Power Equipment Green Certification Method Based on Carbon Emission Data Analysis

Abstract

The present application discloses a green certification method for power equipment based on carbon emission data analysis, which relates to the technical field of equipment green certification. The method includes constructing a carbon emission environmental impact relationship based on carbon emission data, local environmental data, and global environmental data under historical time series; obtaining the equipment combination probability according to the similarity of the life cycle data of each equipment under historical time series, and constructing an equipment carbon emission change model based on the equipment combination probability and its life cycle change factor; obtaining carbon emission dynamic constraint data based on the carbon emission environmental impact relationship according to the current time series; obtaining carbon emission prediction data based on the current equipment information according to the equipment carbon emission change model; and performing green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data. The present application adaptively fluctuates the carbon emission constraint according to the absorption of carbon emissions by the local environment and the global environment under the actual time series, and compensates and corrects the equipment carbon emission prediction data with the equipment combination probability, thereby improving the accuracy and adaptability of power equipment certification.

Description

Technical Field

[0001] This application relates to the technical field of equipment green certification, and particularly to a green certification method for power equipment based on carbon emission data analysis. Background Art

[0002] In the power industry, carbon emissions from power equipment are one of the important sources of greenhouse gas emissions. Therefore, green certification of power equipment to ensure that its carbon emissions meet environmental protection requirements is an important measure to promote the low-carbon development of the power industry.

[0003] Currently, green certification of power equipment mainly relies on traditional carbon emission measurement and assessment methods, usually estimated based on overall power consumption or energy consumption, with relatively coarse data granularity, making it difficult to accurately reflect the carbon emissions of different equipment and their emissions under different environmental and operating conditions.

[0004] In related technologies, power equipment is usually regarded as an independent entity, ignoring the mutual influence between equipment, resulting in double counting or omission of carbon emissions, leading to deviations in the certification of whether the carbon emissions of power equipment meet green emissions under complex equipment combinations and dynamic environmental conditions, and the certification accuracy is poor.

[0005] Patent "A Green Power - CCER Mutual Recognition Trading System Based on Dynamic Emission Reduction Factors", Publication Number: CN119444242A, Publication Date: February 14, 2025, specifically discloses a green power - CCER mutual recognition system, which consists of four modules: a data interaction module, a dynamic carbon emission reduction factor publicity module, a CCER certificate issuance module, and a transfer and cancellation module, and can systematically implement mutual recognition technology, making the green power - CCER mutual recognition transparent and executable. This solution publishes the dynamic carbon emission reduction factors of each user node in real time, but when users conduct load planning, they still based on fixed carbon emission constraints, with poor environmental adaptability.

[0006] Patent "A Linkage Mechanism for Green Power Consumption Certification and Carbon Emission Monitoring and Accounting", Publication Number: CN118036882A, Publication Date: May 14, 2024, specifically discloses determining the carbon emission accounting scope according to the logistics business activities of power grid enterprises; establishing a carbon - green certificate joint trading model considering conditional value at risk; establishing a total carbon emission accounting model for the logistics space of power equipment; the green certificate management agency uses blockchain to verify the sampled data in the database, and issues green certificates if the sampled data is true. Although this solution includes the total carbon emission accounting for the logistics space of power equipment, there are still fixed carbon emission constraints and it cannot adapt to the actual environmental carbon emission requirements. Summary of the Invention

[0007] In view of the technical problem that there are deviations in the certification of whether the carbon emissions of power equipment meet the green emissions under complex equipment combinations and dynamic environmental conditions in the prior art, this application provides a green certification method for power equipment based on carbon emission data analysis. By constructing the carbon emission environmental impact relationship according to the respective influence of local environmental data and global environmental data on carbon emission data in the historical time series, adaptively fluctuating the carbon emission constraint according to the carbon emission absorption of the local environment and the global environment in the actual time series, so that the carbon emission constraint adapts to the actual environment. At the same time, based on the equipment combination probability, the carbon emission prediction data of the equipment is compensated and corrected, and the equipment that may be combined is included in the consideration of the current equipment green certification, avoiding the problem of excessive carbon emissions caused by the subsequent addition of equipment, and improving the accuracy and adaptability of the power equipment green certification.

[0008] To achieve the above technical objectives, a technical solution provided by this application is a green certification method for power equipment based on carbon emission data analysis, including the following steps: constructing a carbon emission environmental impact relationship based on carbon emission data, local environmental data, and global environmental data in the historical time series; obtaining the equipment combination probability according to the similarity of the life cycle data of each device in the historical time series, and constructing a carbon emission change model of the device based on the equipment combination probability and its life cycle change factor; obtaining carbon emission dynamic constraint data based on the carbon emission environmental impact relationship according to the current time series; obtaining carbon emission prediction data based on the current device information according to the carbon emission change model of the device; performing green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data.

[0009] Further, constructing the carbon emission environmental impact relationship based on the carbon emission data, local environmental data, and global environmental data in the historical time series includes: obtaining the local influence relationship according to the superposition value of the carbon emission data and the carbon emission influence data of the local environment in the historical time series; obtaining the global influence relationship according to the superposition value of the carbon emission data and the carbon emission influence data of the global environment in the historical time series.

[0010] Further, constructing the carbon emission environmental impact relationship based on the carbon emission data, local environmental data, and global environmental data in the historical time series includes: obtaining local environmental data according to a preset area, and for each preset area, performing: obtaining carbon emission data that changes synchronously with the local environmental data; obtaining the corresponding device spatial position according to the carbon emission data, and constructing a spatial influence relationship; constructing a time influence relationship according to the local environmental parameters in the local environmental data and the superposition value of the carbon emission data; obtaining the local influence relationship with the spatial influence relationship and the time influence relationship.

[0011] Further, obtaining the corresponding device spatial position according to the carbon emission data and constructing a spatial influence relationship includes: obtaining carbon emission data with a fluctuation similarity to the local environmental data; obtaining the corresponding device spatial position of the carbon emission data; constructing a spatial influence relationship according to the device spatial position, the fluctuation value of the local environmental data, and the fluctuation value of the carbon emission data.

[0012] Furthermore, constructing the carbon emission environmental impact relationship based on carbon emission data, local environmental data, and global environmental data in historical time series includes: obtaining the vegetation impact coefficient according to the animal species ratio, biological abundance index, vegetation coverage rate, and vegetation carbon absorption rate; obtaining the carbon emission environmental impact relationship by combining the carbon emission data superposition value with the temperature, precipitation, wind speed, wind direction, humidity, and vegetation impact coefficient.

[0013] Furthermore, obtaining the equipment combination probability according to the similarity of the life cycle data of each device in historical time series includes: obtaining the transportation similarity of each device based on the transportation type, transportation time series, and transportation path of each device in historical time series; obtaining the operation similarity of each device based on the operation dependence relationship and operation time series of each device in historical time series; obtaining the scrapping similarity of each device based on the structure type, scrapping time series, and scrapping area of each device in historical time series; obtaining the equipment combination probability based on multi-dimensional scaling analysis combined with the equipment transportation similarity, equipment operation similarity, and equipment scrapping similarity.

[0014] Furthermore, constructing the equipment carbon emission change model based on the equipment combination probability and its life cycle change factors includes: constructing an equipment combination probability calculation layer based on the initial transportation weight, initial operation weight, and initial scrapping weight; constructing a factor matching layer for the equipment combination probability, initial probability threshold, and life cycle change factors; constructing an initial carbon emission change model with the equipment combination probability calculation layer and the factor matching layer; iteratively updating the initial transportation weight, initial operation weight, initial scrapping weight, and initial probability threshold in the initial carbon emission change model according to the carbon emission data of each device in historical time series to obtain the carbon emission change model.

[0015] Furthermore, constructing the equipment carbon emission change model based on the equipment combination probability and its life cycle change factors also includes: constructing a data compensation layer based on the merge unique strategy; constructing an initial equipment carbon emission change model with the neural network model architecture and the data compensation layer; training the initial equipment carbon emission change model with the processed carbon emission data to obtain the equipment carbon emission change model.

[0016] Furthermore, in the data compensation layer, execute: merging the carbon emission data according to the equipment combination probability, and outputting the merged equipment information according to the merging result of the carbon emission data; outputting the equipment carbon emission compensation value according to the merged equipment information and the equipment carbon emission calculation time series.

[0017] Further, the obtaining of carbon emission dynamic constraint data based on the carbon emission environment impact relationship according to the current time series includes: obtaining a local environment impact value based on the current time series, the time impact relationship, the current device spatial location, and the spatial impact relationship; obtaining a global environment impact value based on the current time series and the global impact relationship; obtaining local environment constraints and global environment constraints according to the current device spatial location; obtaining local carbon emission constraints with the local environment constraints and the local environment impact value, and obtaining global carbon emission constraints with the global environment constraints and the global environment impact value.

[0018] Further, the performing of the green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data includes: if the carbon emission prediction data meets the carbon emission dynamic constraint data, the green certification of the power equipment passes; if the carbon emission prediction data does not meet the carbon emission dynamic constraint data, the green certification of the power equipment fails.

[0019] Advantages of the present application: 1. Construct a carbon emission environment impact relationship based on the impact of carbon emissions on the local environment and the global environment under historical time series, and quantify the continuous impact of carbon emissions on the local environment and the global environment. At the same time, construct a device carbon emission change model for each device type according to the device combination probability, comprehensively account for and monitor the carbon emission data of the entire process of power equipment, and use the device combination probability to incorporate the carbon emissions generated by the possible synchronous operation of other devices with this device into the carbon emission prediction data of this device, incorporate the impact of device combination operation, consider the overall carbon emission status of the device, consider the carbon emission amount and energy-saving effect of a single individual from a macroscopic perspective, improve the accuracy of the certification effect, and avoid ignoring the global carbon emission impact. And output carbon emission dynamic constraint data as the certification standard based on the dynamic tolerance of the environment, improve the time series adaptability and spatial adaptability of the green certification, and improve the reliability of the green certification of power equipment.

[0020] 2. Through the local environment data and carbon emission data with similar fluctuations, obtain the device spatial locations that may affect the current local environment, and then construct a spatial impact relationship according to the correlation between the historical carbon emission data fluctuation values and the local environment data fluctuation values of each device spatial location, so as to show the impact of different devices on different local environments and avoid the single calculation of the device carbon emission impact.

[0021] 3. Obtain the impact of device carbon emissions on different local environments through the spatial impact relationship and the time impact relationship respectively, and obtain the impact of device carbon emissions on the overall environment according to the global time impact. This can not only avoid ignoring the impact on the remaining adjacent or relevant local environments when performing the green certification of power equipment, but also avoid double-counting the environmental impact during the overall carbon emission calculation, improve the pertinence while avoiding double-counting in the overall environment consideration, and ensure the accuracy.

[0022] 4. Denote the devices with a combined probability of two devices exceeding the probability threshold as merged devices, and retrieve the carbon emission data of the combination of the two in the historical carbon emission data and the previously calculated device carbon emission data according to the merged device information to output the compensation value of the device carbon emission data calculated later. If there is no online calculated device carbon emission data, the output device carbon emission data is the value calculated based on the merged carbon emission data. Thus, when conducting green certification of power equipment, carbon emission reservation can be made in advance according to the necessity of device combination, avoiding the excessive carbon emission of combined operation caused by ignoring the combination situation during the certification process.

[0023] 5. Obtain the overall carbon absorption capacity through the vegetation carbon absorption rate and vegetation coverage, obtain the grazing pressure on the vegetation according to the ratio of carnivores to herbivores, reflect the biodiversity within the region according to the biological abundance index, comprehensively calculate the impact of vegetation carbon absorption, and use the vegetation impact coefficient to reflect the vegetation carbon absorption capacity, ecological balance situation, and biodiversity amplification effect, so as to quantify the vegetation carbon sink potential, thereby accurately obtaining the carbon emission absorption and treatment capacity of the current region and improving the calculation accuracy of subsequent carbon emission constraints. Description of the Drawings

[0024] Figure 1 It is a schematic flowchart of the method for green certification of power equipment based on carbon emission data analysis of this application. Detailed Embodiment

[0025] To make the purpose, technical solution, and advantages of this application clearer, the following further elaborates on this application in combination with the drawings and embodiments. It should be understood that the specific embodiment described here is only the best embodiment of this application, only used to explain this application, and does not limit the protection scope of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of this application.

[0026] As Figure 1 shown, as Embodiment 1 of this application, the method for green certification of power equipment based on carbon emission data analysis includes the following steps:

[0027] Construct a carbon emission environmental impact relationship based on carbon emission data, local environmental data, and global environmental data in historical time series;

[0028] Obtain the device combination probability according to the similarity of the life cycle data of each device in historical time series, and construct a device carbon emission change model based on the device combination probability and its life cycle change factor;

[0029] Obtain carbon emission dynamic constraint data based on the carbon emission environmental impact relationship according to the current time series;

[0030] Obtain carbon emission prediction data based on the current device information according to the device carbon emission change model;

[0031] Perform green certification of power equipment based on carbon emission dynamic constraint data and carbon emission prediction data.

[0032] In this embodiment, the carbon emission environmental impact relationship is constructed based on the impact of carbon emissions on the local environment and the global environment in the historical time series, and the continuous impact of carbon emissions on the local environment and the global environment is quantified. At the same time, based on the equipment combination probability, a targeted equipment carbon emission change model is constructed according to each equipment type, and the carbon emission data of the entire process of the power equipment is comprehensively calculated and monitored. The equipment combination probability is used to retrieve the carbon emissions generated by the synchronous operation of the remaining equipment with the equipment as a life cycle change factor and included in the carbon emission prediction data of the equipment, and the impact of the equipment combination operation is included. The overall equipment carbon emission status is considered, and the carbon emissions and energy-saving effects of a single individual are considered from a macro perspective, so as to improve the accuracy of the certification effect and avoid ignoring the global carbon emission impact. And based on the dynamic tolerance of the environment, the carbon emission dynamic constraint data is output as the certification standard to improve the temporal adaptability and spatial adaptability of green certification, and improve the reliability of green certification of power equipment.

[0033] Specifically, the carbon emission environmental impact relationship is constructed based on the historical time series carbon emission data, local environmental data and global environmental data, including:

[0034] Obtain the local impact relationship based on the superposition value of carbon emission data in historical time series and the carbon emission impact data of the local environment;

[0035] Obtain the global impact relationship based on the carbon emission data superposition value in the historical time series and the carbon emission impact data of the global environment;

[0036] The environmental impact relationship of carbon emissions is obtained through local impact relationship and global impact relationship.

[0037] The impact of carbon emissions from power equipment on the local environment is quantified through local impact relationships, and the impact of carbon emissions from power equipment on the global environment is quantified through global environmental impact relationships. Carbon emission constraints are considered from the local and global impacts respectively, thereby avoiding certification bias caused by macro data ignoring local impacts.

[0038] In some cases, the carbon emission environmental impact relationship is constructed based on historical carbon emission data, local environmental data, and global environmental data, including:

[0039] Get local environment data based on the preset area and execute for each preset area:

[0040] Obtain carbon emission data that changes synchronously with local environmental data;

[0041] Obtain the corresponding equipment spatial location based on carbon emission data and build spatial impact relationships;

[0042] Construct a time impact relationship based on the local environmental parameters in the local environmental data and the carbon emission data superposition value;

[0043] Obtain the local impact relationship based on the spatial impact relationship and the time impact relationship.

[0044] Construct the spatial impact through the carbon emission impact corresponding to the device spatial position, and construct the time impact relationship through the carbon emission data superposition impact under the time series change. In this way, obtain the local impact relationship corresponding to a single region to quantify the impact relationship of the carbon emissions of power equipment on each region at different positions and different time sequences. In this case, it also includes:

[0045] Construct a global impact relationship based on the global environmental parameters in the global environmental data and the carbon emission data superposition value;

[0046] Obtain the carbon emission environmental impact relationship based on the global impact relationship and the local impact relationship.

[0047] The local environmental data at least includes the carbon emission impact data of the local environment and the local environmental parameters. The global environmental data at least includes the carbon emission impact data of the global environment and the global environmental parameters. The carbon emission impact data at least includes the temperature impact data, the air quality impact data, the water quality impact data, the soil pollution impact data, and the ecosystem impact data. The local environmental parameters at least include the local weather data and the local ecological data. The global environmental parameters at least include the global weather data and the global ecological fluctuation data. The carbon emission impact data can obtain the impact value of carbon emissions on the environment under the condition of excluding other interference factors according to expert experience. The local weather data at least includes the local temperature, the local precipitation, the local wind speed, the local wind direction, and the local humidity. The global weather data at least includes: the temperature combination, the precipitation combination, the wind speed combination, the wind direction combination, and the humidity combination. The local ecological data at least includes the local vegetation coverage rate and the local vegetation carbon absorption rate. The global ecological data at least includes the global vegetation coverage rate and the global vegetation carbon absorption rate. It can be understood that currently, the positive effect of the vegetation coverage rate on carbon emissions has been considered in terms of the carbon storage capacity of plants and microorganisms. Therefore, the local or global vegetation carbon impact situation is calculated here by using the vegetation coverage rate and the carbon absorption ability of vegetation.

[0048] Among them, obtaining the corresponding device spatial position according to the carbon emission data, and constructing the spatial impact relationship includes:

[0049] Obtain the carbon emission data with a fluctuation similarity to the local environmental data;

[0050] Obtain the device spatial position corresponding to the carbon emission data;

[0051] Construct the spatial impact relationship according to the device spatial position, the local environmental data fluctuation value, and the carbon emission data fluctuation value.

[0052] By using local environmental data and carbon emission data with similar fluctuations, the spatial locations of devices that may affect the current local environment are obtained. Then, based on the correlation between the historical carbon emission data fluctuation values and the local environmental data fluctuation values at each device's spatial location, a spatial impact relationship is constructed to demonstrate the impacts of different devices on different local environments and avoid single calculations of device carbon emission impacts. It can be understood that in this embodiment, the impacts of device carbon emissions on different local environments are obtained through the spatial impact relationship and the time impact relationship respectively, while the impact of device carbon emissions on the overall environment is obtained based on the global time impact. This can not only avoid ignoring the impacts on other adjacent or relevant local environments when conducting green certification for power devices, but also avoid double-counting environmental impacts during overall carbon emission calculations, improve pertinence, avoid double-counting in overall environmental considerations, and ensure accuracy.

[0053] In some cases, obtaining the device combination probability based on the similarity of the life cycle data of each device in the historical time series, and constructing a device carbon emission change model based on the device combination probability and its life cycle change factor includes:

[0054] Obtaining the transportation data, operation data, and scrapping data of each device in the historical time series;

[0055] Based on the clustering algorithm, obtaining the device combination probability according to the similarity of the transportation data, operation data, and scrapping data of each device in the historical time series;

[0056] For each device type, performing carbon emission data correction processing according to the device combination probability, and obtaining a device carbon emission change model based on the processed carbon emission data.

[0057] In this case, the carbon emissions of the equipment life cycle are calculated through the processes of equipment transportation, operation, and scrapping. The transportation data includes at least the transportation mode, transportation distance, transportation energy consumption, and transportation carbon emission data. The operation data includes at least the operation energy consumption, operation time, load data, maintenance data, operation efficiency fluctuation data, and operation carbon emission data. The scrapping data includes at least the scrapping time, scrapping factors, scrapping mode, and scrapping carbon emission data. The clustering algorithm can be in the form of K-means, DBSCAN, etc. According to the similarity of the transportation data, operation data, and scrapping data of each equipment, the combined characteristics of each equipment in the entire life cycle of transportation, operation, and scrapping are obtained. For example, whether the transportation of two equipment can be carried out simultaneously, whether the operation of a certain equipment is accompanied by the operation of another equipment, and whether the scrapping of a certain equipment means the scrapping of another equipment. Thus, according to each equipment type, the carbon emission data correction process is performed according to the equipment combination probability. For example: for equipment combinations that can be transported simultaneously, calculate the carbon emissions of the combined transportation, and accordingly correct the carbon emission data of individual equipment; for equipment with mutual influence in operation and scrapping, calculate the carbon emissions of the equipment combined. By analyzing the transportation, operation, and scrapping data of the equipment, comprehensively and accurately evaluate the carbon emissions of the equipment in the entire life cycle, identify the deviations and omissions in the carbon emission data, and improve the accuracy of power equipment certification. In other cases, the entire life cycle can also include processes such as raw material extraction and production.

[0058] In this embodiment, obtaining the equipment combination probability according to the similarity of the life cycle data of each equipment in the historical time series includes:

[0059] Obtain the transportation similarity of each equipment based on the transportation type, transportation time series, and transportation path of each equipment in the historical time series;

[0060] Obtain the operation similarity of each equipment based on the operation dependency relationship and operation time series of each equipment in the historical time series;

[0061] Obtain the scrapping similarity of each equipment based on the structure type, scrapping time series, and scrapping area of each equipment in the historical time series;

[0062] Obtain the equipment combination probability based on multi-dimensional scaling analysis combined with equipment transportation similarity, equipment operation similarity, and equipment scrapping similarity.

[0063] When two pieces of equipment are both transported in the same way and the transportation time series and transportation starting points are the same, it is considered that the transportation between these two pieces of equipment is the same, that is, the carbon emissions generated during transportation at this time are jointly generated by the two pieces of equipment. When two pieces of equipment have a section of the same transportation path (the transportation time series and transportation type are the same on this transportation path), it is considered that the transportation between these two pieces of equipment is similar, and the operation similarity is calculated according to the same proportion of the transportation path.

[0064] A running dependency means that there is a situation where devices are used in combination. When there is a running dependency between two devices and they have the same running time sequence, it is considered that the two devices are running the same. At this time, the running of one device necessarily implies the running of another device, and the same is true for carbon emissions. When there is a running dependency between two devices during a certain time sequence, the running of one device necessarily implies the running of another device during this time sequence. At this time, the carbon emissions of the two devices are correlated during this time sequence, and the two devices are considered to be running similarly. The running similarity is calculated based on the proportion of the same running time sequence. Of course, in some cases, the running dependency between two devices is not inevitable. At this time, the running similarity is calculated based on the proportion of the existence of the running dependency and the proportion of the same running time sequence. For example, in the past year, there were n A devices running, but only n - k B devices were running with the A devices. At this time, the running similarity of A and B devices is , if the B device only runs with the A device for a period of time and at this time, the running similarity of A and B devices is , where is the running time sequence of the A device. The running dependency can be determined according to the device type and the distance between devices.

[0065] When two devices have the same structural type and the same scrapping time sequence and scrapping area, it is considered that the two devices are scrapped the same. For example, redundant devices with the same scrapping time sequence. At this time, the carbon emissions generated by scrapping are jointly generated by the two devices. When two devices have some same structural types, it is considered that the two devices are scrapped similarly. The scrapping similarity is calculated based on the proportion of the same structural type. For example, the hardware structure of device C contains 30% metal materials and 70% non-metal materials, and the hardware structure of device D contains 100% non-metal materials. The scrapping similarity of C and D devices is . In some other cases, the scrapping time sequence includes the actual scrapping time sequence and the shutdown time sequence. The actual scrapping time sequence is the time point when the device is actually disassembled and recycled, and the shutdown time sequence is the time point when the device stops running. Since there are some cases where the actual scrapping time sequences of some devices are the same but the shutdown time sequences are different, therefore, the difference between the actual scrapping time sequence and the shutdown time sequence is taken into account in the scrapping similarity. For example, the shutdown time sequence of device C is the same as the actual scrapping time sequence, and the shutdown time sequence of device D is earlier than that of device C, and the actual scrapping time sequences are the same. At this time, the scrapping similarity of C and D devices is , represents the actual scrapping time sequence of the device, represents the shutdown time sequence of the device, represents the actual scrapping time sequence of device C, represents the shutdown time sequence of device D.

[0066] Transport similarity, operation similarity, and scrapping similarity reflect the combined probability among devices during the processes of transportation, operation, and scrapping. Multidimensional scaling analysis is used to fuse the combined probabilities in these three aspects of transportation, operation, and scrapping to obtain the device combined probability, so as to comprehensively evaluate the overall characteristics of the device combination. The device combined probability P is as follows:

[0067] ;

[0068] Among them, represents the transportation weight, represents the operation weight, represents the scrapping weight, , represents the transportation similarity in the device transportation dimension, represents the operation similarity in the device operation dimension, represents the scrapping similarity in the device scrapping dimension.

[0069] In some cases, constructing a device carbon emission change model based on the device combined probability and its life cycle change factors includes:

[0070] Constructing a device combined probability calculation layer based on the initial transportation weight, initial operation weight, and initial scrapping weight;

[0071] Constructing a factor matching layer for the device combined probability, initial probability threshold, and life cycle change factors;

[0072] Constructing an initial carbon emission change model with the device combined probability calculation layer and the factor matching layer;

[0073] Iteratively updating the initial transportation weight, initial operation weight, initial scrapping weight, and initial probability threshold in the initial carbon emission change model according to the carbon emission data of each device under historical time series to obtain the carbon emission change model.

[0074] Training the carbon emission change model with the carbon emission data of each device under historical time series to obtain the transportation weight, operation weight, scrapping weight, and probability threshold that conform to the historical situation. When the device combined probability between two devices is greater than the probability threshold, the carbon emission data of another device is taken as a life cycle change factor and incorporated into the consideration of the carbon emission data of the current device.

[0075] In other cases, constructing a device carbon emission change model based on the device combined probability and its life cycle change factors also includes:

[0076] Constructing a data compensation layer based on the merged unique strategy;

[0077] Constructing an initial device carbon emission change model with the neural network model architecture and the data compensation layer;

[0078] Train the initial model of equipment carbon emission changes with the processed carbon emission data to obtain the equipment carbon emission change model.

[0079] Execute at the data compensation layer:

[0080] Merge the carbon emission data according to the equipment combination probability, and output the merged equipment information according to the merged result of the carbon emission data;

[0081] Calculate the equipment carbon emission compensation value according to the merged equipment information and the equipment carbon emission calculation time series.

[0082] When the equipment combination probability is greater than the probability threshold, the carbon emission data of the corresponding equipment is considered for merging, and only a single carbon emission value is output. Through the construction of carbon emission data merging and the data processing layer, the associated equipment carbon emissions will only be calculated once, thus avoiding duplicate calculations of carbon emission data. In this embodiment, for example: when the equipment combination probability of equipment E and F is greater than the probability threshold, the transportation carbon emission data of equipment E and F is merged according to transportation similarity, the operation carbon emission data of equipment E and F is merged according to operation similarity, and the scrapping carbon emission data of E and F is merged according to scrapping similarity. If equipment E performs green certification first, the transportation carbon emission data of equipment E is:

[0083] ;

[0084] Among them, represents the transportation carbon emission data of equipment E predicted according to historical data;

[0085] The operation carbon emission data of equipment E is:

[0086] ;

[0087] Among them, represents the operation carbon emission data of equipment E predicted according to historical data, represents the operation carbon emission data of equipment F predicted according to historical data;

[0088] The scrapping carbon emission data of equipment E is:

[0089] ;

[0090] Among them, represents the scrapping carbon emission data of equipment E predicted according to historical data;

[0091] And the transportation carbon emission data of equipment F is:

[0092] ;

[0093] Among them, represents the carbon emission data of F equipment transportation predicted according to historical data, represents the transportation similarity between E and F equipment, represents the actual carbon emission data of E equipment transportation;

[0094] The carbon emission data of F equipment operation is:

[0095] ;

[0096] Among them, represents the carbon emission data of F equipment operation predicted according to historical data, represents the actual carbon emission data of E equipment operation;

[0097] The carbon emission data of F equipment scrapping is:

[0098] ;

[0099] Among them, represents the carbon emission data of F equipment scrapping predicted according to historical data, represents the scrapping similarity between E and F equipment, represents the actual carbon emission data of E equipment scrapping.

[0100] In this embodiment, the equipment with the equipment combination probability exceeding the probability threshold is recorded as the merged equipment. According to the merged equipment information, the carbon emission data of the combination of the two in the historical carbon emission data and the previously calculated equipment carbon emission data are retrieved to output the compensation value of the equipment carbon emission data calculated later, that is, the parts of transportation, operation, and scrapping that have been calculated by the previous equipment, so as to ensure the uniqueness of the merger. If there is no previously calculated equipment carbon emission data, the output equipment carbon emission data is the value calculated according to the merged carbon emission data. Thus, when conducting green certification for power equipment, carbon emission reservation can be made in advance according to the necessity of equipment combination, avoiding the over-standard of combined operation carbon emissions caused by ignoring the combination situation during the certification process. And for the power equipment certified later, since the equipment carbon emission compensation value is output according to the carbon emission reservation situation of the previous equipment in the data compensation layer, power equipment certification is carried out in this way, avoiding the repeated calculation of equipment carbon emissions and ensuring the accuracy of carbon emission calculation in the case of equipment combination.

[0101] It can be understood that in some cases, the probability thresholds include transportation probability threshold, operation probability threshold, scrapping probability threshold, and comprehensive probability threshold. When any similarity exceeds the corresponding probability threshold, the corresponding process carbon emissions of the corresponding device can also be incorporated into the life cycle change factor of the current device. Although only the real-time situation of pairwise device combinations is given in this embodiment, in practical applications, the combination situations of multiple devices can also be considered, that is, the respective processes of multiple devices can be separately incorporated into the consideration of the life cycle change factor of the current device.

[0102] Furthermore, obtaining carbon emission dynamic constraint data based on the carbon emission environment impact relationship according to the current time sequence includes:

[0103] Obtaining a local environment impact value based on the current time sequence, the time impact relationship, the current device spatial position, and the spatial impact relationship;

[0104] Obtaining a global environment impact value based on the current time sequence and the global impact relationship;

[0105] Obtaining local environment constraints and global environment constraints according to the current device spatial position;

[0106] Obtaining local carbon emission constraints with the local environment constraints and the local environment impact value, and obtaining global carbon emission constraints with the global environment constraints and the global environment impact value.

[0107] Among them, the local environment constraints and the global environment constraints can be obtained according to the carbon emission requirements of the actual region and the carbon emission requirements of the overall region. The local environment impact value and the global environment impact value at the current time sequence are respectively used to obtain the absorption capacity of carbon emissions by the local environment and the global environment, so as to obtain the actually acceptable carbon emissions considering environmental impacts under the local environment constraints and the global environment constraints. At the same time, the changes in the time series and spatial position are considered, enabling the carbon emission management to be dynamically adjusted with the change of environmental conditions, and realizing the adaptation of local and global carbon emission requirements.

[0108] In this embodiment, carbon emission prediction data is obtained based on the current device information according to the device carbon emission change model. The carbon emission prediction data is calculated according to whether there is a corresponding merged device for the device and whether the corresponding merged device has been calculated previously. If so, the carbon emission prediction data output by the device is the difference between the prediction data initial value calculated according to the neural network model and the carbon emission prediction data output by the merged device. If not, the carbon emission prediction data output by the device is the prediction data initial value calculated according to the neural network model.

[0109] Performing green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data includes:

[0110] If the carbon emission prediction data meets the carbon emission dynamic constraint data, the green certification of the power equipment passes;

[0111] If the carbon emission prediction data does not meet the carbon emission dynamic constraint data, the green certification of the power equipment fails.

[0112] Specifically, if the carbon emission prediction data meets both the local carbon emission constraint and the global carbon emission constraint, the green certification of the power equipment passes. If the carbon emission prediction data does not meet any one of the local carbon emission constraint and the global carbon emission constraint, the green certification of the power equipment fails.

[0113] By obtaining the local environmental impact value and the global environmental impact value, as well as the corresponding local environmental constraint and global environmental constraint, the impact of equipment carbon emissions on the local and global environments is comprehensively considered. At the same time, by calculating the difference between the initial value of the prediction data and the carbon emission prediction data output by the combined equipment, the prediction result is further corrected to ensure that the predicted carbon emission data can not only reserve the necessary carbon emission situations that may exist in the combination, but also avoid duplicate calculation of carbon emissions and improve the prediction accuracy.

[0114] As the second embodiment of the present application, the local ecological data further includes the local animal species ratio and the local biological abundance index. The global ecological data includes at least the global animal species ratio and the global biological abundance index. The local animal species ratio is obtained according to the ratio of the number of carnivores to herbivores in the local environment, and the local biological abundance index is calculated according to the species richness in the local environment, and can be calculated according to the species counting method, the Shannon-Wiener diversity index, etc.; the global animal species ratio is obtained according to the ratio of the number of carnivores to herbivores in the global environment, and the global biological abundance index is calculated according to the species richness in the global environment, and can also be calculated according to the species counting method, the Shannon-Wiener diversity index, etc.

[0115] Among them, constructing the carbon emission environment impact relationship based on the carbon emission data, local environmental data, and global environmental data under historical time series further includes:

[0116] Obtaining the vegetation impact coefficient according to the animal species ratio, biological abundance index, vegetation coverage rate, and vegetation carbon absorption rate;

[0117] Obtaining the carbon emission environment impact relationship according to the air temperature, precipitation, wind speed, wind direction, humidity, and vegetation impact coefficient combined with the carbon emission data superposition value.

[0118] Specifically, constructing the time impact relationship according to the local environmental parameters in the local environmental data and the carbon emission data superposition value includes:

[0119] Obtaining the local vegetation impact coefficient according to the local animal species ratio, local biological abundance index, local vegetation coverage rate, and local vegetation carbon absorption rate;

[0120] Obtain the influence weight of the time influence relationship based on the local temperature, local precipitation, local wind speed, local wind direction, local humidity, and the combined carbon emission data superposition value with the local vegetation influence coefficient.

[0121] Construct the global influence relationship according to the global environmental parameters in the global environmental data and the carbon emission data superposition value, including:

[0122] Obtain the global vegetation influence coefficient according to the global animal species proportion, global biological abundance index, global vegetation coverage rate, and global vegetation carbon absorption rate;

[0123] Obtain the influence weight of the global influence relationship according to the temperature combination, precipitation combination, wind speed combination, wind direction combination, humidity combination, and the global vegetation influence coefficient combined with the carbon emission data superposition value.

[0124] Specifically, calculate the local vegetation influence coefficient according to the following formula:

[0125] ;

[0126] Among them, represents the local vegetation influence coefficient, I represents the number of local vegetation species, represents the vegetation coverage rate of the i-th type of vegetation at time t, represents the carbon absorption rate of the i-th type of vegetation, represents the number of carnivores at time t, represents the number of herbivores at time t, represents the local biological abundance index.

[0127] Obtain the influence weight of the time influence relationship according to the local temperature, local precipitation, local wind speed, local wind direction, local humidity, and the local vegetation influence coefficient combined with the carbon emission data superposition value:

[0128] ;

[0129] Among them, represents the local carbon emission influence at time t; represents the carbon emission data superposition value at time t; represents the local temperature influence weight, represents the local temperature at time t; represents the local precipitation influence weight, represents the local precipitation at time t; represents the local wind speed influence weight, represents the wind speed at time t; represents the local wind direction influence weight, represents the wind direction coefficient at time t; represents the local humidity influence weight, represents the local humidity; represents the local ecological impact coefficient.

[0130] In this embodiment, the overall carbon absorption capacity is obtained through the vegetation carbon absorption rate and the vegetation coverage rate, the grazing pressure of the vegetation is obtained according to the ratio of carnivores to herbivores, and the biodiversity within the region is reflected according to the biological abundance index. The impact of vegetation carbon absorption is comprehensively calculated, and the vegetation impact coefficient is used to reflect the vegetation carbon absorption capacity, the ecological balance situation, and the biodiversity amplification effect, so as to quantify the vegetation carbon sink potential, thereby accurately obtaining the carbon emission absorption and treatment capacity of the current region and improving the calculation accuracy of subsequent carbon emission constraints. It can be understood that in some other cases, the relationship between the animal ratio and the grazing pressure is not linear. After converting the non-linear relationship between the animal ratio and the grazing pressure by using the linear regression method, the calculation of the local vegetation impact coefficient and the impact weight is carried out.

[0131] Similarly, the same temperature, precipitation, wind speed, wind direction, and humidity in the global weather data are combined to obtain the temperature combination, precipitation combination, wind speed combination, wind direction combination, and humidity combination. According to the proportional change of each value in the combination, the impact relationship between the global weather conditions and carbon emissions is further calculated. Then, based on the temperature combination, precipitation combination, wind speed combination, wind direction combination, humidity combination, and the global vegetation impact coefficient, the impact weight of the global impact relationship is obtained by combining with the carbon emission data superposition value.

[0132] Based on the multi-source data fusion of ecology and climate, the carbon emission absorption capacity under different weather factors and ecological factors is obtained, so as to accurately calculate the impact of the unabsorbed carbon emissions on the actual environment, improve the dynamic adaptability of carbon emission constraints, and ensure the accuracy of the green certification of power equipment.

[0133] As the third embodiment of the present application, a computer-readable storage medium stores a computer program thereon. When the program is executed by a processor, it implements the steps of the power equipment green certification method based on carbon emission data analysis in any of the above embodiments. The computer-readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0134] In this embodiment, an electronic device is further provided. The electronic device may include a processor and a memory for storing executable instructions of the processor. Among them, the processor is configured to execute the steps of the power equipment green certification method based on carbon emission data analysis in any of the above embodiments by executing the executable instructions.

[0135] The above-described specific embodiments are the preferred embodiments of the power equipment green certification method based on carbon emission data of the present application, and do not limit the specific implementation scope of the present application. The scope of the present application includes but is not limited to this specific embodiment. Any equivalent changes made according to the shape and structure of the present application are within the protection scope of the present application.

Claims

1. A green certification method for power equipment based on carbon emission data analysis, characterized in that: The method includes the following steps: Construct a carbon emission environment impact relationship based on carbon emission data, local environment data, and global environment data in historical time series; Obtain the equipment combination probability according to the similarity of the life cycle data of each device in historical time series, and construct an equipment carbon emission change model based on the equipment combination probability and its life cycle change factor; Obtain carbon emission dynamic constraint data based on the carbon emission environment impact relationship according to the current time series; Obtain carbon emission prediction data based on the equipment carbon emission change model according to the current equipment information; Execute green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data; The life cycle change factor includes the carbon emissions synchronously generated by each device and the remaining devices during the whole life cycle; The carbon emission dynamic constraint data includes environmental constraints corresponding to carbon emission requirements and environmental impact values corresponding to the environmental carbon emission absorption capacity in the global environment and the local environment; Among them, construct an initial equipment carbon emission change model with a neural network model architecture and a data compensation layer; train the initial equipment carbon emission change model with the processed carbon emission data to obtain the equipment carbon emission change model; Execute in the data compensation layer: Merge the carbon emission data according to the equipment combination probability, output the merged equipment information according to the merged result of the carbon emission data; output the equipment carbon emission compensation value according to the merged equipment information and the equipment carbon emission calculation time series.

2. The method for green certification of power equipment based on carbon emission data analysis according to claim 1, characterized in that: The construction of the carbon emission environment impact relationship based on the carbon emission data, local environment data, and global environment data in historical time series includes: Obtain the local impact relationship according to the carbon emission data superposition value in historical time series and the carbon emission impact data of the local environment; Obtain the global impact relationship according to the carbon emission data superposition value in historical time series and the carbon emission impact data of the global environment.

3. The method for green certification of power equipment based on carbon emission data analysis according to claim 1, characterized in that: The construction of the carbon emission environment impact relationship based on the carbon emission data, local environment data, and global environment data in historical time series includes: Obtain the local environment data according to the preset area, and execute for each preset area: Obtain the carbon emission data that changes synchronously with the local environment data; Obtain the corresponding equipment spatial position according to the carbon emission data, and construct a spatial impact relationship; Construct a time impact relationship according to the local environment parameters in the local environment data and the carbon emission data superposition value; Obtain the local impact relationship with the spatial impact relationship and the time impact relationship.

4. The method for green certification of power equipment based on carbon emission data analysis according to claim 3, characterized in that: Obtaining the corresponding equipment spatial position according to the carbon emission data and constructing a spatial impact relationship includes: Obtain the carbon emission data with a fluctuation similarity to the local environment data; Obtain the equipment spatial position corresponding to the carbon emission data; Construct a spatial impact relationship according to the equipment spatial position, the local environment data fluctuation value, and the carbon emission data fluctuation value.

5. The method for green certification of power equipment based on carbon emission data analysis according to claim 1, characterized in that: The construction of the carbon emission environment impact relationship based on the carbon emission data, local environment data, and global environment data in historical time series includes: Obtain the vegetation impact coefficient based on the proportion of animal species, the biological abundance index, the vegetation coverage rate, and the vegetation carbon absorption rate; Obtain the relationship between carbon emission environmental impacts by combining the carbon emission data superposition value with the air temperature, precipitation, wind speed, wind direction, humidity, and the vegetation impact coefficient.

6. The method for green certification of power equipment based on carbon emission data analysis according to claim 1, characterized in that: The obtaining of the equipment combination probability based on the similarity of the life cycle data of each device in historical time series includes: Obtain the transportation similarity of each device based on the transportation type, transportation time series, and transportation path of each device in historical time series; Obtain the operation similarity of each device based on the operation dependency and operation time series of each device in historical time series; Obtain the scrapping similarity of each device based on the structure type, scrapping time series, and scrapping area of each device in historical time series; Obtain the equipment combination probability based on multi-dimensional scaling analysis combined with the equipment transportation similarity, equipment operation similarity, and equipment scrapping similarity.

7. The method for green certification of power equipment based on carbon emission data analysis according to claim 6, characterized in that: The constructing of the equipment carbon emission change model based on the equipment combination probability and its life cycle change factors includes: Construct an equipment combination probability calculation layer based on the initial transportation weight, initial operation weight, and initial scrapping weight; Construct a factor matching layer for the equipment combination probability, initial probability threshold, and life cycle change factors; Construct an initial carbon emission change model with the equipment combination probability calculation layer and the factor matching layer; Iteratively update the initial transportation weight, initial operation weight, initial scrapping weight, and initial probability threshold in the initial carbon emission change model according to the carbon emission data of each device in historical time series to obtain the carbon emission change model.

8. The method for green certification of power equipment based on carbon emission data analysis according to claim 3, characterized in that: The obtaining of the carbon emission dynamic constraint data based on the current time series according to the carbon emission environmental impact relationship includes: Obtain the local environmental impact value based on the current time series and the time impact relationship, the current device spatial position, and the spatial impact relationship; Obtain the global environmental impact value based on the current time series and the global impact relationship; Obtain the local environmental constraint and the global environmental constraint according to the current device spatial position; Obtain the local carbon emission constraint with the local environmental constraint and the local environmental impact value, and obtain the global carbon emission constraint with the global environmental constraint and the global environmental impact value.

9. The method for green certification of power equipment based on carbon emission data analysis according to claim 1, characterized in that: The performing of the green certification of power equipment based on the carbon emission dynamic constraint data and the carbon emission prediction data includes: If the carbon emission prediction data meets the carbon emission dynamic constraint data, the green certification of the power equipment passes; If the carbon emission prediction data does not meet the carbon emission dynamic constraint data, the green certification of the power equipment fails.

Images