Regional carbon emission collaborative optimization method and system

Abstract

The invention discloses a regional carbon emission collaborative optimization method and system, and the method comprises the steps: obtaining multi-dimensional flow and carbon emission data of different cities in a region, and carrying out the processing of the data; based on the influence of different multi-dimensional flow data on carbon emission, dynamically adjusting the weight of each flow element, and performing weighted synthesis to generate a comprehensive flow weight of each city node; based on the comprehensive flow weight and the carbon emission data of each city, constructing a city group total carbon emission prediction model; determining a total emission reduction target amount based on the total emission amount predicted by the model, and constructing an emission reduction amount model of each city in combination with the carbon emission amount of each city and the comprehensive flow weight of each city; and performing global optimization calculation on the collaborative emission reduction of each city by combining an optimization algorithm with the constructed emission reduction model of each city, and obtaining the carbon emission reduction of each city by maximizing the total carbon emission reduction. According to the invention, the urban agglomeration carbon emission can be accurately predicted and optimized, and the accuracy and collaboration of the urban agglomeration in the aspect of carbon emission management and control are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of environmental protection, and particularly relates to a regional carbon emission collaborative optimization method and system. BACKGROUND

[0002] Traditional carbon emission management methods mostly focus on single element analysis, such as relying only on traffic data or energy consumption for carbon emission monitoring and control. Although this method can play a certain role in certain specific situations, its limitations are also obvious: 1. Most current carbon emission optimization systems only focus on a certain type of element and fail to consider the interaction between multiple elements within a city cluster. This analysis method ignores the complexity within the city cluster and cannot fully reflect the comprehensive influence of carbon emissions between cities.

[0003] 2. Traditional carbon emission optimization methods often assign fixed weights to various flow elements. This approach cannot effectively respond to changes in flow characteristics at different development stages or different situations in different cities, leading to deviations in emission reduction effects.

[0004] 3. Most existing carbon emission optimization schemes only optimize emission reduction for a single city or a single region, lacking a collaborative emission reduction mechanism within a city cluster or across regions, resulting in suboptimal emission reduction effects within the entire city cluster.

[0005] 4. Existing technologies mostly use static data analysis, ignoring the dynamic influence of flow elements (such as population flow, economic activity, and information flow) on carbon emissions over time. This makes traditional schemes less adaptable and less accurate in predicting the rapidly changing city cluster ecosystem.

[0006] Therefore, there is an urgent need to develop a regional carbon emission collaborative optimization method to solve the above problems. SUMMARY

[0007] The present application aims to address the shortcomings of existing technologies by providing a regional carbon emission collaborative optimization method that can accurately predict and optimize city cluster carbon emissions, effectively improving the precision and collaboration of city cluster carbon emission control.

[0008] To solve the above technical problems, the present application adopts the following technical solutions: A regional carbon emission collaborative optimization method, comprising the following steps: Step 1, obtaining multi-dimensional flow data and carbon emission data of different cities in the region, and processing the data; Step 2, dynamically adjusting the weights of each flow element based on the influence of different multi-dimensional flow data on carbon emissions, and performing weighted synthesis to generate the comprehensive flow weight of each city node; Step 3, based on the comprehensive flow weight of each city obtained in step 2 and the carbon emission data obtained in step 1, a city group carbon total emission prediction model is constructed; Step 4, based on the total emission amount predicted by the model in step 3, the total emission reduction target amount is determined, and combined with the carbon emission amount of each city and its comprehensive flow weight, a city emission reduction model is constructed; Step 5, through the optimization algorithm combined with the constructed city emission reduction model, the global optimization calculation of the cooperative emission reduction amount of each city is carried out, and the optimization target is to maximize the total carbon emission reduction amount, and the carbon emission reduction amount of each city is obtained.

[0009] Further, in step 1, the multi-dimensional flow data includes economic flow data, traffic flow data, population flow data and information flow data.

[0010] Further, in step 1, the data processing includes: first, the multi-dimensional flow data is denoised by using the sliding window Z-score method respectively, then the missing values in the denoised multi-dimensional flow data are filled by using the time linear interpolation method respectively, and finally the multi-dimensional flow data is normalized.

[0011] Further, in step 2, the weight of each flow element is dynamically adjusted by using a genetic algorithm; the genetic algorithm first initializes a population, each individual is a vector containing the weight of each flow element, in each generation, the following genetic operations are performed: According to the fitness value, the individuals with fitness meeting the requirements are reserved to enter the next generation; Single-point crossover or uniform crossover is used, and in the above selected individuals, a position is randomly selected as a crossover point from the two parent individuals with the highest fitness, and the weights after the position are exchanged to generate new individuals; During the evolution process of each generation, a mutation probability is set for part of the individuals , and according to the set mutation probability, the perturbation is performed to perturb part of the weight variables of the newly generated individuals; The weight combination of all individuals in each generation population is evaluated by the fitness function, the fitness value of the optimal individual in the current generation is compared with that in the last generation, and it is judged whether to terminate iteration: when the iteration termination condition is reached, the iteration is stopped; otherwise, the iteration is continued until the termination condition is reached; After the iteration is finished, the optimal weight combination in the iteration process is reserved as the weight of each flow element.

[0012] Further, the fitness function is:

[0013] In the formula, is the fitness function, nis the number of social and economic indicators; is the determination coefficient between the comprehensive flow node intensity of the city and the jth social and economic indicator, and the calculation formula is as follows:

[0014] , wherein, is the observation value of the city on the jth social and economic indicator; is the fitting value predicted by the comprehensive flow intensity; is the average value of the jth indicator in all cities; is the number of cities; , the greater the value, the better the fitting effect, and the more reasonable the current flow element weight combination. m

[0015] Further, the iteration termination condition is one or more of the following conditions: If the preset maximum number of iterations is reached, the iteration is terminated; If the optimal individual fitness degree of the continuous several generations is improved by less than a set threshold, the iteration is terminated; If the fitness degree of a certain individual has reached a set optimal target, the iteration is terminated.

[0016] Further, in step 3, the constructed carbon total emission prediction model of the urban agglomeration is:

[0017] , wherein, represents the total carbon emission of the entire urban agglomeration, represents the carbon emission of the ith city, represents the flow element weight of the ith city, which is the weight of each type of flow element optimized in step 2 , and is used to measure the overall flowability of the city:

[0018] is the flow value of the city i in the kth flow; is the weight of each flow element optimized; total represents the total number of flow types.

[0019] Further, the method for constructing the emission reduction model of each city in step 4 includes: According to the proportion of the product of the carbon emission and the flowability weight of each city in the total weight product of the urban agglomeration, the total emission reduction task is allocated , and the calculation formula is as follows:​​​​

[0020] wherein, is the reduction amount of city i; is the carbon emission amount of city i, obtained through public channels; is the liquidity index of city i, calculated based on multi-dimensional flow data; is the total reduction target of the urban agglomeration; n is the total number of cities in the urban agglomeration.

[0021] Further, the method for calculating the carbon reduction amount of each city in step 5 through the optimization algorithm comprises: establishing a target function, the formula of which is:

[0022] f(R) is the fitness function, representing the total carbon emission reduction amount of the reduction strategy; is the initial carbon emission amount of city i; E i (R) is the given reduction strategy is the optimized carbon emission amount of city i; The optimization process is realized through a genetic algorithm, which dynamically adjusts the reduction amount of each city R i .

[0023] Another object of the present application is to provide a system for realizing the above-mentioned regional carbon emission collaborative optimization method, comprising: a data acquisition and processing module for acquiring multi-dimensional flow data and carbon emission data of different cities in the region and processing the data; a comprehensive flow weight calculation module for dynamically adjusting the weight of each flow element based on the influence of different multi-dimensional flow data on carbon emission, and performing weighted synthesis to generate the comprehensive flow weight of each city node; a city group carbon total emission prediction model construction module for constructing a city group carbon total emission prediction model based on the comprehensive flow weight of each city and the carbon emission data; a city reduction amount model construction module for determining the total reduction target amount based on the total emission amount predicted by the city group carbon total emission prediction model, and constructing a city reduction amount model in combination with the carbon emission amount of each city and its comprehensive flow weight; a carbon emission optimization module for performing global optimization calculation on the collaborative reduction amount of each city through an optimization algorithm in combination with the constructed city reduction amount model, with the optimization target being to maximize the total carbon emission reduction amount, to obtain the carbon reduction amount of each city.

[0024] Compared with the existing technology, the beneficial effects of the present invention are: the present invention dynamically adjusts the weights of flow elements through genetic algorithms, can accurately predict and optimize the carbon emissions of urban agglomerations, and can monitor the carbon emission data of each urban agglomeration in real time, effectively improving the accuracy and coordination of urban agglomerations in carbon emission management, and promoting coordinated emission reduction among urban agglomerations in the region, overcoming the limitations of traditional carbon emission management methods that rely on single-element flow analysis and ignore the interactive influence of multi-dimensional flow elements within urban agglomerations. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flow chart of a method for collaborative optimization of regional carbon emissions according to an embodiment of the present invention; Figure 2 Schematic diagram of the genetic algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0028] The present invention will be further described below with reference to specific examples, but they are not intended to limit the present invention.

[0029] like Figure 1 As shown, the embodiment of the present invention discloses a method for collaborative optimization of regional carbon emissions, comprising the following steps: Step 1: Obtain multidimensional flow data and carbon emission data from different cities in the region and process the data; In this step, multi-dimensional flow data and carbon emission data of different cities in the region are collected and processed. Multi-dimensional flow data includes economic flow data, traffic flow data, population flow data and information flow data. Specifically: Economic flow data: collect economic exchange data between cities and calculate economic flows based on the intensity of economic activities in each city:

[0030] Where: Refers to the distance between city i and city j, It represents the economic value of external services; C is the gravity coefficient, h is the distance coefficient, and its value is 1.

[0031] Traffic flow data: Collect traffic flow data between cities, including GaoDe migration willingness index, traffic connection between cities, etc.

[0032]

[0033] In the formula: GaoDe migration willingness index from city i to city j, GaoDe migration willingness index from city j to city i, Sum of GaoDe migration willingness index in city i; Sum of GaoDe migration willingness index in city j.

[0034] Population flow data: Collect data on population migration between cities, especially through Baidu migration index and other platforms.

[0035]

[0036] In the formula: Baidu migration index from city i to city j, Baidu migration index from city j to city i, Sum of Baidu migration index in city i; Sum of Baidu migration index in city j.

[0037] Information flow data: Collect information dissemination and sharing data, such as Baidu index, microblog check-in data, etc., to evaluate information flow; information flow:

[0038] In the formula: Baidu index search volume from city i to j, Baidu index search volume from city j to i, Sum of Baidu index search volume in city i; Sum of Baidu index search volume in city j.

[0039] Preprocess the original multi-dimensional flow data obtained above, including: De-noising: Use sliding window Z-score method to de-noise economic flow, traffic flow, population flow and information flow data to identify local outliers:

[0040] Where, Indicates the flow data on the t-th day; , are the local mean and standard deviation with as the center and window size 2k+1; if If the value is greater than the maximum value of the corresponding flow element or less than the minimum value of the corresponding flow element, it is considered an outlier and is replaced using linear interpolation; Missing value filling: The missing values in the denoised economic flow, traffic flow, population flow, and information flow data are filled using a time linear interpolation method:

[0041] The nearest two non-missing time points before and after; The time point to be filled The estimated value.

[0042] Standardization: Each type of flow data (economic flow, traffic flow, population flow, and information flow) is standardized to the [0, 1] interval through normalization to eliminate the impact of order-of-magnitude differences on weight learning.

[0043] Step 2, based on the influence of different multi-dimensional flow data on carbon emissions, dynamically adjust the weights of each flow element and perform weighted synthesis to generate the comprehensive flow weight of each city node ; In this step, based on the influence of different flow elements on carbon emissions, the weights of the flow elements are dynamically adjusted through a genetic genetic algorithm. The genetic genetic algorithm optimizes the weights of the flow elements through selection, crossover, and mutation operations, thereby improving the accuracy of carbon emission prediction. Specifically, as shown in Figure 2 the genetic genetic algorithm first initializes a population, with each individual being a vector containing 4 flow element weights corresponding to economic flow, traffic flow, population flow, and information flow; in each generation, the following genetic operations are performed: Selection operation: evaluate the quality of individuals according to the fitness value (such as the fitting degree of carbon emission prediction) , retain the individuals with satisfactory fitness (such as ≥ 0.95) to enter the next generation, thereby guiding the search process to converge towards the global optimal solution; Crossover operation: using single-point crossover or uniform crossover, randomly select a position as the crossover point in the two parent individuals with the highest fitness from the screened individuals, exchange the weights after the crossover point to generate new individuals; Mutation operation: to enhance the global search ability of the algorithm and avoid falling into local optimal solutions, a small probability perturbation is performed on part of the individuals in each generation evolution process. Specifically, set the mutation probability (in this embodiment, the value is 0.01–0.1), and perform perturbation operation on part of the weight variables of the newly generated individuals; The weight combination of all individuals in each generation population is evaluated by the fitness function, and the fitness value usually adopts the goodness of fit of the carbon emission prediction model (such as the coefficient of determination ) as the criterion. The fitness function is given by:

[0044] wherein, is the fitness function, is the number of socio-economic indicators; is the determination coefficient (goodness of fit) between the intensity of the urban comprehensive flow node and the i-th socio-economic indicator, which is calculated by the linear regression of:

[0045] wherein, is the observed value of the i-th socio-economic indicator of the city; is the fitted value predicted by the intensity of the comprehensive flow; is the average value of the i-th indicator in all cities; is the number of cities; , the larger the value, the better the fitting effect, and the more reasonable the current flow element weight combination. Finally, compare the fitness value of the optimal individual in the current generation with the previous generation. If any one or more of the following conditions are met, terminate the iteration: a. Maximum iteration number: reach the preset maximum iteration number (such as 100 generations or 200 generations); b. Convergence criterion: the optimal individual fitness improvement amplitude of consecutive generations (such as 10 generations) is less than the set threshold (such as Δ

[0046] <10 -4 ), which is considered to have converged; c. Target fitness threshold: if the fitness of a certain individual has reached the set optimal target (such as >0.99), terminate early; If not, continue iteration; after iteration, keep the optimal weight combination in the iteration history .

[0047] According to the obtained optimal weight combination , construct the inter-city flow index, i.e., the comprehensive flow weight of each city . Specifically, based on multi-source data such as economic flow, traffic flow, population flow, and information flow, construct the comprehensive flow weight of each city through weighted fusion , to depict the ability of resource exchange and information transmission of the city in the regional network. This index not only reflects the interaction intensity between cities, but also quantifies the potential impact on carbon emission diffusion and linkage, and its calculation formula is:​​​​​

[0048] In the formula, is the city The flow value in the k type flow (economic flow, traffic flow, etc.); is the optimized weight of each flow element; total Indicates the total number of flow types.

[0049] Step 3, based on the comprehensive flow weight of each city obtained in step 2 and the carbon emission data obtained in step 1, a city group carbon total emission prediction model is constructed; On the basis of step 2, further differentiated and executable carbon emission optimization strategies are proposed, according to the relative relationship between the carbon emission intensity and the flow capacity of each city, the emission reduction target is allocated, and the dynamic regulation and control mechanism of taking the city with high flow and high emission as the key to drive the coordinated emission reduction of surrounding cities is realized. Through this strategy, the emission reduction path and intensity can be scientifically set, and the coordinated allocation and dynamic optimization of regional carbon emission resources can be realized.

[0050] Specifically, the following city group carbon total emission prediction model is constructed to predict and quantify the overall carbon emission level of the city group:

[0051] Among them, indicates the total carbon emission of the entire city group, indicates the carbon emission of the i-th city, indicates the flow element weight of the i-th city, which is the result of weighting and synthesizing the optimized flow element weights of each type obtained in step 2 to measure the flow capacity of the city as a whole.

[0052] This model reflects the comprehensive status of the city in the multi-dimensional flow system and can dynamically reflect its "carbon output elasticity". Unlike the static or expert-assigned flow weight setting in traditional methods, the optimized flow element weight is obtained by evolutionary calculation in the embodiment of the application, which ensures that the contribution of each city in the coordinated emission reduction is more in line with the objective dynamic pattern, effectively improving the scientificity and fairness of regional carbon emission allocation.

[0053] Step 4, determining the total emission reduction target based on the total emission predicted by the model in step 3 , and combining the carbon emission of each city and its comprehensive flow weight , a city emission reduction model is constructed; ​On the basis of completing the total carbon emission prediction of urban agglomeration and the calculation of inter-city mobility weight, further differentiated emission reduction strategies at the city level are developed to achieve coordinated regulation across urban agglomerations and build city-level emission reduction models The core idea is that cities with high emissions and high mobility should bear more responsibility for emission reduction, while taking into account the development foundation and mobility capacity of each city to ensure fair and efficient emission reduction distribution.

[0054] The total emission reduction target is determined based on the total emission predicted by the model in step 3 Then, the total emission reduction task is allocated according to the proportion of the product of each city's carbon emissions and its mobility weight in the total weight product of the urban agglomeration This strategy takes into account historical emission responsibilities and introduces the flow of influence of cities in the regional network, thereby improving the scientificity and operability of the emission reduction strategy, and its calculation formula is as follows:

[0055] Where, is the emission reduction of city ; is the carbon emissions of city , obtained through public channels (such as China Carbon Emission Database CEADs, National Bureau of Statistics carbon accounting table, etc.); is the mobility index of city , calculated based on multi-dimensional flow data; is the total emission reduction target of the urban agglomeration; is the total number of cities in the urban agglomeration; Step 5, through optimization algorithm combined with the constructed city emission reduction model, the global optimization calculation of the coordinated emission reduction of each city is carried out, and the optimization target is to maximize the total carbon emission reduction, and the carbon emission reduction of each city is obtained; In this step, first, the objective function is constructed, and in constructing the objective function, the target is to maximize the total carbon emission reduction, and the formula is:

[0056] f(R) is the fitness function, which represents the total carbon emission reduction of the emission reduction strategy; is the initial carbon emissions of city ; E i (R) is the optimized carbon emissions of city under the given emission reduction strategy ; The optimization process is realized by genetic algorithm, which dynamically adjusts the emission reduction of each city Ri A genetic algorithm for allocating urban emission reduction performs global search, further improving the overall effect and fairness of urban group emission reduction. The optimization steps include: The urban emission reduction R i is encoded as an individual; The population is constructed and initialized; The fitness function f(R) is used to evaluate the individual; The selection, crossover and mutation operations are used to generate a new generation of emission reduction strategies; Each individual must meet the total emission reduction and city constraints; After multiple generations of evolution, the optimal coordinated emission reduction strategy is obtained R* ; The dynamic adjustment capability of the algorithm enables it to update the emission reduction scheme according to real-time data, improving the efficiency and fairness of regional coordinated emission reduction.

[0057] The dynamic monitoring and optimization of carbon emissions in the Yangtze River Delta urban agglomeration The Yangtze River Delta urban agglomeration includes Shanghai, Nanjing, Hangzhou and other important economic centers, with dense traffic and large population flow. With the continuous growth of population and economy in this region, carbon emissions have become an urgent environmental problem. The method of the present invention is applied to the dynamic monitoring and coordinated optimization of carbon emissions in the Yangtze River Delta region to improve the efficiency of carbon emission reduction.

[0058] Implementation steps: (1) Data collection and integration: The system obtains multi-dimensional flow data in the Yangtze River Delta urban agglomeration through real-time monitoring, including traffic flow (through Gaode traffic data), population flow (through Baidu migration index to analyze the flow between different cities), economic flow (through GDP data from the National Bureau of Statistics and local governments) and information flow (through analysis of search indexes of major Internet platforms). After data collection, the data is processed to remove outliers and interpolate missing data to ensure data accuracy.

[0059] (2) Genetic algorithm for weight optimization: In the data processing module, the genetic algorithm is used to dynamically adjust the weights of different flow elements. The algorithm gradually optimizes the weights of flow elements through multiple generations of evolution, ensuring that the influence of each flow element on carbon emissions is fully considered, thereby improving the accuracy of carbon emission prediction. After weight optimization, the urban group carbon total emission prediction model is generated according to the adjusted flow element weights.

[0060] (3) Prediction and optimization of carbon emissions: The total carbon emission of the Yangtze River Delta region is calculated through the urban agglomeration carbon total emission prediction model, and the regional carbon emission is optimized and adjusted based on the carbon emission data and flow element weight of each city. The carbon emission optimization strategy reasonably allocates the emission reduction targets of each city to ensure that the emission reduction tasks of each city meet the regional coordinated emission reduction target.

[0061] (4) Coordination emission reduction scheme and policy optimization: According to the carbon emission prediction data and flow data of each city, a coordinated emission reduction scheme is provided. The scheme includes carbon emission coordinated reduction among cities, traffic optimization, energy structure adjustment and other contents. Through the coordinated emission reduction strategy, the carbon emission of the urban agglomeration is optimized as a whole, and the carbon emission reduction strategy of each city can also be dynamically adjusted according to real-time data.

[0062] (5) Implementation effect and verification: The carbon emission of the Yangtze River Delta urban agglomeration is effectively controlled by the method of the embodiment of the present application, especially during the peak period, the carbon emission is reduced by about 15%, and the carbon emission data can be dynamically monitored in real time, and the optimization strategy is adjusted in time to ensure the emission reduction effect. The successful application of the method not only improves the carbon emission control efficiency of the Yangtze River Delta region, but also provides a reference for carbon emission reduction of other urban agglomerations.

[0063] The embodiment of the present application also provides a system for the above-mentioned regional carbon emission coordinated optimization method, comprising: A data acquisition and processing module is used to acquire multi-dimensional flow data and carbon emission data of different cities in the region, and process the data; A comprehensive flow weight calculation module is used to dynamically adjust the weight of each flow element based on the influence of different multi-dimensional flow data on carbon emission, and to perform weighted synthesis to generate the comprehensive flow weight of each city node; A city group carbon total emission prediction model construction module is used to construct a city group carbon total emission prediction model based on the comprehensive flow weight of each city and the carbon emission data; A city emission reduction amount model construction module is used to determine the total emission reduction target amount based on the total emission amount predicted by the city group carbon total emission prediction model, and to construct a city emission reduction amount model in combination with the carbon emission amount of each city and its comprehensive flow weight; A carbon emission optimization module is used to perform global optimization calculation on the coordinated emission reduction amount of each city through an optimization algorithm in combination with the constructed city emission reduction amount model, and the optimization target is to maximize the total carbon emission reduction amount to obtain the carbon emission reduction amount of each city.

[0064] The above merely preferred embodiments of the present application, and not therefore limit the embodiments and protection scope of the present application, for those skilled in the art, it should be realized that the equivalent replacement and obvious changes made by the application description, the resulting scheme should be included in the protection scope of the present application.

Claims

1. A regional carbon emission collaborative optimization method, characterized in that: The following steps are involved: Step 1: Obtain multidimensional flow data and carbon emission data from different cities in the region and process the data; Step 2: Based on the impact of different multi-dimensional flow data on carbon emissions, dynamically adjust the weights of each flow element and perform weighted synthesis to generate a comprehensive flow weight for each city node; Step 3: Based on the comprehensive flow weights of each city obtained in step 2 and the carbon emission data obtained in step 1, a total carbon emission prediction model for the urban agglomeration is constructed; Step 4: Determine the total emission reduction target based on the total emissions predicted by the model in Step 3, and build an emission reduction model for each city by combining the carbon emissions of each city with its comprehensive flow weight; Step 5: Perform global optimization calculations on the coordinated emission reductions of each city through the optimization algorithm combined with the constructed emission reduction models of each city. The optimization goal is to maximize the total reduction in carbon emissions and obtain the carbon emission reductions of each city.

2. The regional carbon emission collaborative optimization method according to claim 1, characterized in that: In step 1, the multidimensional flow data includes economic flow data, traffic flow data, population flow data and information flow data.

3. The regional carbon emission collaborative optimization method according to claim 1, characterized in that: In step 1, the data processing includes: firstly denoising the multidimensional stream data using the sliding window Z score method, then filling the missing values ​​in the denoised multidimensional stream data using time linear interpolation, and finally normalizing the multidimensional stream data.

4. The regional carbon emission collaborative optimization method according to claim 1, characterized in that: In step 2, the weights of each flow element are dynamically adjusted through a genetic algorithm. The genetic algorithm first initializes a population, where each individual is a set of vectors containing the weights of each flow element. In each generation, the following genetic operations are performed: Evaluate the quality of individuals based on their fitness values, and retain individuals that meet the fitness requirements to enter the next generation; Using single-point crossover or uniform crossover, randomly select a position as the crossover point from the two parent individuals with the highest fitness among the individuals screened above, exchange the position and the subsequent weights to generate a new individual; Set the mutation probability for some individuals in each generation of evolution , and perform perturbations according to the set mutation probability to adjust some weight variables of the newly generated individuals Perform disturbance operations; The weighted combination of all individuals in each generation of the population is evaluated by the fitness function. The fitness value of the best individual in the current generation is compared with that of the previous generation, and it is determined whether to terminate the iteration: when the iteration termination condition is reached, the iteration is stopped; Otherwise, continue iterating until the termination condition is reached; After the iteration is completed, the optimal weight combination during the iteration is retained , as the weight of each flow element.

5. The regional carbon emission collaborative optimization method according to claim 4, characterized in that: The fitness function is: Where, is the fitness function, n is the number of socioeconomic indicators; The intensity of the city's comprehensive flow nodes and the The coefficient of determination between the socioeconomic indicators is calculated as follows: in, For the city In the Observations on socioeconomic indicators; is the fitting value predicted by the comprehensive flow intensity; For the The average value of the indicator in all cities; m is the number of cities; The larger the value, the better the fitting effect, which means the more reasonable the current flow element weight combination.

6. The method for collaborative optimization of regional carbon emissions according to claim 5, characterized in that: The iteration termination conditions are one or more of the following: If the preset maximum number of iterations is reached, the iteration is terminated; If the improvement of the fitness of the best individual in several consecutive generations is less than the set threshold, the iteration is terminated; If the fitness of an individual has reached the set optimal goal, the iteration is terminated.

7. The method for collaborative optimization of regional carbon emissions according to claim 1, characterized in that: In step 3, the urban agglomeration total carbon emission prediction model is constructed as follows: in, represents the total carbon emissions of the entire urban agglomeration, represents the carbon emissions of the i-th city, represents the flow element weight of the i-th city, which is the weight of each type of flow element optimized in step 2 The weighted composite result is used to measure the overall mobility capacity of the city: is the flow value of city i in the kth type of flow; The weights of each flow element obtained by optimization; total Indicates the total number of flow types.

8. The method for collaborative optimization of regional carbon emissions according to claim 1, characterized in that: The method for constructing the emission reduction model for each city in step 4 includes: According to the carbon emissions of each city Its liquidity weight The overall emission reduction task is shared by the ratio of the product of the total weight of the urban agglomeration , which is calculated as follows: in, is the emission reduction of city i; is the carbon emissions of city i, obtained through public channels; is the mobility index of city i, calculated based on multidimensional flow data; is the total emission reduction target of the urban agglomeration; n is the total number of cities in the urban agglomeration.

9. The method for collaborative optimization of regional carbon emissions according to claim 1, characterized in that: The method for calculating the carbon emission reduction of each city through the optimization algorithm in step 5 includes: Establish the objective function, whose formula is: f(R) is the fitness function, which represents the total carbon emission reduction of the emission reduction strategy; is the initial carbon emissions of city i; E i (R) For a given emission reduction strategy Next, the optimized carbon emissions of city i; The optimization process is achieved through genetic algorithms, dynamically adjusting the emission reduction of each city. R i .

10. A system for implementing the regional carbon emission collaborative optimization method according to any one of claims 1 to 9, characterized in that: include: The data acquisition and processing module is used to obtain multi-dimensional flow data and carbon emission data of different cities in the region and process the data; The comprehensive flow weight calculation module is used to dynamically adjust the weight of each flow element based on the impact of different multi-dimensional flow data on carbon emissions, and perform weighted synthesis to generate the comprehensive flow weight of each city node; The module for constructing a model for predicting the total carbon emissions of urban agglomerations is used to construct a model for predicting the total carbon emissions of urban agglomerations based on the comprehensive flow weights and carbon emission data of each city. The module for constructing emission reduction models for each city is used to determine the total emission reduction target based on the total emissions predicted by the urban agglomeration carbon emission prediction model, and to construct emission reduction models for each city by combining the carbon emissions of each city with its comprehensive flow weight; The carbon emission optimization module is used to perform global optimization calculations on the coordinated emission reductions of each city by combining the optimization algorithm with the emission reduction models of each city. The optimization goal is to maximize the total carbon emission reduction and obtain the carbon emission reduction of each city.

Images