data; the carbon emission data includes the carbon emission data of each enterprise collected by the park management agency, the carbon emission data of the corresponding enterprise calculated based on the energy consumption collected by each enterprise in the park, the carbon emission data of each enterprise obtained from the carbon emission verification report provided by a third-party agency, and the carbon emission data of each enterprise in the park calculated based on the energy consumption data provided by the energy supply enterprise; Verify the carbon emission data of each enterprise, and the execution steps include: obtaining the carbon emission data C1 of enterprise C collected by the park management agency, the carbon emission data C2 calculated based on the energy consumption collected by enterprise C itself, obtaining the carbon emission data C3 of enterprise C from the carbon emission verification report provided by a third-party agency, and the carbon emission data C4 of enterprise C in the park calculated based on the energy consumption data provided by the energy supply enterprise; Calculate the difference between the carbon emission data C1, C2, C3 and C4 The calculated difference With threshold Compare, if all differences All less than or equal to the threshold , then the carbon emission data of enterprise C after verification is the average value of carbon emission data C1, C2, C3 and C4.

[0009] and will be in all the differences All less than or equal to the threshold The enterprise carbon emission data and its corresponding enterprise theoretical energy consumption data are divided into training set and validation set, and a carbon emission prediction model is constructed using a neural network based on the training set and validation set; If there are some differences Greater than threshold , the carbon emission prediction model is used to predict the carbon emission of enterprise C, and the grid closeness of the predicted carbon emission data and carbon emission data C1, C2, C3 and C4 is calculated using the grid closeness identification method, and the mean of the carbon emission data corresponding to the maximum grid closeness and the predicted carbon emission data is used as the verified carbon emission data. Using the carbon emission prediction model to predict carbon emissions means calculating the corresponding theoretical energy consumption data of the enterprise according to the actual production plan of the enterprise, and then using the carbon emission prediction model to predict carbon emissions.

[0010] The carbon database is composed of the verified carbon emission data of each enterprise in the park and its corresponding theoretical energy consumption data, as well as the carbon capture data and carbon storage data obtained by the park.

[0011] Theoretical energy consumption data of an enterprise refers to the theoretical energy consumption data calculated based on the production plan of the enterprise.

[0012] The grid proximity calculation method is: ; In the formula, Represents the predicted carbon emissions Carbon emissions data The closeness of Represents the predicted carbon emissions Carbon emissions data The inner product of Represents the predicted carbon emissions Carbon emissions data The outer product of . =C1, C2, C3, C4.

[0013] If there are some differences greater than the threshold , it means that the accuracy of carbon emissions is relatively poor, and it is temporarily impossible to use simpler methods such as eliminating error values ​​to determine the true value. Therefore, the grid proximity method can be used to quickly calibrate the carbon emission data to improve the accuracy of the calibrated carbon emission data.

[0014] According to the above technical solution, the steps for confirming the park carbon trading model include: Optimize and adjust the carbon emission prediction model based on the verified carbon emission data of each enterprise in the carbon database and its corresponding theoretical energy consumption data; Calculate the theoretical energy consumption data of the park, use the optimized carbon emission prediction model based on the theoretical energy consumption data of the park to predict the total carbon emissions of the park, and compare the predicted total carbon emissions of the park with the total rated carbon emission data of the park; the theoretical energy consumption data of the park is the sum of the theoretical energy consumption data of all enterprises in the park; If the predicted total carbon emissions of the park are less than or equal to the total carbon emissions data of the park, the total carbon emissions data of the park will be used as the constraint condition of carbon emissions. The current carbon trading model is called the internal carbon trading model. If the predicted total carbon emissions of the park are greater than the rated total carbon emissions data of the park, the predicted total carbon emissions of the park will be used as the constraint condition for carbon emissions. The current carbon trading model is called the master-slave carbon trading model.

[0015] According to the above technical solution, the steps of the internal carbon trading model include: Conduct statistical analysis on the verified historical carbon emissions of each enterprise in the database, determine the share of each enterprise's carbon emissions in the park's rated total carbon emissions based on the proportion of each enterprise's historical annual carbon emissions to the park's total carbon emissions, and determine each enterprise's rated carbon emissions; Obtain the current carbon emissions of each enterprise from the database, and predict the total carbon emissions for the remaining years based on the current production plans of each enterprise; the enterprise production plans include industry type, the corresponding scale of the industry, and energy consumption plans; that is, calculate the corresponding theoretical energy consumption data of each enterprise according to its production plan, and use the carbon emission prediction model based on the theoretical energy consumption data of the enterprise to predict the total carbon emissions for the remaining years; each enterprise calculates the corresponding remaining carbon emissions based on the current carbon emissions and rated carbon emissions; Calculate the difference between each company's predicted total carbon emissions for the remaining year and its corresponding remaining carbon emissions , according to the difference Determine the trading role of each enterprise, which includes consumers, sellers and bystanders; the bystanders do not participate in the carbon trading process; if the difference Positive value and greater than or equal to the threshold Called a consumer, if the difference is negative and the difference The absolute value of is greater than or equal to the threshold called sellers, and in other cases bystanders; Calculate the total amount of carbon emissions that consumers need to purchase and the total amount of carbon emissions that sellers can sell, and use the priority optimization method to optimize the purchasing strategies of consumers and sellers in the park.

[0016] According to the above technical solution, the specific steps of the priority optimization method include: The total amount of carbon emissions to be purchased is greater than or equal to the threshold Consumers in the following situations are classified as execution priority 1, and consumers in other situations are classified as execution priority 2; Carbon trading within the park is carried out according to the execution priority of consumers, and consumers with priority 1 are given priority in the park carbon trading. The specific execution steps include: In the first stage, all consumers with execution priority 1 are regarded as a particle swarm, and the minimum amount of particle expenditure is used as the objective function. The particle swarm optimization algorithm is used to calculate the optimal solution set of the particle swarm; Determine whether the difference D between the carbon emission demand of the consumer corresponding to the optimal solution and the total carbon emission that can be sold by the seller at the corresponding position is greater than 0. If the difference D is greater than 0, the consumer with the difference D greater than 0 is recorded as a secondary consumer, and the difference D is used as the carbon emission demand of the corresponding secondary consumer in the second stage; When the first phase mutation operation is completed, the carbon emission demand information of all secondary consumers and consumers with execution priority 2 is used to construct a cooperative game model, and the The cooperative game model is solved by using the distribution value of each consumer obtained by the solution as the payment value of the non-cooperative game, and then the Nash equilibrium non-cooperative game is constructed to carry out the park carbon trading. The particle swarm optimization method belongs to the prior art and will not be described in detail here.

[0017] Because some companies' carbon emission needs are at the boundary of needing to purchase or not needing to purchase, some companies generally choose not to participate in carbon trading for the time being, so these companies are called bystanders. Some companies that are consumers do not have much demand for carbon trading, which makes the carbon trading process of different carbon demands inconsistent. Therefore, priority is adopted, and companies with large carbon demand are considered for carbon trading first. The priority optimization method is more specific to the actual process of carbon trading.

[0018] According to the above technical solution, the master-slave carbon trading model is expressed as: ; Among them, G represents the game model, For leaders to show carbon sales system, for the second-level leader, it represents the park, and M for the follower represents the park’s carbon capture and storage system; represents the strategy set, R represents the payoff set, ; represents a bid for buying carbon, The strategy is the carbon demand of the park, represents the carbon capture and storage capacity of the park, For leaders The objective function is Second level leader The objective function is is the objective function of the follower.

[0019] According to the above technical solution, the leader The objective function : ; The secondary leader The objective function is: ; The follower The objective function is: ; In the formula, is the total revenue of the carbon sales system, represents the price of carbon sold by the carbon selling system in the tth time period, represents the total carbon sales of the carbon sales system in the tth time period, represents the price of carbon purchased by the park in the tth time period, represents the total amount of carbon purchased by the park in the tth time period, is the total amount of energy corresponding to the total amount of carbon purchased by the park in the tth time period, represents the purchase price of energy in the tth time period, is the total amount of carbon captured in the tth time period, is the total amount of carbon sequestration in time t, Indicates the total number of time periods.

[0020] Another embodiment includes a smart park management system based on multi-source data, including: The data acquisition module performs format conversion and normalization processing on the collected multi-source data after denoising, and integrates and reconstructs the multi-source data based on the normalization processing; A data sorting module is used to extract carbon data from the integrated and reconstructed multi-source data, verify the extracted carbon data, and build a carbon database based on the verified carbon data; A carbon trading determination module, using a carbon emission prediction model to predict the total carbon emissions of the park, and comparing the predicted total carbon emissions of the park with the total carbon emissions data of the park rating to determine the carbon trading mode of the park; the carbon trading mode of the park includes an internal carbon trading mode and a master-slave carbon trading mode; The carbon trading optimization management module optimizes the park's carbon trading with the optimal economic benefit as the objective function in the park's carbon trading model, and guides the upgrade management of various industries in the park based on the optimized carbon trading data.

[0021] Compared with the prior art, the beneficial effects achieved by the present invention are: the present invention realizes real-time monitoring and data analysis of various carbon emission sources in the park to correct the carbon emission data, ensuring the accuracy and completeness of the data. The park of the present invention sets an overall carbon emission cap, and allocates carbon emission quotas to enterprises in the park according to certain rules. Through carbon trading, enterprises in the park can flexibly choose emission reduction methods according to their own emission reduction costs and the supply and demand of carbon emission quotas. Enterprises with lower emission reduction costs can choose to reduce emissions in excess and sell excess quotas, while enterprises with higher emission reduction costs can meet their emission needs by purchasing quotas. If there is no carbon emission quota in the park, a suitable carbon sales system is selected from the market through a game method for purchase. This market-oriented resource allocation method helps to achieve the optimal allocation of carbon emissions, thereby improving emission reduction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a flowchart of the steps of the smart park management method based on multi-source data of the present invention; Figure 2 It is a flow chart of the steps of the internal carbon trading model. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Example 1: Carbon database construction The steps to build a carbon database include: L1. Extract carbon emission data from integrated and reconstructed multi-source data; The carbon emission data includes the carbon emission data of each enterprise collected by the park management agency, the carbon emission data of the corresponding enterprise calculated based on the energy consumption collected by each enterprise in the park, the carbon emission data of each enterprise obtained from the carbon emission verification report provided by a third-party agency, and the carbon emission data of each enterprise in the park calculated based on the energy consumption data provided by the energy supply enterprise.

[0025] L2. Verify the carbon emission data of each enterprise and All less than or equal to the threshold The proofread enterprise carbon emission data and its corresponding enterprise theoretical energy consumption data are divided into training set and validation set, and a carbon emission prediction model is constructed based on the training set and validation set using neural network.

[0026] Taking Enterprise C as an example, the steps for verifying carbon emission data include: Obtain the carbon emission data C1 of enterprise C collected by the park management agency, the carbon emission data C2 calculated based on the energy consumption collected by enterprise C itself, the carbon emission data C3 of enterprise C obtained from the carbon emission verification report provided by a third-party agency, and the carbon emission data C4 of enterprise C in the park calculated based on the energy consumption data provided by the energy supply enterprise; Calculate the difference between the carbon emission data C1, C2, C3 and C4 The calculated difference With threshold Compare, if all differences All less than or equal to the threshold , then the carbon emission data of enterprise C after verification is the average value of carbon emission data C1, C2, C3 and C4; If there are some differences Greater than threshold , the carbon emission prediction model is used to predict the carbon emissions of enterprise C, and the grid closeness of the predicted carbon emission data of enterprise C and the carbon emission data C1, C2, C3 and C4 are calculated using the grid closeness identification method, and the average of the carbon emission data corresponding to the maximum grid closeness and the predicted carbon emission data of enterprise C is taken as the verified carbon emission data of enterprise C.

[0027] L3. The verified carbon emission data of each enterprise in the park and its corresponding theoretical energy consumption data, as well as the carbon capture data and carbon storage data obtained by the park, together form a carbon database.

[0028] Embodiment 2: A smart park management method based on multi-source data, the steps of which include: S1a, after denoising, the collected multi-source data is converted into a format and normalized, and the multi-source data based on the normalization processing is integrated and reconstructed.

[0029] S2a. Extract carbon data from the integrated and reconstructed multi-source data and verify the extracted carbon data, and build a carbon database based on the verified carbon data; the carbon data includes carbon emission data, carbon capture data, and carbon sequestration data.

[0030] S3a, using the carbon emission prediction model to predict the total carbon emissions of the park, and comparing the predicted total carbon emissions of the park with the total carbon emissions data of the park's rated carbon emissions to determine the carbon trading model of the park. Using the carbon emission prediction model, the specific steps are as follows: Optimize and adjust the carbon emission prediction model based on the verified carbon emission data of each enterprise in the carbon database and its corresponding theoretical energy consumption data; Calculate the theoretical energy consumption data of the park, use the optimized carbon emission prediction model to predict the total carbon emissions of the park based on the theoretical energy consumption data of the park, and compare the predicted total carbon emissions of the park with the total rated carbon emissions data of the park; If the predicted total carbon emissions of the park are less than or equal to the total carbon emissions data of the park, the total carbon emissions data of the park will be used as the constraint condition for carbon emissions. The current carbon trading model is called the internal carbon trading model.

[0031] S4a, using the internal carbon trading model with the optimal economic benefit as the objective function, optimize the carbon trading of the park, and guide the various industries in the park to upgrade management based on the optimized carbon trading data. The specific steps of the internal carbon trading model include: S401. Perform statistical analysis on the verified historical carbon emissions of each enterprise in the database, determine the share of each enterprise's carbon emissions in the rated total carbon emissions of the park based on the proportion of each enterprise's historical annual carbon emissions to the total carbon emissions of the entire park, and determine the rated carbon emissions of each enterprise.

[0032] S402, obtaining the current carbon emissions of each enterprise from the database, and predicting the total carbon emissions for the remaining years based on the current production plans of each enterprise; the enterprise production plans include industry types, corresponding scales of industries, and energy consumption plans.

[0033] S403. Each enterprise calculates the corresponding remaining carbon emissions based on the current carbon emissions and the rated carbon emissions.

[0034] S404. Calculate the difference between the total carbon emissions of each enterprise predicted for the remaining year and its corresponding remaining carbon emissions , according to the difference Determine the trading role of each enterprise, which includes consumers, sellers and bystanders. Bystanders do not participate in the carbon trading process.

[0035] S405. Calculate the total amount of carbon emissions that consumers need to purchase and the total amount of carbon emissions that sellers can sell, and use the priority optimization method to optimize the purchasing strategies of consumers and sellers in the park. Specifically, perform: A1. The total amount of carbon emissions to be purchased is greater than or equal to the threshold Consumers in the following situations are classified as execution priority 1, and consumers in other situations are classified as execution priority 2.

[0036] A2. Carbon trading within the park is carried out according to the execution priority of consumers. Consumers with priority 1 are given priority in the park carbon trading. The specific execution steps include: In the first stage, all consumers with execution priority 1 are regarded as a particle swarm, and the minimum amount of particle expenditure is used as the objective function. The particle swarm optimization algorithm is used to calculate the optimal solution set of the particle swarm; Determine whether the difference D between the carbon emission demand of the consumer corresponding to the optimal solution and the total carbon emission that can be sold by the seller at the corresponding position is greater than 0. If the difference D is greater than 0, the consumer with the difference D greater than 0 is recorded as a secondary consumer, and the difference D is used as the carbon emission demand of the corresponding secondary consumer in the second stage; When the first phase mutation operation is completed, the carbon emission demand information of all secondary consumers and consumers with execution priority 2 is used to construct a cooperative game model, and the The cooperative game model is solved based on the value, and the allocated value of each consumer obtained is used as the payment value of the non-cooperative game, and then the Nash equilibrium non-cooperative game is constructed to carry out carbon trading in the park.

[0037] Embodiment 3: A smart park management method based on multi-source data, comprising the following steps: S1. After denoising, the collected multi-source data is format converted and normalized, and the multi-source data based on the normalization processing is integrated and reconstructed.

[0038] S2. Extract carbon data from the integrated and reconstructed multi-source data and verify the extracted carbon data, and build a carbon database based on the verified carbon data; the carbon data includes carbon emission data, carbon capture data, and carbon storage data.

[0039] S3. Use the carbon emission prediction model to predict the total carbon emissions of the park, and compare the predicted total carbon emissions of the park with the total carbon emissions data of the park's rated carbon emissions to determine the carbon trading model of the park. Use the carbon emission prediction model, and perform the following specific steps: Optimize and adjust the carbon emission prediction model based on the verified carbon emission data of each enterprise in the carbon database and its corresponding theoretical energy consumption data; Calculate the theoretical energy consumption data of the park, use the optimized carbon emission prediction model to predict the total carbon emissions of the park based on the theoretical energy consumption data of the park, and compare the predicted total carbon emissions of the park with the total rated carbon emissions data of the park; If the predicted total carbon emissions of the park are greater than the rated total carbon emissions data of the park, the predicted total carbon emissions of the park will be used as the constraint condition for carbon emissions. The current carbon trading model is called the master-slave carbon trading model.

[0040] S4. Using the optimal economic benefit as the objective function in the master-slave carbon trading model, the carbon trading of the park is optimized, and the various industries in the park are guided to upgrade management based on the optimized carbon trading data. The master-slave carbon trading model is expressed as: ; Among them, G represents the game model, For leaders to show carbon sales system, for the second-level leader, it represents the park, and M for the follower represents the park’s carbon capture and storage system; represents the strategy set, R represents the payoff set, ; represents a bid for buying carbon, The strategy is the carbon demand of the park, represents the carbon capture and storage capacity of the park, For leaders The objective function is Second level leader The objective function is is the objective function of the follower.

[0041] Among them, leaders The objective function is: ; Secondary Leader The objective function is: ; Followers The objective function is: ; In the formula, is the total revenue of the carbon sales system, represents the price of carbon sold by the carbon selling system in the tth time period, represents the total carbon sales of the carbon sales system in the tth time period, represents the price of carbon purchased by the park in the tth time period, represents the total amount of carbon purchased by the park in the tth time period, is the total amount of energy corresponding to the total amount of carbon purchased by the park in the tth time period, represents the purchase price of energy in the tth time period, is the total amount of carbon captured in the tth time period, is the total amount of carbon sequestration in time t, Indicates the total number of time periods.

[0042] Embodiment 4: A smart park management system based on multi-source data, comprising: The data acquisition module performs format conversion and normalization processing on the collected multi-source data after denoising, and integrates and reconstructs the multi-source data based on the normalization processing; A data sorting module is used to extract carbon data from the integrated and reconstructed multi-source data, verify the extracted carbon data, and build a carbon database based on the verified carbon data; The carbon trading determination module uses the carbon emission prediction model to predict the total carbon emissions of the park, and compares the predicted total carbon emissions of the park with the total carbon emissions data of the park's rated carbon emissions to determine the carbon trading model of the park; the carbon trading model of the park includes the internal carbon trading model and the master-slave carbon trading model; The carbon trading optimization management module optimizes the park's carbon trading with the optimal economic benefit as the objective function in the park's carbon trading model, and guides the upgrade management of various industries in the park based on the optimized carbon trading data.

[0043] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0044] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A smart park management method based on multi-source data, characterized in that: The steps of the smart park management method include: After denoising, the collected multi-source data is formatted and normalized, and the normalized multi-source data is integrated and reconstructed; Extract carbon data from the integrated and reconstructed multi-source data, verify the extracted carbon data, and build a carbon database based on the verified carbon data; The total carbon emissions of the park are predicted using a carbon emission prediction model, and the predicted total carbon emissions of the park are compared with the total carbon emissions data of the park rating to determine the carbon trading model of the park; the carbon trading model of the park includes an internal carbon trading model and a master-slave carbon trading model; In the park's carbon trading model, the optimal economic benefit is used as the objective function to optimize the park's carbon trading, and the optimized carbon trading data is used to guide the upgrade management of various industries within the park.

2. The smart park management method based on multi-source data according to claim 1 is characterized in that: The carbon data includes carbon emission data, carbon capture data and carbon sequestration data.

3. The smart park management method based on multi-source data according to claim 1 is characterized in that: The carbon database construction steps include: Extracting carbon emission data from the integrated and reconstructed multi-source data; the carbon emission data includes the carbon emission data of each enterprise collected by the park management agency, the carbon emission data of the corresponding enterprise calculated based on the energy consumption collected by each enterprise in the park, the carbon emission data of each enterprise obtained from the carbon emission verification report provided by a third-party agency, and the carbon emission data of each enterprise in the park calculated based on the energy consumption data provided by the energy supply enterprise; Verify the carbon emission data of each enterprise, and the execution steps include: obtaining the carbon emission data C1 of enterprise C collected by the park management agency, the carbon emission data C2 calculated based on the energy consumption collected by enterprise C itself, obtaining the carbon emission data C3 of enterprise C from the carbon emission verification report provided by a third-party agency, and the carbon emission data C4 of enterprise C in the park calculated based on the energy consumption data provided by the energy supply enterprise; Calculate the difference between the carbon emission data C1, C2, C3 and C4 The calculated difference With threshold Compare, if all differences All less than or equal to the threshold , then the carbon emission data of enterprise C after verification is the average value of carbon emission data C1, C2, C3 and C4; In all the differences All less than or equal to the threshold The enterprise carbon emission data and its corresponding enterprise theoretical energy consumption data are divided into training set and validation set, and a carbon emission prediction model is constructed using a neural network based on the training set and validation set; If there are some differences Greater than threshold , the carbon emission prediction model is used to predict the carbon emission of enterprise C, and the grid closeness of the carbon emission data predicted by enterprise C and the carbon emission data C1, C2, C3 and C4 is calculated using the grid closeness identification method, and the mean of the carbon emission data corresponding to the maximum grid closeness and the predicted carbon emission data is used as the verified carbon emission data; The carbon database is composed of the verified carbon emission data of each enterprise in the park and its corresponding theoretical energy consumption data, as well as the carbon capture data and carbon storage data obtained by the park.

4. The smart park management method based on multi-source data according to claim 1 is characterized in that: The steps for confirming the park carbon trading model include: Optimize and adjust the carbon emission prediction model based on the verified carbon emission data of each enterprise in the carbon database and its corresponding theoretical energy consumption data; Calculate the theoretical energy consumption data of the park, use the optimized carbon emission prediction model to predict the total carbon emissions of the park based on the theoretical energy consumption data of the park, and compare the predicted total carbon emissions of the park with the total rated carbon emissions data of the park; If the predicted total carbon emissions of the park are less than or equal to the total carbon emissions data of the park, the total carbon emissions data of the park will be used as the constraint condition of carbon emissions. The current carbon trading model is called the internal carbon trading model. If the predicted total carbon emissions of the park are greater than the rated total carbon emissions data of the park, the predicted total carbon emissions of the park will be used as the constraint condition for carbon emissions. The current carbon trading model is called the master-slave carbon trading model.

5. The smart park management method based on multi-source data according to claim 4 is characterized in that: The steps of the internal carbon trading model include: Conduct statistical analysis on the verified historical carbon emissions of each enterprise in the database, determine the share of each enterprise's carbon emissions in the park's rated total carbon emissions based on the proportion of each enterprise's historical annual carbon emissions to the park's total carbon emissions, and determine each enterprise's rated carbon emissions; Obtain the current carbon emissions of each enterprise from the database, and predict the total carbon emissions for the remaining years based on the current production plans of each enterprise; the enterprise production plans include industry type, corresponding scale of the industry, and energy consumption plan; Each enterprise calculates the corresponding remaining carbon emissions based on its current carbon emissions and rated carbon emissions; Calculate the difference between each company's predicted total carbon emissions for the remaining year and its corresponding remaining carbon emissions , according to the difference Determine the trading role of each enterprise, wherein the trading roles include consumers, sellers and bystanders; the bystanders do not participate in the carbon trading process; Calculate the total amount of carbon emissions that consumers need to purchase and the total amount of carbon emissions that sellers can sell, and use the priority optimization method to optimize the purchasing strategies of consumers and sellers in the park.

6. The smart park management method based on multi-source data according to claim 5 is characterized in that: The specific steps of the priority optimization method include: The total amount of carbon emissions to be purchased is greater than or equal to the threshold Consumers in the following situations are classified as execution priority 1, and consumers in other situations are classified as execution priority 2; Carbon trading within the park is carried out according to the execution priority of consumers, and consumers with priority 1 are given priority in the park carbon trading. The specific execution steps include: In the first stage, all consumers with execution priority 1 are regarded as a particle swarm, and the minimum amount of particle expenditure is used as the objective function. The particle swarm optimization algorithm is used to calculate the optimal solution set of the particle swarm; Determine whether the difference D between the carbon emission demand of the consumer corresponding to the optimal solution and the total carbon emission that can be sold by the seller at the corresponding position is greater than 0. If the difference D is greater than 0, the consumer with the difference D greater than 0 is recorded as a secondary consumer, and the difference D is used as the carbon emission demand of the corresponding secondary consumer in the second stage; When the first phase mutation operation is completed, the carbon emission demand information of all secondary consumers and consumers with execution priority 2 is used to construct a cooperative game model, and the The cooperative game model is solved based on the value, and the allocated value of each consumer obtained is used as the payment value of the non-cooperative game, and then the Nash equilibrium non-cooperative game is constructed to carry out carbon trading in the park.

7. The smart park management method based on multi-source data according to claim 4 is characterized in that: The master-slave carbon trading model is expressed as: ; Among them, G represents the game model, For leaders to show carbon sales system, for the second-level leader, it represents the park, and M for the follower represents the park’s carbon capture and storage system; represents the strategy set, R represents the payoff set, ; represents a bid for buying carbon, The strategy is the carbon demand of the park, represents the carbon capture and storage capacity of the park, For leaders The objective function is Second level leader The objective function is is the objective function of the follower.

8. The smart park management method based on multi-source data according to claim 7 is characterized in that: The Leader The objective function is: ; The secondary leader The objective function is: ; The follower The objective function is: ; In the formula, is the total revenue of the carbon sales system, represents the price of carbon sold by the carbon selling system in the tth time period, represents the total amount of carbon sold by the carbon sales system in the tth time period, represents the price of carbon purchased by the park in the tth time period, represents the total amount of carbon purchased by the park in the tth time period, is the total amount of energy corresponding to the total amount of carbon purchased by the park in the tth time period, represents the purchase price of energy in the tth time period, is the total amount of carbon captured in the tth time period, is the total amount of carbon sequestration in time t, Indicates the total number of time periods.

9. The smart park management system based on multi-source data is characterized by: include: The data acquisition module performs format conversion and normalization processing on the collected multi-source data after denoising, and integrates and reconstructs the multi-source data based on the normalization processing; A data sorting module is used to extract carbon data from the integrated and reconstructed multi-source data, verify the extracted carbon data, and build a carbon database based on the verified carbon data; A carbon trading determination module, using a carbon emission prediction model to predict the total carbon emissions of the park, and comparing the predicted total carbon emissions of the park with the total carbon emissions data of the park rating to determine the carbon trading mode of the park; the carbon trading mode of the park includes an internal carbon trading mode and a master-slave carbon trading mode; The carbon trading optimization management module optimizes the park's carbon trading with the optimal economic benefit as the objective function in the park's carbon trading model, and guides the upgrade management of various industries in the park based on the optimized carbon trading data.

Images