Carbon emission online monitoring and quota distribution control method and device based on multi-source data fusion

Abstract

The application discloses a multi-source data fusion carbon emission online monitoring and quota allocation control method and device, relates to the technical field of carbon emission monitoring and quota management, and comprises the following steps: calculating the carbon emission intensity comprehensive score of each process unit based on multi-source data through a comprehensive scoring model, and constructing a urgency mapping function to convert the comprehensive score to obtain the carbon emission urgency of each process unit; preliminarily allocating carbon quotas according to the carbon emission urgency, and calculating the carbon quota demand of each unit; based on the carbon quota demand, constructing a quota allocation optimization model, and solving the optimization model to obtain the final quota loan amount and carbon quota of each unit. The application is used to solve the problems that the existing carbon emission quota allocation is highly subjective and it is difficult to balance fairness and efficiency.

Description

Technical Field

[0001] This invention relates to the field of carbon emission monitoring and quota management technology, and more specifically, to a method and apparatus for online carbon emission monitoring and quota allocation control based on multi-source data fusion. Background Technology

[0002] Power grid construction projects face stringent carbon emission quota restrictions and regulatory requirements during the construction phase. To achieve emission reduction targets while ensuring project progress and construction quality, construction units need to monitor carbon emissions at each construction stage and rationally allocate limited carbon emission quotas. However, current carbon emission monitoring and quota allocation methods still have many shortcomings. For example, traditional methods often allocate carbon quotas statically based on fixed quotas or historical statistical data, lacking consideration for real-time changes in operating conditions during construction. When construction plans are temporarily adjusted or the operating status of machinery and equipment fluctuates, fixed quota allocations cannot respond promptly, easily leading to surplus and waste of carbon quotas in some stages, while other stages suffer from insufficient quotas, triggering the risk of exceeding emission limits.

[0003] In current technologies, some engineering projects still rely on a single data source—the total energy consumption of construction machinery—to estimate carbon emissions. This method has limited accuracy and fails to reflect the actual impact of factors such as different machinery efficiencies, construction conditions, and operating periods on carbon emissions. Furthermore, quota allocation often depends on the experience of project managers or a rough average allocation method, lacking scientific quantitative basis and optimization strategies. This can not only lead to unreasonable quota allocation but also reduce the utilization efficiency of carbon quotas at the construction site, increasing the overall carbon reduction cost of the project. Especially in large-scale power grid construction projects involving multiple processes such as earthwork excavation, foundation construction, equipment installation, and line erection, the carbon emissions of each process unit vary significantly. Therefore, there is an urgent need for a technical solution that can integrate multi-source data, assess the carbon emission intensity of the construction process in real time, and dynamically optimize quota allocation. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for online monitoring and quota allocation control of carbon emissions based on multi-source data fusion. By fusing multi-source operational data to establish a comprehensive scoring model and combining optimized allocation with a private mean gate module, the method addresses the problems of strong subjectivity and difficulty in balancing fairness and efficiency in existing carbon emission quota allocation.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A multi-source data fusion-based online carbon emission monitoring and quota allocation control method includes the following steps: Calculating the comprehensive carbon emission intensity score of each process unit based on multi-source data using a comprehensive scoring model, and constructing an urgency mapping function to transform the comprehensive score to obtain the carbon emission urgency of each process unit; Preliminary allocation of carbon quotas based on the carbon emission urgency, and calculation of the carbon quota demand of each unit; Based on the carbon quota demand, constructing a quota allocation optimization model, and solving the optimization model to obtain the final quota borrowing amount and carbon quota for each unit.

[0007] In a preferred embodiment, the comprehensive scoring model is obtained by weighted calculation of equipment carbon intensity, process carbon intensity, and time period carbon intensity indicators.

[0008] In a preferred embodiment, the weight coefficients of the three types of indicators in the comprehensive scoring model are obtained through a machine learning method based on historical data, including the following steps: constructing a training sample set based on the historical operating data and carbon emission statistics of each process unit; training the training sample set using a machine learning algorithm to establish a fitting relationship between the comprehensive score and the actual carbon emission level, and obtaining the optimal weight coefficients of the three types of indicators: equipment carbon intensity, process carbon intensity, and time period carbon intensity.

[0009] In a preferred embodiment, the preliminary allocation of carbon quotas based on the urgency of carbon emissions specifically involves allocating quotas using an inverse priority weighted allocation method. This method uses the reciprocal of the urgency of carbon emissions as the allocation weight, ensuring that units with higher urgency receive a smaller proportion of quotas in the preliminary allocation.

[0010] In a preferred embodiment, the calculation of quota demand specifically involves calculating the carbon quota demand gap and surplus for each unit; the carbon quota demand gap refers to the insufficient portion obtained by subtracting the initially allocated quota from the carbon quota demand of each unit within its production plan cycle; the surplus is the remaining portion resulting from the initially allocated quota exceeding the actual needs of the unit.

[0011] In a preferred embodiment, the objective function of the quota allocation optimization model is to minimize the borrowing friction cost, the penalty for unmet gaps, and the fairness regularization term.

[0012] In a preferred embodiment, the constraints of the quota allocation optimization model include supply limits, demand coverage, safety buffers, phased quota conservation, and compliance rules.

[0013] In a preferred embodiment, the quota allocation optimization model is solved in a distributed manner using the alternating direction multiplier method (ADMM), comprising the following steps: based on the objective function of the quota allocation optimization model, auxiliary variables and global conserved variables are introduced to obtain an augmented Lagrangian function; through the augmented Lagrangian function, the subproblems of each unit are solved independently and local variables are updated to obtain the local update results of each unit; based on the local update results, the final quota of each unit is obtained after iterative convergence.

[0014] In a preferred embodiment, when calculating the global average quota required for the fairness regularization term during the ADMM iteration process, a private mean gate module is introduced, including the following steps: based on the quota results obtained by each unit in the iterative update, the quota is decomposed into several random shares and distributed to other units; through the random shares, each unit calculates the intermediate aggregation result locally; using the intermediate aggregation result, the global average quota for this round of iteration is reconstructed, thereby avoiding the direct exposure of individual quota data of each unit.

[0015] This invention provides a carbon emission online monitoring and quota allocation control device based on multi-source data fusion, comprising: a scoring module for calculating a comprehensive carbon emission intensity score for each process unit based on multi-source data using a comprehensive scoring model; a urgency mapping module for constructing a urgency mapping function to map the comprehensive carbon emission intensity score to the carbon emission urgency of each process unit; a demand calculation module for initially allocating carbon quotas according to the carbon emission urgency and calculating the carbon quota demand of each unit; a model building module for constructing a quota allocation optimization model based on the carbon quota demand; and an optimization solution module for solving the optimization model and outputting the final quota borrowing amount and carbon quota for each process unit.

[0016] The technical effects and advantages of the multi-source data fusion method for online carbon emission monitoring and quota allocation control in this invention are as follows:

[0017] This invention integrates multi-source carbon emission data at the equipment, process, and time periods. Based on a comprehensive scoring model, it quantifies the emission levels of each process unit and transforms the comprehensive score into a dynamic indicator directly applicable to quota allocation using an urgency mapping function. Preliminary allocation is achieved using an inverse priority weighting method, ensuring timely constraints on high-risk units and rational utilization of quotas for low-risk units, thus realizing dynamic linkage between monitoring results and quota allocation. Furthermore, by constructing a quota allocation optimization model incorporating frictional costs, gap penalties, and fairness constraints, and employing ADMM for distributed solution, it improves computational efficiency and convergence in large-scale engineering scenarios. By introducing a private mean gate module, it calculates the global average quota without exposing individual unit data, helping to maintain fairness in allocation results while protecting data privacy. This effectively solves the problems of strong subjectivity in experience-based allocation, lack of real-time monitoring support, and difficulty in balancing fairness and scalability inherent in traditional methods. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process for online carbon emission monitoring and quota allocation control based on multi-source data fusion provided in an embodiment of the present invention;

[0019] Figure 2 This is a schematic diagram of the carbon emission levels of a unit device provided in an embodiment of the present invention;

[0020] Figure 3 This is a pie chart illustrating the proportion of equipment emissions within a concrete pouring unit, provided in an embodiment of the present invention.

[0021] Figure 4 A line graph showing the comparison of carbon emission intensity between the conventional empirical method and the method of the present invention, provided for an embodiment of the present invention;

[0022] Figure 5 This is a block diagram of the multi-source data fusion carbon emission online monitoring and quota allocation control device provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1, Figure 1 This invention presents a method for online carbon emission monitoring and quota allocation control based on multi-source data fusion, comprising the following steps:

[0025] S1. Based on multi-source data, a comprehensive scoring model is used to calculate the comprehensive score of carbon emission intensity for each process unit, and an urgency mapping function is constructed to transform the comprehensive score to obtain the carbon emission urgency of each process unit.

[0026] S2, make a preliminary allocation of carbon quotas based on the urgency of carbon emissions, and calculate the carbon quota demand of each unit;

[0027] S3. Based on the carbon quota demand, construct a quota allocation optimization model, and solve the optimization model to obtain the final quota borrowing amount and carbon quota for each unit.

[0028] This embodiment introduces a comprehensive scoring model that integrates multi-source data, enabling unified quantification of equipment carbon intensity, process carbon intensity, and time-based carbon intensity. This provides a more comprehensive reflection of the carbon emission characteristics of process units compared to traditional single-indicator evaluation methods. Through an urgency mapping function, the comprehensive score is transformed into a dynamic indicator usable for quota allocation, achieving timely identification of high-emission-risk units. An inverse priority weighted initial allocation method prevents high-urgency units from consuming excessive quotas in the initial stage, leaving room for subsequent adjustments. Furthermore, a quota allocation optimization model is constructed and solved in a distributed manner using the ADMM algorithm, improving computational efficiency and ensuring scalability in large-scale engineering scenarios. Based on this, a private mean gate module is introduced to calculate the global average quota without exposing individual unit data, thus balancing fairness constraints and data privacy protection. In summary, this embodiment achieves accuracy, dynamism, fairness, and feasibility in carbon emission monitoring and quota allocation, demonstrating significant improvements over traditional empirical methods.

[0029] S1. Based on multi-source data, a comprehensive scoring model is used to calculate the comprehensive score of carbon emission intensity for each process unit, and an urgency mapping function is constructed to transform the comprehensive score to obtain the carbon emission urgency of each process unit.

[0030] In this embodiment, the specific steps for calculating the comprehensive score of carbon emission intensity of each process unit are as follows:

[0031] Collect operational data for each work unit during the construction period. Data sources include: electricity consumption, fuel consumption, and utilization rate recorded by the equipment monitoring system; work completion volume and corresponding energy consumption recorded by the work output system; and periodic emission statistics collected by the dispatch platform.

[0032] Based on the above data, calculate the carbon intensity of equipment, carbon intensity of process, and carbon intensity of time period.

[0033] The comprehensive scoring model is obtained by weighted calculation of equipment carbon intensity, process carbon intensity, and time-period carbon intensity indicators. Specifically:

[0034] Equipment carbon intensity is calculated by dividing the total carbon emissions of all equipment within a unit over the assessment period by the output of that equipment. The formula is as follows:

[0035]

[0036] in, Let k be the carbon emissions of device k. Number of unit devices This represents the output of the unit.

[0037] Process carbon intensity is obtained by dividing the carbon emissions of a process by the number of processes completed, using the following formula:

[0038]

[0039] in, This represents the total carbon emissions from the process. To complete the number of processes.

[0040] Time-period carbon intensity is used to characterize the fluctuations in carbon emissions across different construction periods. The assessment period is divided into L time periods, and emissions are collected for each period. and divide by duration The specific formula is as follows:

[0041]

[0042] Furthermore, a comprehensive scoring model is obtained by weighting the carbon intensity of equipment, process, and time period. The specific formula is as follows:

[0043]

[0044] in, For process units The overall score of carbon emission intensity , and These correspond to the weighting coefficients for the three types of carbon intensity.

[0045] The weight coefficients of the three types of indicators in the comprehensive scoring model are obtained through machine learning methods based on historical data. To avoid subjective assignment, the specific steps include:

[0046] A training sample set is constructed based on historical operational data and carbon emission statistics for each process unit. Specifically, this is achieved by collecting indicator sequences from several past engineering units. and its actual carbon emissions Construct a training sample set.

[0047] Machine learning algorithms are used to train the training sample set, establishing a fitting relationship between the comprehensive score and the actual carbon emission level. The optimal weight coefficients for three categories of indicators—equipment carbon intensity, process carbon intensity, and time-period carbon intensity—are obtained. Specifically, a linear regression algorithm is employed, aiming to minimize the error between the comprehensive score and actual emissions. The expression is as follows:

[0048]

[0049] And order , The optimal weighting coefficients can be obtained through the above process.

[0050] The specific calculation steps for obtaining the carbon emission urgency of each process unit by constructing the urgency mapping function to transform the comprehensive score are as follows:

[0051] A monotonically increasing mapping function is constructed to transform the comprehensive score into an urgency value that can be used to allocate the model. The mapping function is a logistic function, and its expression is as follows:

[0052]

[0053] in, For process units The urgency of carbon emissions Slope adjustment coefficient This is an empirical threshold. Parameter , This can be determined through nonlinear regression fitting. Specifically, it utilizes the comprehensive scores of process units from historical data. Corresponding actual carbon emissions Combined with emission limits The above-mentioned logical function is constructed; the optimal parameters are obtained by solving the problem with the objective of minimizing the deviation between the predicted urgency and historical emission levels. After training, when the comprehensive score is... Approaching the threshold corresponding to the emission limit At that time, urgency The value will rapidly approach 1, indicating that this process unit has a high urgency for emission reduction.

[0054] The formula, which aims to minimize the deviation between the forecast urgency and historical emission levels, is as follows:

[0055]

[0056] In the formula, For high and low emission labels, This is a predicted value for the urgency of carbon emissions.

[0057] By following the steps above, based on three measurable data points—equipment monitoring, process statistics, and emissions over time—a comprehensive score and urgency level for each process unit can be obtained, providing a quantitative basis for subsequent carbon quota allocation.

[0058] S2, initially allocates carbon quotas based on the urgency of carbon emissions and calculates the carbon quota requirements of each unit.

[0059] In this embodiment, the preliminary allocation of carbon quotas based on the urgency of carbon emissions specifically involves allocating quotas using an inverse priority weighted allocation method.

[0060] The inverse priority weighted allocation method uses the reciprocal of the carbon emission urgency as the allocation weight, meaning that the higher the urgency of a unit, the smaller the quota ratio it receives in the initial allocation.

[0061] Specifically:

[0062] Let the total carbon allowance be The set of process units is The urgency of each unit is To reflect that units with higher urgency receive a smaller quota in the initial allocation, an allocation weight is defined for each unit. The formula for the reciprocal of urgency is as follows:

[0063]

[0064] Next, the allocation weights are standardized to The formula is as follows:

[0065]

[0066] Accordingly, the initial allocation quota for each unit is obtained using the following formula. :

[0067]

[0068] The above method ensures that, with a fixed total quota, units with high urgency receive a reduced quota share, thus leaving more room for adjustment in subsequent optimization.

[0069] Furthermore, after the initial allocation is completed, it is necessary to calculate the quota requirements for each unit based on production plans and carbon emission forecasts. The specific steps are as follows:

[0070] The carbon quota requirement for each unit within the planning period is obtained by multiplying the planned process quantity provided by the production scheduling system by the process carbon intensity. ;

[0071] Based on the carbon quota requirements of each unit during the planning period Subtract the initial allocation quota The resulting shortfall is used to determine the carbon quota demand gap for each unit. The expression is as follows:

[0072]

[0073] Based on the preliminary allocation of quotas for each unit Exceeding the unit's carbon allowance requirements during the planning cycle The resulting surplus is used to determine the carbon allowance surplus for each unit. The expression is as follows:

[0074]

[0075] Using the above method, this step completes the initial allocation using the urgency index and clearly distinguishes the gaps and surpluses of each unit, providing input conditions for the subsequent optimized allocation model.

[0076] S3. Based on the carbon quota demand, construct a quota allocation optimization model, and solve the optimization model to obtain the final quota borrowing amount and carbon quota for each unit.

[0077] In this embodiment, the quota allocation optimization model uses the gap amount calculated by each process unit in S2. and surplus The input is quota borrowing, and the goal is to minimize the overall gap, reduce borrowing costs, and ensure fairness through quota borrowing.

[0078] First, define the decision variables. For the unit set... Set unit The amount of secondment is ,like This indicates borrowing quotas from other units, if This indicates that quotas are borrowed from other units. The final quota obtained by the calculation unit is... The formula is as follows:

[0079]

[0080] The objective function of the quota allocation optimization model is to minimize the borrowing friction cost, the penalty for unmet gaps, and the fairness regularization term, as expressed below:

[0081]

[0082] in:

[0083] The borrowing friction cost of unit i originates from the difficulty of inter-process transfer or the cost of cross-departmental coordination. It is calculated by statistically analyzing the additional energy, labor coordination, or process adjustment costs incurred by unit i for borrowing 1 unit quota in past cycles, and taking the average value as the cost. If direct economic data is lacking, the difference in carbon emission intensity between processes can be used as a substitute, as shown in the formula:

[0084]

[0085] In the formula, The average carbon emission intensity of all equipment in the unit reflects the difficulty of adjusting the unit.

[0086] The gap penalty coefficient is used to penalize unmet demand. If a gap in a certain unit is not filled, it may lead to excessive emissions, and the regulatory authorities will set penalties based on the amount of excess. Let the unit price of the regulatory penalty be F (yuan / ton CO2), then the formula for calculating the gap penalty coefficient is:

[0087]

[0088] In the formula, This is the sensitivity coefficient of the carbon emissions of this unit to the total emissions of the project;

[0089] This is a fairness regularization term used to balance the final quota distribution across all units. It measures fairness among different units using the mean-variance form, and the calculation formula is as follows:

[0090]

[0091] In the formula, The global average quota;

[0092] The weighting coefficients are calculated based on the allocation results of three historical periods, taking the average proportion of fairness variance in the overall objective function, and set within the range of 10% to 20%. For example, if a project aims to balance "fairness" and "economic cost," this can be determined empirically. .

[0093] The constraints of the quota allocation optimization model include supply limits, demand coverage, safety buffers, phased quota conservation, and compliance rules, as expressed in the following expressions:

[0094] The supply limit is that the amount lent out cannot exceed the unit's surplus:

[0095]

[0096] Demand coverage is limited to borrowing amounts that do not exceed the shortfall.

[0097]

[0098] The final quota for each unit in the security buffer should be greater than the preset security threshold. :

[0099]

[0100] The phased quota conservation means that the total amount of borrowing from the entire system is zero.

[0101]

[0102] The compliance rule stipulates that the final quota for each unit should meet the upper and lower limits set by the regulatory authorities. and :

[0103]

[0104] The quota allocation optimization model is solved in a distributed manner using the alternating direction multiplier method (ADMM). The specific steps are as follows:

[0105] Based on the objective function, auxiliary variables are introduced. With globally conserved variables Construct the augmented Lagrange function, as shown in the following expression:

[0106]

[0107] in, and As dual variables, This is the penalty parameter.

[0108] The iterative process of the alternating direction multiplier method is as follows:

[0109] In each iteration, each unit first solves its subproblem independently and updates its local variables. Its optimization subproblem is a quadratic programming problem, and the closed-form solution is:

[0110]

[0111] And it needs to be projected onto the feasible region. The internal constraints are satisfied. Then, update... The subproblems are:

[0112]

[0113] This subproblem can be solved analytically using the soft thresholding operator, as shown in the following formula:

[0114]

[0115] Simultaneously, based on the local update results, the global conserved variables are updated using the following formula:

[0116]

[0117] Then iteratively update the dual variable:

[0118]

[0119] It then checks whether the residuals converge. The iterative convergence criterion is: the original variable residual, the dual residual, and the quota conservation residual are all less than a set threshold, i.e.:

[0120]

[0121] in , , .

[0122] Once the convergence condition is met, the element The final quota is determined as follows:

[0123]

[0124] in This represents the optimal borrowing amount after convergence. Thus, efficient and secure distributed quota allocation is achieved under the constraints of global quota conservation and fairness.

[0125] During the ADMM iteration process, the global average quota required to compute the fairness regularization term is calculated. To avoid units directly exposing their respective quota data, this embodiment introduces a private mean gate module, with the specific steps as follows:

[0126] Each unit i will set its final quota It is decomposed into several random shares, specifically:

[0127] Generate m-1 random numbers ;

[0128] Set local reserved share for:

[0129]

[0130] for ,make And send it to the corresponding unit j.

[0131] In summary, all shares satisfy the following formula:

[0132]

[0133] Furthermore, each unit j receives a share from all units. And calculate locally to obtain intermediate aggregation results. :

[0134]

[0135] Finally, all units will... Adding them together, we get:

[0136]

[0137] This leads to the global average quota:

[0138]

[0139] Through the above steps, no single unit can recover the quota value of other units individually; they can only obtain the summed result, thus ensuring the privacy of quota data for each unit.

[0140] Example 2: To verify the applicability and effectiveness of the method of the present invention, two typical process units in the construction project of a 110 kV substation were selected as examples, including the earthwork excavation and transportation unit and the concrete pouring unit.

[0141] 1) Description of the case study scenario

[0142] The main equipment in the earthwork excavation and transportation unit includes two 1.0 m... 3 The equipment includes a hydraulic excavator, one loader, and five dump trucks. The construction period is 10 days, with approximately 8 hours of effective working time per day. To achieve online monitoring, fuel consumption sensors are installed on the excavator and loader, and onboard remote data terminals are installed on the trucks to collect real-time data on fuel consumption and operating time. Simultaneously, the completion rate of each work process is recorded through a construction log.

[0143] The main equipment in the concrete pouring unit includes 3 forced mixers, 2 concrete pump trucks, and 7 electric vibrators. This process is concentrated in operation for 7 days, with 12 hours of work per day. The equipment is connected to the smart meters of the temporary construction power distribution system to collect real-time data on the power consumption of the pump trucks and mixers. This data, combined with manually calculated pouring volume data, constitutes the carbon emission monitoring information for this process.

[0144] All real-time monitoring data is collected through an online monitoring platform at the construction site and integrated with manual process statistics and emission quotas to form an online updated carbon emission data stream.

[0145] 2) Carbon emission calculation and quota allocation

[0146] Based on energy consumption conversion factors, it was calculated that the earthwork excavation and transportation unit emits approximately 42.5 tons of CO2, with excavators accounting for 40%, transport trucks for 50%, and loaders for 10%; the concrete pouring unit emits approximately 28.3 tons of CO2, with mixers accounting for 55%, pump trucks for 35%, and vibrators for 10%. These results were obtained by integrating online monitoring data with quota parameters, as shown in the following figures. Figure 2 bar chart and Figure 3 The pie chart is shown below.

[0147] Following the process described in S1–S3, a comprehensive score is first calculated based on the equipment carbon intensity, process carbon intensity, and time-period carbon intensity obtained from online monitoring. Then, the carbon emission urgency value for each unit is obtained through an urgency mapping function: Earthwork excavation unit. Concrete pouring unit Then, using the inverse priority weighted allocation method, combined with the project's total CO2 quota of 60 tons, a preliminary allocation was made to the two units. The calculation results are: an initial quota of 28 tons for the excavation unit and 32 tons for the pouring unit. Compared with the predicted demand, there is a shortfall of 14.5 tons in the excavation unit and a surplus of 3.7 tons in the pouring unit.

[0148] Finally, the gap and surplus were input into the quota allocation optimization model, and the final quota was obtained by solving the ADMM and private mean gate module: 31.2 tons for excavation unit and 28.8 tons for pouring unit.

[0149] 3) Results Comparison and Analysis

[0150] Compared with the traditional quota allocation method based on experience ratio, the method of the present invention reduces the gap of the excavation unit by about 20% and improves the balance of the overall quota allocation. Figure 4 The figure shows a comparison curve of carbon emission intensity under the two methods. It can be seen that the traditional method leads to insufficient allocation of high-emission processes, while the method of this invention, driven by online monitoring data, achieves more reasonable dynamic adjustment through urgency mapping and optimization solution.

[0151] In summary, this example demonstrates that the method of the present invention can not only achieve online monitoring of carbon emissions at process units, but also ensure the fairness and rationality of quota allocation while maintaining the total quota conservation.

[0152] Example 3, Figure 5 A multi-source data fusion-based online carbon emission monitoring and quota allocation control device is presented, including:

[0153] The scoring module is used to calculate the comprehensive carbon emission intensity score of each process unit based on multi-source data and a comprehensive scoring model;

[0154] The urgency mapping module is used to construct the urgency mapping function, which maps the comprehensive carbon emission intensity score to the carbon emission urgency of each process unit;

[0155] The demand calculation module is used to initially allocate carbon quotas based on the urgency of carbon emissions and calculate the carbon quota demand of each unit.

[0156] The model building module is used to build a quota allocation optimization model based on carbon quota demand.

[0157] The optimization solution module is used to solve the optimization model and output the final quota borrowing amount and carbon quota for each process unit.

[0158] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0159] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0160] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0161] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0162] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0163] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for online monitoring and quota allocation control of carbon emissions through multi-source data fusion, characterized in that, Includes the following steps: The system collects operational data from each work unit during the construction period to enable online monitoring. Data sources include: electricity consumption, fuel consumption, and utilization rate recorded by the equipment monitoring system; the amount of work completed and the corresponding energy consumption recorded by the work output system; and periodic emission statistics collected by the dispatching platform. The comprehensive score of carbon emission intensity for each process unit is calculated based on multi-source data using a comprehensive scoring model. Construct an urgency mapping function to map the comprehensive carbon emission intensity score to the carbon emission urgency of each process unit; Based on the urgency of carbon emissions, the total carbon quota is initially allocated using the reverse priority weighted allocation method, and the carbon quota demand gap and surplus of each unit are calculated. The reverse priority weighted allocation method uses the reciprocal of the urgency of carbon emissions as the allocation weight. Based on the aforementioned demand gap and surplus, a quota allocation optimization model is constructed. The objective function of the optimization model is to minimize the borrowing friction cost, the penalty for unmet gaps, and the fairness regularization term. The constraints of the optimization model include supply limits, demand coverage, safety buffers, phased quota conservation, and compliance rules. The alternating direction multiplier method is used to solve the quota allocation optimization model in a distributed manner. During the iteration of the alternating direction multiplier method, a private mean gate module is introduced to calculate the global average quota required for the fairness regularization term, and the final quota borrowing amount and carbon quota of each process unit are output. The specific steps for calculating the global average quota are as follows: decompose the quota result obtained by each unit in the iterative update into several random shares and distribute them to other units; through the random shares, each unit calculates the intermediate aggregation result locally; and using the intermediate aggregation result, reconstruct the global average quota for this round of iteration.

2. The method for online carbon emission monitoring and quota allocation control based on multi-source data fusion according to claim 1, characterized in that, The comprehensive scoring model is obtained by weighted calculation of equipment carbon intensity, process carbon intensity, and time period carbon intensity indicators.

3. The method for online carbon emission monitoring and quota allocation control based on multi-source data fusion according to claim 2, characterized in that, The weight coefficients of the three categories of indicators in the comprehensive scoring model are obtained through machine learning methods based on historical data, including the following steps: A training sample set was constructed based on the historical operation data and carbon emission statistics of each process unit. Machine learning algorithms were used to train the training sample set to establish a fitting relationship between the comprehensive score and the actual carbon emission level, and the optimal weight coefficients of three types of indicators, namely equipment carbon intensity, process carbon intensity and time period carbon intensity, were obtained.

4. The method for online carbon emission monitoring and quota allocation control based on multi-source data fusion according to claim 3, characterized in that, The carbon quota demand gap refers to the insufficient portion obtained by each unit after subtracting the quota obtained from the initial allocation based on its carbon quota demand within its production planning cycle. The surplus is the remaining amount resulting from the initial allocation exceeding the actual needs of the unit.

5. The method for online monitoring and quota allocation control of carbon emissions based on multi-source data fusion according to claim 4, characterized in that, The distributed solution of the quota allocation optimization model using the alternating direction multiplier method includes the following steps: Based on the objective function of the quota allocation optimization model, an augmented Lagrange function is obtained by introducing auxiliary variables and global conserved variables. By using the augmented Lagrangian function, the subproblems of each element are solved independently and local variables are updated to obtain the local update results of each element. Based on the local update results, the final quota for each unit is obtained after iterative convergence.

6. A carbon emission online monitoring and quota allocation control device based on multi-source data fusion, characterized in that, include: The online monitoring module is used to collect operational data of each process unit during the construction cycle to achieve online monitoring. Data sources include: power consumption, fuel consumption and utilization rate recorded by the equipment monitoring system; process output records of process completion and corresponding energy consumption; and time-period emission statistics collected by the dispatching platform. The scoring module is used to calculate the comprehensive carbon emission intensity score of each process unit based on multi-source data and a comprehensive scoring model; The urgency mapping module is used to construct the urgency mapping function, which maps the comprehensive carbon emission intensity score to the carbon emission urgency of each process unit; The demand calculation module is used to initially allocate the total carbon quota based on the urgency of carbon emissions using an inverse priority weighted allocation method, and to calculate the carbon quota demand gap and surplus of each unit; the inverse priority weighted allocation method uses the reciprocal of the urgency of carbon emissions as the allocation weight. The model building module is used to construct a quota allocation optimization model based on the demand gap and surplus. The objective function of the optimization model is to minimize the borrowing friction cost, the penalty for unmet gap and the fairness regularization term. The constraints of the optimization model include supply limit, demand coverage, safety buffer, phase quota conservation and compliance rules. The optimization solution module is used to perform distributed solution of the quota allocation optimization model using the alternating direction multiplier method. During the iteration process of the alternating direction multiplier method, a private mean gate module is introduced to calculate the global average quota required for the fairness regularization term, and outputs the final quota borrowing amount and carbon quota of each process unit. The specific steps for calculating the global average quota are as follows: decompose the quota result obtained by each unit in the iterative update into several random shares and distribute them to other units; through the random shares, each unit calculates the intermediate aggregation result locally; and using the intermediate aggregation result, reconstruct the global average quota for this round of iteration.

Images