A method, device, electronic equipment, and storage medium for coordinated control of carbon emissions from industrial microgrids.

By generating a comprehensive feature sequence and inputting it into a carbon emission prediction model, and combining the carbon emission control threshold and the benchmark carbon emission factor to calculate the target regulation power, the problem of inaccurate coordinated control of carbon emissions in industrial microgrids is solved, and coordinated control and accuracy of equipment are improved.

CN123073944APending Publication Date: 2026-11-10JIANGMEN POWER SUPPLY BUREAU OF GUANGDONG POWER GRID CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611253478.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-18
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing industrial microgrid control schemes lack a calculation and allocation mechanism between carbon emissions and equipment regulation power, making it impossible to predict carbon emission trends in advance and determine carbon emission deviations based on control thresholds, resulting in inaccurate coordinated carbon emission control.

Method used

By acquiring operational data from conventional energy-consuming equipment, renewable energy output, and grid-type energy storage equipment, a comprehensive feature sequence is generated. This sequence is then input into a carbon emission prediction model to generate a prediction sequence. The target adjustment power is calculated by combining carbon emission control thresholds and benchmark carbon emission factors, and power allocation is performed to control the equipment.

Benefits of technology

It enables coordinated control of conventional energy-consuming equipment and grid-type energy storage equipment, improves the accuracy and coordination of carbon emissions from industrial microgrids, and solves the problem of isolated equipment regulation in existing technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123073944A_ABST
    Figure CN123073944A_ABST
Patent Text Reader

Abstract

This invention discloses a method, device, electronic equipment, and storage medium for the coordinated control of carbon emissions in industrial microgrids, belonging to the field of microgrid control technology. The method includes: acquiring the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in the industrial microgrid; constructing a comprehensive feature sequence based on the above data and inputting it into a carbon emission prediction model to generate a carbon emission prediction sequence; determining the carbon emission deviation based on the prediction sequence and the carbon emission control threshold, and calculating the target regulation power in conjunction with a benchmark carbon emission factor; and determining the energy storage regulation power and load regulation power through a power allocation strategy, thereby controlling the conventional energy-consuming equipment and the grid-type energy storage equipment. By implementing this invention, the problem of inaccurate coordinated control of carbon emissions in industrial microgrids in existing technologies can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of microgrid control technology, specifically to a method, device, electronic equipment, and storage medium for the coordinated control of carbon emissions from an industrial microgrid. Background Technology

[0002] With increasing industrial demands for green production, the low-carbon operation of industrial microgrids has become a crucial aspect of energy conservation and emission reduction. Industrial microgrids typically include conventional energy-consuming equipment, renewable energy generation equipment, and grid-connected energy storage devices. Coordinated control of these devices during microgrid operation is not only essential for system safety and stability but also a primary means for enterprises to meet carbon emission control requirements. Therefore, establishing a coordinated carbon emission control mechanism for industrial microgrids has significant practical application value.

[0003] However, current industrial microgrid control schemes typically focus on balancing power supply and demand. When controlling equipment, they often only reduce the power of conventional energy-consuming devices based on real-time power shortages, while allowing grid-based energy storage devices to operate independently according to fixed charging and discharging rules. This makes it difficult to directly coordinate the control of equipment based on carbon emission control requirements. The reason for this problem is that existing technologies lack a calculation and allocation mechanism between carbon emissions and equipment regulation power. They cannot predict carbon emission trends in advance and determine carbon emission deviations based on control thresholds. Consequently, they cannot convert carbon emission deviations into specific power allocation instructions for grid-based energy storage devices and conventional energy-consuming devices. As a result, industrial microgrids can only perform isolated power regulation on each device in actual operation, failing to effectively coordinate the control of conventional energy-consuming devices and grid-based energy storage devices, leading to inaccurate coordinated control of carbon emissions in industrial microgrids. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for the coordinated control of carbon emissions from industrial microgrids, which can solve the problem of insufficient accuracy in the coordinated control of carbon emissions from industrial microgrids in the prior art.

[0005] One embodiment of the present invention provides a method for coordinated control of carbon emissions from industrial microgrids, comprising: Obtain the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in industrial microgrids; Based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, a comprehensive characteristic sequence is determined. The comprehensive feature sequence is input into a preset carbon emission prediction model so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence. Based on the carbon emission forecast sequence and the preset carbon emission control threshold, determine the carbon emission deviation; based on the carbon emission deviation and the preset benchmark carbon emission factor, determine the target adjustment power. Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculation is performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment. Based on the energy storage regulation power of grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, control is applied to both conventional energy-consuming equipment and grid-type energy storage equipment.

[0006] Furthermore, based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, a comprehensive characteristic sequence is determined, including: The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

[0007] Furthermore, the pre-defined carbon emission prediction model is trained in the following manner: Acquire several training samples; the preset carbon emission prediction model is trained from these training samples; each training sample includes a historical comprehensive feature sequence and a carbon emission label; the carbon emission label is the historical carbon emission sequence corresponding to the historical comprehensive feature sequence. Each training sample is sequentially input into the carbon emission prediction model to train it until a preset number of training iterations are reached. The carbon emission prediction model outputs a carbon emission prediction sequence corresponding to each training sample it receives. A loss function value is calculated based on the carbon emission prediction sequence and its corresponding carbon emission label. The carbon emission prediction model is then updated based on the loss function value.

[0008] Furthermore, the carbon emission prediction sequence includes carbon emission prediction values ​​for each time period; Based on the carbon emission forecast sequence and the preset carbon emission control threshold, the carbon emission deviation is determined, including: The carbon emission forecast values ​​for each period in the carbon emission forecast sequence are summed to generate the total cumulative carbon emission forecast. The carbon emission deviation is determined based on the cumulative predicted total carbon emissions and the preset carbon emission control threshold.

[0009] Furthermore, based on the carbon emission deviation and the preset baseline carbon emission factor, the target regulation power is determined, including: Calculate the quotient between the carbon emission deviation and the preset benchmark carbon emission factor to generate the target regulation power. The target adjustment power is determined based on the target adjustment power and the preset scheduling cycle duration.

[0010] Furthermore, the energy storage operating condition sequence of grid-type energy storage equipment includes: the state of charge, maximum discharge limit power, and maximum charging limit power of the grid-type energy storage equipment; Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculations are performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, including: Boundary calculations are performed based on the state of charge, maximum discharge limit power, and maximum charge limit power to determine the upper and lower boundaries of power regulation for grid-type energy storage devices. The minimum value between the target regulation power and the upper boundary of power regulation is taken as the initial energy storage regulation power; The maximum value between the initial energy storage regulation power and the lower boundary of power regulation is determined as the energy storage regulation power of the grid-type energy storage device; Calculate the difference between the target regulation power and the energy storage regulation power of the grid-type energy storage device to generate the load regulation power of conventional energy-consuming equipment.

[0011] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0012] An embodiment of the present invention provides a collaborative control device for carbon emissions from an industrial microgrid, comprising: a data acquisition module, a feature sequence determination module, a carbon emission prediction module, a target power determination module, a power allocation module, and a collaborative control module; The data acquisition module is used to acquire the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in the industrial microgrid. The feature sequence determination module is used to determine a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment. The carbon emission prediction module is used to input the comprehensive feature sequence into the preset carbon emission prediction model, so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence. The target power determination module is used to determine the carbon emission deviation based on the carbon emission prediction sequence and the preset carbon emission control threshold; and to determine the target adjustment power based on the carbon emission deviation and the preset benchmark carbon emission factor. The power allocation module is used to perform power allocation calculations based on the target adjustment power, the energy storage operating condition sequence of the grid-type energy storage device, and the preset power allocation strategy, to determine the energy storage adjustment power of the grid-type energy storage device and the load adjustment power of the conventional energy-consuming device. The collaborative control module is used to control conventional energy-consuming equipment and grid-type energy storage equipment based on the energy storage regulation power of grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment.

[0013] Furthermore, the feature sequence determination module determines a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, including; The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

[0014] Based on the above method embodiments, the present invention provides corresponding electronic device embodiments.

[0015] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the coordinated control method for carbon emissions of industrial microgrids according to any one of the above-described method embodiments.

[0016] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments.

[0017] One embodiment of the present invention provides a storage medium storing a computer program thereon, wherein, when the computer program is running, it controls the device where the storage medium is located to execute any of the above-described method embodiments of the collaborative control method for carbon emissions of industrial microgrids.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method, apparatus, electronic device, and storage medium for the coordinated control of carbon emissions from industrial microgrids. The method acquires the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage condition sequence of grid-type energy storage equipment in the industrial microgrid; constructs a comprehensive feature sequence based on the above data and inputs it into a carbon emission prediction model to generate a carbon emission prediction sequence; determines the carbon emission deviation based on the prediction sequence and the carbon emission control threshold, and calculates the target regulation power in conjunction with a benchmark carbon emission factor; and determines the energy storage regulation power and load regulation power through a power allocation strategy, thereby controlling the conventional energy-consuming equipment and the grid-type energy storage equipment.

[0019] This invention generates a comprehensive feature sequence by acquiring the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage condition sequence of grid-type energy storage equipment. This sequence is then input into a preset carbon emission prediction model to generate a prediction sequence to determine the carbon emission deviation. This solves the problem that existing technologies cannot predict carbon emission trends in advance and determine the carbon emission deviation based on control thresholds. Furthermore, this invention determines the target regulation power based on the carbon emission deviation and the benchmark carbon emission factor, and determines the regulation power of grid-type energy storage equipment and conventional energy-consuming equipment through power allocation calculations. This solves the problem that existing technologies lack a calculation and allocation mechanism between carbon emissions and equipment regulation power, and cannot convert the deviation into specific power allocation instructions. Controlling conventional energy-consuming equipment and grid-type energy storage equipment based on the determined regulation power solves the problem that existing industrial microgrids can only regulate each device in isolation, improving the accuracy of coordinated carbon emission control in industrial microgrids. Attached Figure Description

[0020] Figure 1 This is a schematic flowchart of a collaborative control method for carbon emissions from an industrial microgrid, provided in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of a collaborative control device for carbon emissions from an industrial microgrid, provided in an embodiment of the present invention. Detailed Implementation

[0022] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, to address the problem of inaccurate coordinated control of carbon emissions from industrial microgrids in existing technologies, an embodiment of the present invention provides a method for coordinated control of carbon emissions from industrial microgrids, comprising at least the following steps: Step S1: Obtain the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in the industrial microgrid.

[0024] Specifically, the operating parameters of different types of equipment within the industrial microgrid are obtained during historical sampling periods. The industrial microgrid is equipped with conventional energy-consuming equipment to maintain routine production operations. This conventional energy-consuming equipment includes industrial manufacturing machine tools for material processing, production drive motors to maintain assembly line operation, factory lighting facilities to ensure nighttime operations, and environmental temperature control equipment to regulate workshop temperature and humidity. The total electrical power consumed by these conventional energy-consuming equipment is collected in real time at each consecutive sampling moment to construct a conventional power consumption sequence characterizing the fluctuations in load power consumption. The calculation formula for the conventional power consumption sequence is set as follows: In the formula, This represents the typical power consumption sequence. This indicates the total number of sampling points included within the historical sampling period. Indicates the first One sampling point, This indicates that conventional energy-consuming equipment is in the first... Total power consumption of each sampling point.

[0025] The industrial microgrid simultaneously deploys distributed photovoltaic (PV) power generation equipment and wind power generation equipment as new energy generators. By continuously monitoring the sum of the actual power output of the distributed PV and wind power generation equipment through measurement endpoints, a new energy output sequence reflecting fluctuations in green energy is generated. The specific mathematical expression for the new energy output sequence is: In the formula, Indicates the power output sequence of new energy sources. This indicates that distributed photovoltaic power generation equipment and wind power generation equipment are in the first The sum of the actual power generation of each sampling point.

[0026] For grid-based energy storage devices that provide power support within industrial microgrids, comprehensive data on their underlying operation is obtained. Grid-based energy storage devices are electrochemical energy storage physical entities capable of independently constructing grid voltage and frequency. The state of charge (SOC), maximum discharge limit power reflecting the physical discharge limit of the device's inverter topology, and maximum charge limit power reflecting the physical charging limit of the internal battery entities of the grid-based energy storage devices are extracted. The SOC, maximum discharge limit power, and maximum charge limit power are combined and integrated in chronological order to obtain the energy storage operating condition sequence. The mathematical matrix expression of the energy storage operating condition sequence is as follows: In the formula, Indicates the energy storage operating condition sequence. This indicates that grid-type energy storage devices are in the first The comprehensive working condition feature vector of each sampling point This indicates that grid-type energy storage devices are in the first The state of charge at each sampling point This indicates that grid-type energy storage devices are in the first Maximum discharge limit power at each sampling point This indicates that grid-type energy storage devices are in the first Maximum charging limit power for each sampling point.

[0027] The operation of acquiring conventional power consumption sequences, renewable energy output sequences, and energy storage operating condition sequences objectively and completely quantifies the actual operating conditions inside the industrial microgrid from three dimensions: energy consumption characteristics, green power generation characteristics, and energy storage physical operating boundary characteristics. This provides multi-source and reliable data support for subsequent accurate prediction of future carbon emission evolution trends and scientific formulation of coordinated power allocation instructions.

[0028] Step S2: Determine the comprehensive characteristic sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment.

[0029] In a preferred embodiment, a comprehensive characteristic sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, including: The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

[0030] Specifically, based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, a comprehensive characteristic sequence is determined, including the following execution steps.

[0031] A baseline carbon emission factor is pre-defined to measure the carbon emission equivalent generated by a unit of conventional electricity consumption. The conventional power consumption sequence of conventional energy-consuming equipment is read. The total power consumption at each sampling point in the conventional power consumption sequence is multiplied by the pre-defined baseline carbon emission factor to calculate the baseline carbon emission for the corresponding sampling point. A baseline carbon emission sequence is constructed based on the baseline carbon emission for all sampling points. The specific calculation formula for the baseline carbon emission sequence is set as follows: In the formula, Indicates basic carbon emissions. This represents the preset baseline carbon emission factor. This represents the baseline carbon emission sequence. The preset baseline carbon emission factor and the preset green electricity conversion factor mentioned below can be determined by querying the annual regional power grid baseline emission factor regularly published by the local power grid company or the ecological environment department, or by statistical calculation based on historical carbon inventory data of industrial microgrids.

[0032] The renewable energy output sequence of the industrial microgrid is extracted, and a green electricity conversion factor is pre-defined, measuring the equivalent reduction in carbon emissions per unit of renewable energy power generation. The sum of the actual power generation at each sampling point in the renewable energy output sequence is multiplied by the pre-defined green electricity conversion factor to obtain the equivalent emission reduction for that sampling point. The equivalent emission reductions at each sampling point are then arranged chronologically to determine the equivalent emission reduction sequence. The mathematical expression for the equivalent emission reduction sequence is: In the formula, This represents the equivalent emission reduction. This represents the preset green electricity conversion factor. This represents the equivalent emission reduction sequence.

[0033] Extract the energy storage operating condition sequence of grid-type energy storage devices. For the same sampling point, concatenate the dimensions of the base carbon emissions, equivalent emission reductions, and comprehensive operating condition feature vectors to form a multi-dimensional feature vector. Following the temporal evolution order, integrate the multi-dimensional feature vectors corresponding to all sampling points to generate a multi-dimensional feature matrix. The construction logic of the multi-dimensional feature matrix is ​​as follows: In the formula, Represents a multidimensional feature vector. This represents the multidimensional feature matrix generated after the splicing process.

[0034] Since the multidimensional feature matrix contains data with different physical dimensions, to eliminate the interference of these dimensional differences on the subsequent feature extraction by the prediction model, normalization is performed on each column of feature data within the multidimensional feature matrix. The minimum-maximum normalization algorithm is used to map the feature values ​​in the multidimensional feature matrix to a unified numerical range, generating a comprehensive feature sequence. The normalization calculation formula is as follows: In the formula, Indicates the feature dimension index number. This represents the feature values ​​after normalization. This represents the original values ​​in the multidimensional feature matrix. This represents the minimum value in the corresponding feature dimension column of the multidimensional feature matrix. This represents the maximum value in the corresponding feature dimension column of the multidimensional feature matrix. This represents the comprehensive feature sequence generated after normalization.

[0035] The operation of determining the comprehensive feature sequence involves converting physical-level electrical power data into carbon emission dimension feature data and combining it with the operating conditions of energy storage devices to complete dimensionless feature splicing. This effectively eliminates the physical dimension barriers between multi-source heterogeneous data, providing a data input foundation with unified dimensions and physical meaning correlation for subsequent carbon emission prediction models.

[0036] It should be noted that the underlying operational data acquired from the field measurement endpoints of industrial microgrids often contains communication noise interference or faces data gaps. To ensure the accuracy of input features, data cleaning and preprocessing operations are performed on the collected raw operational data during the construction of conventional power consumption sequences, renewable energy output sequences, and energy storage operating condition sequences. For sampling points with missing data due to communication delays or sensor malfunctions, the average power consumption or power generation values ​​from adjacent historical time periods are extracted to fill in the data, generating a complete and continuous time sequence. For anomalous data points caused by strong electromagnetic interference, the isolated forest algorithm is used to identify and remove outliers, and smoothed effective values ​​are used for replacement.

[0037] Step S3: Input the comprehensive feature sequence into the preset carbon emission prediction model so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence.

[0038] In a preferred embodiment, a pre-defined carbon emission prediction model is trained in the following manner: Acquire several training samples; the preset carbon emission prediction model is trained from these training samples; each training sample includes a historical comprehensive feature sequence and a carbon emission label; the carbon emission label is the historical carbon emission sequence corresponding to the historical comprehensive feature sequence. Each training sample is sequentially input into the carbon emission prediction model to train it until a preset number of training iterations are reached. The carbon emission prediction model outputs a carbon emission prediction sequence corresponding to each training sample it receives. A loss function value is calculated based on the carbon emission prediction sequence and its corresponding carbon emission label. The carbon emission prediction model is then updated based on the loss function value.

[0039] Specifically, the preset carbon emission prediction model is a neural network architecture built based on deep learning algorithms. The normalized output comprehensive feature sequence is fed into the preset carbon emission prediction model as input feature data. The preset carbon emission prediction model performs nonlinear mapping and temporal feature extraction on the multi-dimensional operational features contained in the comprehensive feature sequence, outputting a set of predicted carbon emission values ​​for a specific future time period. The set of carbon emission prediction values ​​is arranged chronologically to form a carbon emission prediction sequence. The generation logic of the carbon emission prediction sequence is expressed as follows: In the formula, Represents the carbon emission prediction sequence. This represents the forward propagation function of the pre-defined carbon emission prediction model. Represents a comprehensive feature sequence. This represents the optimal network weight matrix after the preset carbon emission prediction model has reached a convergent state following training.

[0040] In one specific embodiment, a pre-defined carbon emission prediction model is trained using a set method. Low-level operational data of the industrial microgrid during its historical operation phases are acquired, and multiple sets of historical comprehensive feature sequences are generated following the same processing steps as constructing comprehensive feature sequences. Simultaneously, real historical carbon emission value sequences corresponding to the historical operation phases are collected and used as carbon emission labels. Each set of historical comprehensive feature sequences and its corresponding carbon emission label is combined into a training sample. All training samples are aggregated to construct a training sample set containing several training samples. The mathematical expression for the training sample set is: In the formula, Represents the training sample set, This represents the total number of training samples contained in the training sample set. Indicates the first training samples, Indicates the first Historical comprehensive feature sequences in training samples Indicates the first The carbon emission labels corresponding to each training sample.

[0041] Each training sample in the training sample set is sequentially input into the network layer of a pre-defined carbon emission prediction model to train the model. Upon receiving a historical comprehensive feature sequence from each training sample, the pre-defined carbon emission prediction model performs forward computation using the network weight matrix of the current period, outputting a sample carbon emission prediction sequence corresponding to the received training sample. The formula for calculating the sample carbon emission prediction sequence is as follows: In the formula, This indicates that the preset carbon emission prediction model is for the first... The sample carbon emission prediction sequence output from each training sample. This represents the network weight matrix of the preset carbon emission prediction model in the current training period.

[0042] After obtaining the sample carbon emission prediction sequence, the carbon emission labels within the corresponding training samples are extracted. The mean squared error algorithm is selected to measure the numerical difference between the sample carbon emission prediction sequence and the carbon emission labels, and the loss function value for the current training period is calculated. The calculation logic for the loss function value is as follows: In the formula, This represents the calculated loss function value. This represents the total number of prediction time points contained within the sample carbon emission prediction sequence. Indicates the first Predicted time points Indicating the first carbon emission prediction sequence in the sample The value at each predicted time point. This indicates the corresponding number in the carbon emission label. The values ​​for each predicted time point.

[0043] Based on the calculated loss function value, the gradient descent algorithm is used to update the network weight matrix within the pre-defined carbon emission prediction model. The update formula for the network weight matrix is ​​set as follows: In the formula, This represents the updated network weight matrix for the next training cycle. This represents the preset learning rate parameter. This represents the gradient of the partial derivative of the loss function value with respect to the current network weight matrix. After one weight update, the system continues to read the next training sample for iterative training until a preset training iteration limit is reached, at which point the update operation stops. The preset learning rate parameter can be set as a conventional hyperparameter (e.g., 0.01 or 0.001), or dynamically set using an adaptive learning rate decay strategy. After one weight update, the system continues to read the next training sample for iterative training until a preset training iteration limit is reached, at which point the update operation stops.

[0044] By inputting the comprehensive feature sequence into the preset carbon emission prediction model and performing standardized model training, the carbon emission prediction model is able to mine the deep physical mapping relationship between multi-source heterogeneous data of microgrids and the evolution law of carbon emissions, thus ensuring the accuracy and reliability of the subsequent output carbon emission prediction sequence.

[0045] Step S4: Determine the carbon emission deviation based on the carbon emission prediction sequence and the preset carbon emission control threshold; determine the target adjustment power based on the carbon emission deviation and the preset benchmark carbon emission factor.

[0046] In a preferred embodiment, the carbon emission prediction sequence includes carbon emission prediction values ​​for each time period; Based on the carbon emission forecast sequence and the preset carbon emission control threshold, the carbon emission deviation is determined, including: The carbon emission forecast values ​​for each period in the carbon emission forecast sequence are summed to generate the total cumulative carbon emission forecast. The carbon emission deviation is determined based on the cumulative predicted total carbon emissions and the preset carbon emission control threshold.

[0047] In a preferred embodiment, determining the target adjustment power based on the carbon emission deviation and a preset benchmark carbon emission factor includes: Calculate the quotient between the carbon emission deviation and the preset benchmark carbon emission factor to generate the target regulation power. The target adjustment power is determined based on the target adjustment power and the preset scheduling cycle duration.

[0048] In an optional embodiment, the carbon emission prediction sequence includes predicted carbon emission values ​​for each time period. Based on the carbon emission prediction sequence and a preset carbon emission control threshold, the carbon emission deviation is determined, including the following steps: Extract the predicted carbon emission values ​​for each time period from the carbon emission prediction sequence, and sum all the predicted carbon emission values ​​to generate a cumulative total carbon emission prediction covering the entire prediction period. The formula for calculating the cumulative total carbon emission prediction is set as follows: In the formula, This represents the cumulative predicted total carbon emissions. This indicates the total number of time periods included in the carbon emission forecast series. Indicates the first Each period, Indicates the first Carbon emission forecasts for each time period.

[0049] Obtain the pre-set carbon emission control threshold. The carbon emission control threshold is the maximum allowable carbon equivalent emission quota issued by the relevant management department for industrial microgrids. The cumulative predicted total carbon emissions are subtracted from the carbon emission control threshold to determine the carbon emission deviation. The mathematical expression for the carbon emission deviation is: In the formula, Indicates the carbon emission deviation. This indicates the preset carbon emission control threshold.

[0050] In an optional embodiment, the target regulation power is determined based on the carbon emission deviation and a preset benchmark carbon emission factor, including the following calculation steps: The calculated carbon emission deviation and the stored benchmark carbon emission factor are read, and the quotient of the carbon emission deviation and the preset benchmark carbon emission factor is calculated. Using the quotient calculation, the deviation value in the carbon emission dimension can be equivalently converted into a regulation demand in the physical electrical energy dimension, generating the target regulation power. The conversion formula for the target regulation power is as follows: In the formula, This indicates the target power adjustment.

[0051] The preset dispatch cycle length of the industrial microgrid is retrieved. The preset dispatch cycle length refers to the fixed time span during which a control command is issued and the underlying physical equipment maintains stable operation. Based on the target regulation power and the preset dispatch cycle length, the ratio of the target regulation power to the preset dispatch cycle length is calculated to determine the target regulation power that can be directly used to guide the operation of the physical equipment. The formula for calculating the target regulation power is: In the formula, Indicates the target adjustment power. This indicates the preset scheduling cycle duration. The preset scheduling cycle duration refers to the fixed time span during which a control command is issued and the underlying physical equipment is kept running stably. Specifically, it can be set to a fixed physical duration such as 15 minutes, 30 minutes, or 1 hour, depending on the actual control accuracy requirements of the industrial microgrid.

[0052] Based on the predicted sequence and control threshold, a series of operations are performed to determine the deviation and convert it into the target regulation power. This constructs a direct mathematical conversion path from abstract carbon emission control indicators to the regulation power of concrete physical devices. This overcomes the limitation of previous microgrid control processes that only focused on power supply and demand fluctuations. It provides a clear and quantitative power allocation benchmark for achieving precise and coordinated regulation of underlying heterogeneous energy consumption and energy storage devices.

[0053] Step S5: Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, perform power allocation calculation to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of the conventional energy-consuming equipment.

[0054] In a preferred embodiment, the energy storage operating condition sequence of the grid-type energy storage device includes: the state of charge of the grid-type energy storage device, the maximum discharge limit power, and the maximum charging limit power. Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculations are performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, including: Boundary calculations are performed based on the state of charge, maximum discharge limit power, and maximum charge limit power to determine the upper and lower boundaries of power regulation for grid-type energy storage devices. The minimum value between the target regulation power and the upper boundary of power regulation is taken as the initial energy storage regulation power; The maximum value between the initial energy storage regulation power and the lower boundary of power regulation is determined as the energy storage regulation power of the grid-type energy storage device; Calculate the difference between the target regulation power and the energy storage regulation power of the grid-type energy storage device to generate the load regulation power of conventional energy-consuming equipment.

[0055] In an optional embodiment, the energy storage operating condition sequence of the grid-type energy storage device includes: the state of charge of the grid-type energy storage device, the maximum discharge limit power, and the maximum charge limit power. Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage device, and a preset power allocation strategy, power allocation calculations are performed to determine the energy storage regulation power of the grid-type energy storage device and the load regulation power of conventional energy-consuming equipment, specifically covering the following actions.

[0056] After obtaining the target regulation power, for the preset scheduling period to be executed, the state of charge, maximum discharge limit power, and maximum charging limit power of the grid-type energy storage device at the start of the preset scheduling period are extracted. Combined with the rated energy capacity of the grid-type energy storage device, boundary calculations are performed based on the state of charge, maximum discharge limit power, and maximum charging limit power to determine the upper and lower boundaries of the power regulation for the grid-type energy storage device. The upper boundary of power regulation represents the maximum forward discharge power allowed to be output by the grid-type energy storage device in the current physical state and the corresponding scheduling period. The lower boundary of power regulation represents the maximum reverse charging power allowed to be withstood by the grid-type energy storage device in the current physical state and the corresponding scheduling period. The mathematical expression for the boundary calculation is set as follows: In the formula, Indicates the upper limit of power regulation. Indicates the lower limit of power regulation. This indicates the real-time state of charge of the grid-connected energy storage device at the start of a preset scheduling cycle. This indicates the maximum discharge limit power of the grid-type energy storage device at the start of a preset dispatch cycle. This indicates the maximum charging limit power of the grid-type energy storage device at the start of a preset scheduling cycle. This indicates the preset safety threshold for the state of charge. This indicates the preset upper limit safety threshold for the state of charge. This indicates the rated electrical capacity of the grid-type energy storage device. This represents the function that takes the minimum value. This represents the function that takes the maximum value.

[0057] It should be noted that the preset safety threshold for the state of charge is... and the preset upper limit safety threshold for the state of charge. These are crucial physical boundary parameters for ensuring the electrochemical lifespan of batteries in grid-connected energy storage devices and the physical safety of the system. Specifically, these safety thresholds can be directly obtained from the factory-set configuration table of the Battery Management System (BMS) within the grid-connected energy storage device; alternatively, they can be dynamically calculated and preset by the industrial microgrid's operation and maintenance terminal based on the battery's current State of Health (SOH) and ambient temperature. For example, in typical industrial microgrid applications, to prevent irreversible electrochemical over-discharge damage to the battery pack and to reserve backup physical capacity for emergency power support in the microgrid, a preset state-of-charge (SOH) limit safety threshold can be used. Set a fixed value within the range of 10% to 15% (e.g., 10%); at the same time, to prevent the risk of thermal runaway caused by overcharging, a preset upper limit safety threshold for the state of charge can be set. The value is set to a fixed value within the range of 90% to 95% (e.g., 90%). By introducing the aforementioned underlying safety boundary parameters for amplitude limiting calculation, it is ensured that after the cooperative control command is issued, the operating trajectory of the underlying physical hardware is strictly limited within the electrochemical safety range.

[0058] After determining the upper and lower boundaries of power regulation, grid-based energy storage devices are prioritized for carbon emission power regulation according to a preset power allocation strategy. The minimum value between the target regulation power and the upper boundary of power regulation is used as the initial energy storage regulation power. The portion of the target regulation power exceeding the upper boundary of power regulation is truncated to ensure that the initially allocated power does not exceed the discharge limit of the energy storage hardware. The formula for calculating the initial energy storage regulation power is as follows: In the formula, This indicates the initial energy storage regulation power.

[0059] Subsequently, the maximum value between the initial energy storage regulation power and the lower boundary of power regulation is determined as the energy storage regulation power of the grid-type energy storage device. By calculating the maximum value, values ​​below the lower boundary of power regulation are intercepted at the lower boundary, thus ensuring that the final power command issued to the grid-type energy storage device will never exceed the physical charging safety boundary. The formula for calculating the energy storage regulation power is expressed as: In the formula, This indicates the energy storage regulation power.

[0060] After power allocation for grid-type energy storage devices is completed, the difference between the target regulation power and the energy storage regulation power of the grid-type energy storage devices is calculated to generate the load regulation power of conventional energy-consuming devices. Conventional energy-consuming devices serve as a supplementary regulation means to handle the remaining regulation power that the grid-type energy storage devices cannot absorb. The calculation logic for load regulation power is expressed as follows: In the formula, This indicates the load regulation power.

[0061] By comparing and calculating the difference between the target regulation power and the energy storage operating condition sequence, a control logic was established that prioritizes the use of energy storage devices to buffer carbon emission fluctuations and allows conventional energy-consuming devices to take over the remaining power. This not only maximizes the flexible response potential of energy storage devices as passive regulation nodes, but also minimizes the power outage interference caused by carbon reduction scheduling actions to continuous production operations within the industrial microgrid.

[0062] It should be noted that, considering the fluctuation direction of carbon emission deviation, the target regulation power is divided into positive and negative numerical states. A positive numerical state indicates that the industrial microgrid has a carbon emission deficit exceeding the control targets, requiring external power output or reduction of internal power consumption to lower overall carbon emissions. A negative numerical state indicates that the industrial microgrid has surplus carbon emission allowances, possessing the physical space to absorb external green energy or increase internal power consumption. The upper boundary of power regulation set for grid-based energy storage devices defines the absolute physical upper limit value for positive discharge operations, while the lower boundary of power regulation establishes the absolute physical lower limit value for negative charging operations with a negative sign.

[0063] For different directional states of the target regulating power, the power allocation calculation fully executes the limiting and truncation actions of continuously extracting the smaller value and extracting the larger value. The target regulating power is compared with the upper boundary of power regulation, and the smaller value is extracted to generate the initial energy storage regulating power. The operation of extracting the smaller value constitutes the first stage of positive limiting logic. When the target regulating power is a positive value greater than the upper boundary of power regulation, it indicates that the discharge demand has exceeded the hardware's capacity. The mathematical operation of directly extracting the smaller value can forcibly suppress the discharge power command allocated to the grid-type energy storage device downwards and lock it at the position of the upper boundary of power regulation, preventing the grid-type energy storage device from exceeding the physical discharge limit and triggering over-discharge protection. If the target regulating power is a negative value or a positive value less than the upper boundary of power regulation, the operation of extracting the smaller value will completely preserve the original numerical state of the target regulating power, ensuring that it does not interfere with normal negative charging or slight positive discharging demand.

[0064] After obtaining the initial energy storage regulation power, the initial energy storage regulation power is compared with the lower boundary of power regulation, and the larger value is extracted to determine the final energy storage regulation power issued to the grid-type energy storage equipment. The calculation of extracting the larger value constitutes the negative limiting logic of the second stage. Since both the lower boundary of power regulation indicating charging operation and the negative regulation demand are quantified using values ​​with negative signs, the smaller the value with a negative sign, the greater the actual physical charging power required. When the initial energy storage regulation power is a deeply deviated negative value and its absolute magnitude exceeds the absolute magnitude of the lower boundary of power regulation, the mathematical operation of directly extracting the larger value can raise the excessive negative value and accurately intercept it to the lower boundary of power regulation. After the above two consecutive numerical comparisons and truncation operations in a purely mathematical dimension, the energy storage regulation power issued to the grid-type energy storage equipment is strictly limited within the safe physical operating range jointly constructed by the upper boundary and the lower boundary of power regulation.

[0065] After completing the bidirectional limiting allocation of the grid-type energy storage equipment, the difference between the target regulating power and the energy storage regulating power of the grid-type energy storage equipment is calculated to generate the load regulating power of the conventional energy-consuming equipment, thereby achieving coordinated regulation under different over-limit conditions. When the target regulating power does not exceed the physical limit of the grid-type energy storage equipment, the original values ​​are not truncated in the two extreme value calculations, the difference calculation result is zero, and the conventional energy-consuming equipment maintains its original production and operation load unchanged. When the target regulating power is a positive value and greater than the upper limit of power regulation, the grid-type energy storage equipment discharges at full load according to the upper limit of power regulation. The positive value difference generated by the difference calculation is converted into a power reduction control command for the conventional energy-consuming equipment, driving the conventional energy-consuming equipment to unload part of the load. When the target regulating power is a negative value and its absolute value is greater than the absolute value of the lower limit of power regulation, the grid-type energy storage equipment charges at full load according to the lower limit of power regulation. The negative value difference generated by the difference calculation is converted into a power increase control command for the conventional energy-consuming equipment, driving the conventional energy-consuming equipment to actively increase its electricity load to absorb the surplus indicators.

[0066] By executing bidirectional mathematical comparison logic to extract smaller and larger values, combined with difference allocation operations, the complex bidirectional charging and discharging physical safety boundary of the underlying hardware is directly transformed into a simplified extreme value determination step. This ensures that the grid-type energy storage equipment always operates within the allowable physical safety range and maintains the stable state of continuous production operations of conventional energy-consuming equipment to the greatest extent.

[0067] Step S6: Control the conventional energy-consuming equipment and the grid-type energy storage equipment according to the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of the conventional energy-consuming equipment.

[0068] Specifically, for grid-based energy storage devices, the energy storage regulation power is encapsulated into inverter active power control commands that can be recognized by the underlying physical hardware. These inverter active power control commands are then sent to the converter control module inside the grid-based energy storage device. The converter control module parses the inverter active power control commands and adjusts the conduction state of its internal power switching devices based on the parsing results, forcing the grid-based energy storage device to strictly perform forward discharge or reverse charging operations according to the set energy storage regulation power values. The underlying control command conversion logic of the grid-based energy storage device is expressed as follows: In the formula, This refers to the inverter active power control command generated for grid-type energy storage devices.

[0069] For conventional energy-consuming equipment, the initial operating baseline power of the equipment at the current scheduling moment is measured in real time. The initial operating baseline power represents the total normal physical electricity consumption of the conventional energy-consuming equipment before any carbon emission control intervention. Based on the calculation relationship between the initial operating baseline power and the load regulation power, the target total operating power that the conventional energy-consuming equipment should achieve within the scheduled period is calculated. The mathematical derivation formula for the target total operating power is as follows: In the formula, This indicates the target total operating power of conventional energy-consuming equipment. This represents the initial operating baseline power of conventional energy-consuming equipment at the current scheduling moment.

[0070] Based on the change in the target total operating power, corresponding physical activation / deactivation commands are generated and sent to the distribution control terminals of conventional energy-consuming equipment. Depending on the specific direction of the load regulation power, the distribution control terminals drive the conventional energy-consuming equipment to exhibit differentiated physical action responses. When the load regulation power is positive, the target total operating power will inevitably be less than the initial operating baseline power due to subtraction. The distribution control terminal actively disconnects non-core production lines associated with the conventional energy-consuming equipment to reduce the overall energy consumption of the microgrid. When the load regulation power is negative, the negative value participating in the subtraction operation results in a positive value, increasing the target total operating power to exceed the initial operating baseline power. The distribution control terminal activates standby production equipment in standby mode within the microgrid to increase overall energy consumption and fully utilize excess equivalent carbon emission allowances. When the load regulation power is zero, the target total operating power is completely equal to the initial operating baseline power. The distribution control terminal does not issue any change commands, maintaining the existing operating state of the conventional energy-consuming equipment unchanged.

[0071] By directly performing physical interventions on the bottom energy-consuming and energy-storage components based on energy storage regulation power and load regulation power, a closed-loop link is completely opened from the upper-level carbon emission prediction indicators to the end-point hardware power execution, ensuring that the microgrid can achieve safe and coordinated operation of multi-source heterogeneous devices under the condition of meeting the stringent carbon emission constraints.

[0072] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0073] like Figure 2 As shown, an embodiment of the present invention provides a collaborative control device for carbon emissions of an industrial microgrid, comprising: a data acquisition module, a feature sequence determination module, a carbon emission prediction module, a target power determination module, a power allocation module, and a collaborative control module; The data acquisition module is used to acquire the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in the industrial microgrid. The feature sequence determination module is used to determine a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment. The carbon emission prediction module is used to input the comprehensive feature sequence into the preset carbon emission prediction model, so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence. The target power determination module is used to determine the carbon emission deviation based on the carbon emission prediction sequence and the preset carbon emission control threshold; and to determine the target adjustment power based on the carbon emission deviation and the preset benchmark carbon emission factor. The power allocation module is used to perform power allocation calculations based on the target adjustment power, the energy storage operating condition sequence of the grid-type energy storage device, and the preset power allocation strategy, to determine the energy storage adjustment power of the grid-type energy storage device and the load adjustment power of the conventional energy-consuming device. The collaborative control module is used to control conventional energy-consuming equipment and grid-type energy storage equipment based on the energy storage regulation power of grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment.

[0074] In a preferred embodiment, the feature sequence determination module determines a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, including: The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

[0075] In an optional embodiment, the carbon emission prediction module trains a preset carbon emission prediction model in the following manner: Acquire several training samples; the preset carbon emission prediction model is trained from these training samples; each training sample includes a historical comprehensive feature sequence and a carbon emission label; the carbon emission label is the historical carbon emission sequence corresponding to the historical comprehensive feature sequence. Each training sample is sequentially input into the carbon emission prediction model to train it until a preset number of training iterations are reached. The carbon emission prediction model outputs a carbon emission prediction sequence corresponding to each training sample it receives. A loss function value is calculated based on the carbon emission prediction sequence and its corresponding carbon emission label. The carbon emission prediction model is then updated based on the loss function value.

[0076] In an optional embodiment, the target power determination module includes carbon emission prediction sequences that include predicted carbon emission values ​​for each time period. Based on the carbon emission forecast sequence and the preset carbon emission control threshold, the carbon emission deviation is determined, including: The carbon emission forecast values ​​for each period in the carbon emission forecast sequence are summed to generate the total cumulative carbon emission forecast. The carbon emission deviation is determined based on the cumulative predicted total carbon emissions and the preset carbon emission control threshold.

[0077] In an optional embodiment, the target power determination module determines the target adjustment power based on the carbon emission deviation and a preset benchmark carbon emission factor, including: Calculate the quotient between the carbon emission deviation and the preset benchmark carbon emission factor to generate the target regulation power. The target adjustment power is determined based on the target adjustment power and the preset scheduling cycle duration.

[0078] In an optional embodiment, the power distribution module includes the energy storage operating condition sequence of the grid-type energy storage device, which includes: the state of charge of the grid-type energy storage device, the maximum discharge limit power, and the maximum charging limit power. Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculations are performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, including: Boundary calculations are performed based on the state of charge, maximum discharge limit power, and maximum charge limit power to determine the upper and lower boundaries of power regulation for grid-type energy storage devices. The minimum value between the target regulation power and the upper boundary of power regulation is taken as the initial energy storage regulation power; The maximum value between the initial energy storage regulation power and the lower boundary of power regulation is determined as the energy storage regulation power of the grid-type energy storage device; Calculate the difference between the target regulation power and the energy storage regulation power of the grid-type energy storage device to generate the load regulation power of conventional energy-consuming equipment.

[0079] It should be noted that the embodiments of the device described above correspond to the embodiments of the present invention described above, and can realize the collaborative control method for carbon emissions of industrial microgrids described in any one of the above embodiments of the present invention. Furthermore, the embodiments of the device described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without creative effort.

[0080] Based on the above-described method embodiments of the present invention, a corresponding embodiment of an electronic device is provided.

[0081] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the collaborative control method for carbon emissions of industrial microgrids according to any one of the present invention, or, when the processor executes the computer program, it implements the functions of each module in the above-described device embodiments.

[0082] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0083] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0084] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0085] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0086] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments; Another embodiment of the present invention provides a storage medium including a stored computer program, wherein, when the computer program is executed, the device where the storage medium is located controls the execution of any of the above-described industrial microgrid carbon emission collaborative control methods of the present invention.

[0087] The aforementioned storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0088] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0089] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for coordinated control of carbon emissions from industrial microgrids, characterized in that, include: Obtain the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in industrial microgrids; Based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, a comprehensive characteristic sequence is determined. The comprehensive feature sequence is input into a preset carbon emission prediction model so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence. Based on the carbon emission forecast sequence and the preset carbon emission control threshold, determine the carbon emission deviation; based on the carbon emission deviation and the preset benchmark carbon emission factor, determine the target adjustment power. Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculation is performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment. Based on the energy storage regulation power of grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, control is applied to both conventional energy-consuming equipment and grid-type energy storage equipment.

2. The method for coordinated control of carbon emissions from industrial microgrids as described in claim 1, characterized in that, Based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment, a comprehensive characteristic sequence is determined, including: The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

3. The method for coordinated control of carbon emissions from industrial microgrids as described in claim 2, characterized in that, The pre-defined carbon emission prediction model is trained using the following methods: Acquire several training samples; the preset carbon emission prediction model is trained from these training samples; each training sample includes a historical comprehensive feature sequence and a carbon emission label; the carbon emission label is the historical carbon emission sequence corresponding to the historical comprehensive feature sequence. Each training sample is sequentially input into the carbon emission prediction model to train it until a preset number of training iterations are reached. The carbon emission prediction model outputs a carbon emission prediction sequence corresponding to each training sample it receives. A loss function value is calculated based on the carbon emission prediction sequence and its corresponding carbon emission label. The carbon emission prediction model is then updated based on the loss function value.

4. The method for coordinated control of carbon emissions from industrial microgrids as described in claim 3, characterized in that, The carbon emission prediction sequence includes carbon emission prediction values ​​for each time period; Based on the carbon emission forecast sequence and the preset carbon emission control threshold, the carbon emission deviation is determined, including: The carbon emission forecast values ​​for each period in the carbon emission forecast sequence are summed to generate the total cumulative carbon emission forecast. The carbon emission deviation is determined based on the cumulative predicted total carbon emissions and the preset carbon emission control threshold.

5. The method for coordinated control of carbon emissions from industrial microgrids as described in claim 4, characterized in that, Based on the carbon emission deviation and the preset baseline carbon emission factor, the target regulation power is determined, including: Calculate the quotient between the carbon emission deviation and the preset benchmark carbon emission factor to generate the target regulation power. The target adjustment power is determined based on the target adjustment power and the preset scheduling cycle duration.

6. The method for coordinated control of carbon emissions from industrial microgrids as described in claim 5, characterized in that, The energy storage operating condition sequence of grid-type energy storage equipment includes: the state of charge, maximum discharge limit power, and maximum charging limit power of the grid-type energy storage equipment; Based on the target regulation power, the energy storage operating condition sequence of the grid-type energy storage equipment, and the preset power allocation strategy, power allocation calculations are performed to determine the energy storage regulation power of the grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment, including: Boundary calculations are performed based on the state of charge, maximum discharge limit power, and maximum charge limit power to determine the upper and lower boundaries of power regulation for grid-type energy storage devices. The minimum value between the target regulation power and the upper boundary of power regulation is taken as the initial energy storage regulation power; The maximum value between the initial energy storage regulation power and the lower boundary of power regulation is determined as the energy storage regulation power of the grid-type energy storage device; Calculate the difference between the target regulation power and the energy storage regulation power of the grid-type energy storage device to generate the load regulation power of conventional energy-consuming equipment.

7. A collaborative control device for carbon emissions from an industrial microgrid, characterized in that, include: The system includes a data acquisition module, a feature sequence determination module, a carbon emission prediction module, a target power determination module, a power allocation module, and a collaborative control module. The data acquisition module is used to acquire the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment in the industrial microgrid. The feature sequence determination module is used to determine a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment. The carbon emission prediction module is used to input the comprehensive feature sequence into the preset carbon emission prediction model, so that the carbon emission prediction model generates a carbon emission prediction sequence based on the comprehensive feature sequence. The target power determination module is used to determine the carbon emission deviation based on the carbon emission prediction sequence and the preset carbon emission control threshold; and to determine the target adjustment power based on the carbon emission deviation and the preset benchmark carbon emission factor. The power allocation module is used to perform power allocation calculations based on the target adjustment power, the energy storage operating condition sequence of the grid-type energy storage device, and the preset power allocation strategy, to determine the energy storage adjustment power of the grid-type energy storage device and the load adjustment power of the conventional energy-consuming device. The collaborative control module is used to control conventional energy-consuming equipment and grid-type energy storage equipment based on the energy storage regulation power of grid-type energy storage equipment and the load regulation power of conventional energy-consuming equipment.

8. The collaborative control device for carbon emissions from industrial microgrids as described in claim 7, characterized in that, The feature sequence determination module determines a comprehensive feature sequence based on the conventional power consumption sequence of conventional energy-consuming equipment, the power output sequence of new energy sources, and the energy storage operating condition sequence of grid-type energy storage equipment. This includes: The basic carbon emission sequence is determined based on the conventional power consumption sequence of conventional energy-consuming equipment and the preset benchmark carbon emission factor; Based on the new energy output sequence and the preset green electricity conversion factor, the equivalent emission reduction sequence is determined; The basic carbon emission sequence, the equivalent emission reduction sequence, and the energy storage operating condition sequence of grid-type energy storage equipment are spliced ​​together to generate a multi-dimensional feature matrix. The multidimensional feature matrix is ​​normalized to generate a comprehensive feature sequence.

9. An electronic device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the coordinated control method for carbon emissions from an industrial microgrid as described in any one of claims 1 to 6.

10. A storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the storage medium to perform the collaborative control method for carbon emissions of industrial microgrids as described in any one of claims 1 to 6.