Multi-dimensional evaluation method, system and equipment of electrical-carbon coupling system and storage medium

Abstract

The invention discloses a multi-dimensional evaluation method, system and device for an electricity-carbon coupling system and a storage medium, and the method comprises the steps: collecting multi-source data of an electric power region, carrying out the preprocessing of the multi-source data, and constructing an electric power and external influence combined data set; the method comprises the following steps: extracting weather, day type and load combinations in each time period in a power and external influence joint data set, and clustering combinations with consistent load change directions, marking a lifting interval and frequency, screening typical modes, and forming a load trend segmentation identification set; acquiring multi-energy output data, matching the multi-energy output data with a load change direction, identifying an abnormal time period based on energy proportion fluctuation, and classifying according to dominant energy to generate an energy dominant type power-carbon offset interval set; and matching the load rising and falling interval in the load trend segmentation identification set with a dominant path in the energy-dominant electrical-carbon offset interval set, optimizing path parameters and generating an electrical-carbon coupling transfer structure table. The energy use efficiency and the carbon emission control precision can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy management, and in particular to a multi-dimensional evaluation method, system, device and storage medium of an electricity-carbon coupling system. BACKGROUND

[0002] The technical field of energy management includes the whole process of energy acquisition, conversion, storage, distribution and utilization. Its core content is to realize efficient allocation and coordinated operation of energy resources through scientific methods. This technical field systematically covers power system management, energy consumption monitoring, carbon emission calculation, energy flow and carbon flow optimization, energy utilization efficiency evaluation, and energy planning strategy formulation. The overall goal is to establish a comprehensive management framework covering electricity and carbon emissions, to balance and link energy between different links, and to ensure that the relationship between energy consumption and carbon emissions can be fully and accurately quantified and evaluated, thereby providing clear data basis and method support for energy system planning and operation. Among them, a method for multi-dimensional evaluation of electricity-carbon coupling system refers to a method for systematically measuring and decomposing the correlation between power consumption and carbon emissions in the context of energy management. The technical matters covered include power load collection, energy flow recording, carbon emission factor matching, and data interaction modeling in different dimensions. The method is to combine power consumption data and carbon emission coefficients to establish a multi-dimensional index system, to subdivide and evaluate the electricity conversion process, energy use structure, and carbon emission distribution. The specific contents include time-series recording of power consumption, classification selection of carbon emission factors, segmented expression of energy flow path, and matrix description of the relationship between multi-dimensional indexes, thereby forming a comprehensive evaluation method for electricity-carbon coupling system.

[0003] The existing technology relies too much on single-dimensional data monitoring and fails to fully consider the synergistic effect of different energy, climate change and day types. The existing method has limitations in comprehensive evaluation of electricity-carbon coupling system and cannot accurately capture the combined effect of various external factors on load fluctuation and energy output, resulting in insufficient precision in carbon emission prediction and energy consumption optimization. In actual operation, it may lead to inaccurate prediction of power load fluctuation, affecting energy planning decision-making and limiting the accuracy and feasibility of carbon emission regulation. SUMMARY

[0004] In view of the above existing problems, the present application is proposed. Therefore, the present application provides a multi-dimensional evaluation method, system, device and storage medium of an electricity-carbon coupling system to solve the problems mentioned in the background art.

[0005] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, embodiments of the present invention provide a multi-dimensional evaluation method for an electric-carbon coupled system, comprising: collecting multi-source data of an electric power region, preprocessing the multi-source data, and constructing a joint dataset of electric power and external influences; Extract meteorological data, daily data types, and load combinations for each time period from the combined data set of power and external influences; cluster combinations with consistent load change directions; mark the rise and fall intervals and frequencies; filter typical patterns; and form a load trend segmentation identifier set. Acquire multi-energy output data, match multi-energy output data with load change direction, identify abnormal periods based on energy proportion fluctuations, and generate energy-dominant type carbon offset interval sets according to dominant energy. The load trend segmentation identifiers are matched with the dominant paths of the energy-dominant electric carbon offset intervals, the path parameters are optimized, and an electric carbon coupling transfer structure table is generated.

[0006] As a preferred embodiment of the multi-dimensional evaluation method for the electric-carbon coupling system described in this invention, the construction of the joint dataset of power and external influences includes: acquiring the regional power load sequence, temperature record value, air humidity value, working and non-working labels in the public calendar, and real-time output data of wind and photovoltaic power; detecting the timestamps of each data item; comparing differences by calling the timestamp field; adjusting the time axis format of the data and unifying it to the same time base; and generating a unified time series value. Based on the unified time series values, the spliced ​​power load, temperature, humidity, day type and wind and photovoltaic output data are compared field by field to determine missing values ​​in the fields, and values ​​from adjacent time points are called for interpolation. Abnormal fields that fail to match are deleted to generate valid spliced ​​interval values. Based on the effective splicing interval values, the values ​​of power load, temperature, humidity, day type and wind and photovoltaic output are compared time-by-time to filter out the complete time period set without anomalies, thus obtaining the joint information set of power and external influences.

[0007] As a preferred embodiment of the multi-dimensional evaluation method for the electricity-carbon coupling system described in this invention, the following steps are taken: meteorological data, daily data types, and load combinations for each time period are extracted from the joint data set of electricity and external influences; combinations with consistent load change directions are clustered; rising and falling intervals and frequencies are marked; typical patterns are selected; and a load trend segmentation identifier set is formed, including: The temperature, humidity, day type, and power load value for each time period are extracted from the combined information set of power and external influences. A set of combinations of temperature, humidity, day type, and load value for each time period is constructed, the changing trend of each combination is identified, and a set of load change combinations is generated. Based on the load change combination set, directional analysis is performed on each combination to identify the combination clusters with the same load change direction, and the load increase or decrease intervals in each cluster are marked. The frequency of each interval is calculated to obtain the load change direction clusters. Based on the load change direction cluster, typical load change patterns are selected, mapped to the target time interval, and a load trend segmentation structure identifier set is generated.

[0008] As a preferred embodiment of the multi-dimensional evaluation method for the electric-carbon coupling system described in this invention, the method includes: acquiring multi-energy output data, matching the multi-energy output data with the load change direction, identifying abnormal periods based on energy proportion fluctuations, and generating an energy-dominant electric-carbon offset interval set according to the dominant energy source, including: Acquire actual power output data from wind, solar, and thermal power, compare it with the direction of power load change, perform unified time alignment on the data of different energy sources based on timestamps, match the directional relationship between power output of each energy source and load change, and generate an energy and load matching dataset. Based on the energy and load matching dataset, the changes in the proportion of wind, photovoltaic and thermal power energy in each time period are analyzed, the relationship between energy proportion fluctuation and load change direction is identified, and the fluctuation amplitude in each time period is compared and judged in combination with the set fluctuation threshold. Abnormal fluctuations exceeding the threshold are screened out, and an energy fluctuation abnormal dataset is generated. Based on the energy fluctuation anomaly dataset, the filtered time period data are classified, energy-dominated carbon offset intervals are identified and marked, and analysis is performed on different intervals to establish corresponding classification rules, thereby generating a set of energy output-dominated carbon offset intervals.

[0009] As a preferred embodiment of the multi-dimensional evaluation method for the electric-carbon coupling system described in this invention, the method includes: matching the concentrated load rise and fall intervals of the load trend segmentation identifier with the dominant paths of the energy-dominant electric-carbon offset intervals, optimizing path parameters, and generating an electric-carbon coupling transfer structure table, including: Based on the load rise interval and load fall interval in the load trend segmentation structure identifier set, and combined with the wind energy dominant path in the energy output-dominant type carbon offset interval set, the load rise interval and the wind energy dominant path are matched to obtain the wind energy and load rise matching dataset. Based on the load decline intervals in the load trend segmentation structure identifier set and the thermal power dominant path in the energy output-dominant type electric carbon offset interval set, the load decline intervals and thermal power dominant paths are matched to obtain the thermal power and load decline matching dataset. Based on the aforementioned wind power and load increase matching dataset and thermal power and load decrease matching dataset, the matching relationships of each energy-dominant path are sorted out and optimized, the optimized path is constructed, and an energy-driven type electric carbon coupling transfer structure table is generated.

[0010] As a preferred embodiment of the multi-dimensional evaluation method for the electric-carbon coupling system described in this invention, the method further includes: based on the energy driving path in the electric-carbon coupling transfer structure table, marking the dominant periods of wind power, photovoltaic power, and thermal power within the daily load cycle, calculating the consistency between the load fluctuation direction and external factors, classifying the coupling state, and generating the electric-carbon coupling state relationship results. The results of the electricity-carbon coupling state relationship include wind-dominated periods, photovoltaic-dominated periods, thermal power-dominated periods, consistency between load fluctuations and external factor fluctuations, coupling state categories, state characteristic summaries, and energy-dominated period identifiers within daily load cycles.

[0011] As a preferred embodiment of the multi-dimensional evaluation method for the electro-carbon coupling system described in this invention, the acquisition of the electro-carbon coupling state relationship results includes: Based on the energy drive path in the energy drive type electro-carbon coupling transfer structure table, the time periods dominated by wind power, photovoltaic power and thermal power are marked within the daily load cycle. The dominant energy for each time period is distinguished, and the dominant energy for each time period is clearly identified by combining the time dimension, generating a dominant energy time period identifier set. Based on the dominant energy period identifier set, the load fluctuation direction and the fluctuation consistency of external factors within each period are calculated to obtain the quantitative value of fluctuation consistency for each period and generate a fluctuation consistency dataset. Based on the fluctuation consistency dataset, the coupling states of each time period are classified, and the characteristics of different coupling states are identified by combining the fluctuation consistency values. The state characteristics of each time period are then summarized to generate the electric-carbon coupling state relationship results.

[0012] Secondly, the present invention provides a multi-dimensional evaluation system for an electro-carbon coupling system, comprising: The multi-source data fusion and preprocessing module is used to collect multi-source data from the power area and preprocess the multi-source data to construct a joint dataset of power and external influences. The load trend pattern filtering module is used to extract the weather, daily type and load combination of each time period in the power and external influences joint dataset, cluster combinations with the same load change direction, mark the rise and fall intervals and frequencies, filter typical patterns, and form a load trend segmentation identifier set. The energy-dominant type carbon offset identification module is used to acquire multi-energy output data, match the multi-energy output data with the direction of load change, identify abnormal periods based on energy proportion fluctuations, and generate a set of energy-dominant type carbon offset intervals according to the dominant energy. The electric carbon coupling path optimization and structure generation module is used to match the load trend segmentation identification concentrated load rise and fall intervals with the dominant paths concentrated in the energy-dominant electric carbon offset intervals, optimize path parameters, and generate an electric carbon coupling transfer structure table.

[0013] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the multi-dimensional evaluation method for the electric-carbon coupling system.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the multi-dimensional evaluation method for the electro-carbon coupling system.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: By integrating multi-dimensional data and achieving unified time standard processing, this invention can more accurately analyze the correlation between load fluctuations and external factors, form a load trend structure, and identify the dominant role of energy output in carbon emission fluctuations by combining the dominant energy output path with load changes. This provides more accurate data support for power system optimization and carbon emission prediction, enhances the ability to synergistically optimize the relationship between energy system complexity and carbon emission, and improves energy use efficiency and carbon emission control accuracy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the process flow of a multi-dimensional evaluation method for an electro-carbon coupling system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the evaluation process of a multi-dimensional evaluation method for an electro-carbon coupling system according to an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0018] Example 1, referring to Figures 1-2 As one embodiment of the present invention, this embodiment provides a multi-dimensional evaluation method for an electric-carbon coupling system, including: S100: Collect multi-source data of the power area, preprocess the multi-source data, and construct a joint dataset of power and external influences; S200: Extract meteorological data, daily types, and load combinations from the joint data set of power and external influences for each time period, cluster combinations with consistent load change directions, mark the rise and fall intervals and frequencies, filter typical patterns, and form a load trend segmentation identifier set; S300: Acquire multi-energy output data, match multi-energy output data with load change direction, identify abnormal periods based on energy proportion fluctuations, and generate energy-dominant type electric carbon offset interval sets according to dominant energy. S400: Match the load trend segmentation identifier concentrated load rise and fall intervals with the dominant paths of the energy-dominant electric carbon offset intervals, optimize path parameters, and generate an electric carbon coupling transfer structure table.

[0019] It should be noted that existing technologies rely too heavily on single-dimensional data monitoring and fail to fully consider the synergistic effects of different energy sources, climate change, and daily patterns. Existing methods have limitations in the comprehensive assessment of electricity-carbon coupling systems, failing to accurately capture the combined impact of multiple external factors on load fluctuations and energy output. This results in insufficient accuracy in carbon emission prediction and energy consumption optimization, potentially leading to inaccurate electricity load fluctuation predictions in practical operation, affecting energy planning decisions, and limiting the accuracy and feasibility of carbon emission control. This invention acquires and integrates multi-source data, including electricity load, meteorological data, and renewable energy data, to form an information set. Through cluster analysis, it identifies representative load change patterns, matches the relationship between wind, solar, and thermal energy output and load fluctuations to determine the energy-dominant offset range, matches and optimizes load rise and fall trends with specific energy-dominant paths, marks each energy-dominant period within the daily cycle, and analyzes the coupling state to derive the electricity-carbon relationship. By integrating multi-dimensional data and performing unified time processing, load change patterns are subdivided and identified to analyze their correlation with external factors. Combined with energy output paths, their dominant role in carbon emissions is identified, providing accurate data support for power system optimization and carbon emission prediction. This enhances the ability to synergistically optimize the relationship between energy systems and carbon emissions, and improves energy efficiency and carbon emission control accuracy.

[0020] Reference Figure 2 In this embodiment of the invention, the construction of the joint dataset of power and external influences in step S100 includes: obtaining regional power load sequence, temperature record value, air humidity value, working and non-working labels in the public calendar, and real-time output data of wind power and photovoltaic power; detecting the timestamps of each data item; calling the timestamp field to compare differences; adjusting the time axis format of the data and unifying it to the same time base; and generating a unified time series value. In one feasible approach, the system retrieves a pre-defined statistical period (e.g., seven consecutive days) of power load data from a regional power dispatch center database. This load is time-resolutiond at 15 minutes and formatted as {timestamp, load value (MW)}. Simultaneously, it retrieves temperature records for the same region at the same time from a meteorological monitoring system. This temperature record is time-resolutiond at 30 minutes and formatted as {timestamp, temperature value (°C)}. It also retrieves humidity values ​​with a time-resolution of 1 hour and formatted as {timestamp, humidity value (%RH)}. Finally, it obtains weekday and non-weekday labels from a public calendar service interface. The data format is {date, day type label (0 for non-working day, 1 for working day)}. Real-time output data of wind and solar power is then accessed from the new energy monitoring platform, with a time resolution of 5 minutes. The data format is {timestamp, wind power output (MW), solar power output (MW)}. To unify the time base of all data, the timestamp field of all data is checked. For example, at 08:00:00 on a certain day within this period, the timestamp for power load data is "08:00:00", while the timestamp for temperature data is "08:00:00", and the timestamp for air humidity data is "08:00:00". At 08:00:00, wind power data is displayed as "08:00:00", and solar power data is displayed as "08:00:00". No adjustment is needed at this point. However, at 08:05:00, only wind power and solar power data are available; data for power load, temperature, and humidity are missing. Therefore, a unified time axis format adjustment is required. A 15-minute timeframe is selected as the unified time base. The 5-minute resolution wind power and solar power data are then aggregated. Specifically, this involves aggregating three 5-minute data points between 08:00:01 and 08:15:00 (e.g., wind power outputs of 50.1MW, 50.3MW, and 50.5MW respectively). The arithmetic mean of 50.3MW is used as the representative value for 08:15:00. For the 30-minute temperature data and the 1-hour humidity data, the forward filling method is used. The temperature value of 20.5℃ at 08:00:00 is assigned to the time point of 08:15:00, and the humidity value of 65% at 08:00:00 is assigned to the three time points of 08:15:00, 08:30:00, and 08:45:00. For the day type label, the label value of "first day" (working day) is assigned to all 15-minute time points of the day. All data is processed in this way to generate a unified time series value.

[0021] In this embodiment of the invention, in step S100, based on the unified time series value, the spliced ​​power load, temperature, humidity, day type and wind and photovoltaic output data are compared field by field to determine the missing values ​​in the fields, the values ​​of adjacent time points are called for interpolation judgment, the abnormal fields that fail to match are deleted, and the valid splicing interval value is generated. In one feasible approach, based on time-series values, the spliced ​​wide table data, including electricity load, temperature, humidity, day type, and wind and solar power output, is compared field by field to determine if any fields contain missing values. For example, when checking the data table up to the row "2nd day 14:30:00", the electricity load value is found to be 1500MW, the temperature value is 28.2℃, the day type is 1, the wind power output is 120MW, and the solar power output is 350MW, but the humidity value field is NULL. In this case, the values ​​of adjacent time points for this field are used for interpolation, extracting the humidity value of 72% at the previous time point "14:15:00" and the humidity value of 70% at the next time point "14:45:00", and calculating their arithmetic mean. (72+70) / 2=71%. Fill the humidity value field of "14:30:00" with the calculated result 71%. After this operation, continue to scan the data to delete the abnormal fields that cannot be matched. The judgment rule for abnormal fields is set as follows: if the three key energy fields of power load, wind power, and photovoltaic output are all empty at the same time point, the data at that time point is judged as an invalid record. For example, after the interpolation process, it is found that in the data row of the time point "02:00:00 on the third day", the three items of power load, wind power output, and photovoltaic output are all NULL. Even if the temperature and humidity values ​​exist, the data in this row is still considered invalid and deleted from the dataset as a whole. The missing value interpolation and invalid record deletion of the entire dataset are completed in turn to generate valid spliced ​​interval values.

[0022] In this embodiment of the invention, in step S100, based on the effective splicing interval values, the values ​​of power load, temperature, humidity, day type and wind and photovoltaic output are compared time-by-time to filter out the complete time period set without any abnormal points, and obtain the joint information set of power and external influences.

[0023] In one feasible approach, based on the effective splicing interval values, the values ​​of electricity load, temperature, humidity, day type, and wind and solar power output are compared time-by-time. This process aims to filter out the complete set of time periods without anomalies. The criteria for judging anomalies are set based on the statistical characteristics of historical data, specifically: the rate of change of electricity load value is defined as the difference between the current value and the previous value divided by the previous value, and its absolute value should not exceed 15%; the absolute value of the rate of change of temperature value should not exceed 5%; the absolute value of the rate of change of humidity value should not exceed 10%; and the absolute value of the rate of change of wind and solar power output should not exceed 50%. For example, at "day five, 10:00:00", the electricity load value is 1800MW, while at the previous time "09:00", the value is 1800MW. At 45:00, the load value was 1750MW, with a change rate of (1800-1750) / 1750=2.86%, which is within the 15% threshold range and is therefore considered normal. However, at 10:15:00, the load value suddenly increased to 2100MW, with a change rate of (2100-1800) / 1800=16.67%, which exceeds the 15% threshold. Therefore, the data at 10:15:00 was marked as an outlier. The same comparison and filtering logic was applied to all the values ​​at all time points in the dataset. All the time point data that were not marked as outliers were collected to form a time series set that is internally continuous and has reasonable numerical fluctuations, thus obtaining a joint information set of power and external influences. Table 1. Example of a joint information set on power and external impacts.

[0024] As shown in Table 1, this table displays a portion of the data in the joint information set of power and external influences formed after data collection, alignment, cleaning and interpolation, which forms the basis for subsequent analysis.

[0025] Furthermore, the joint information set of electricity and external influences includes electricity load sequences, temperature records, air humidity values, working and non-working day labels, wind power output data, and photovoltaic output data. The load trend segmentation structure identifier set includes load increase intervals, load decrease intervals, load change patterns, time interval mapping, load change frequency, temperature and humidity combination patterns, and the correlation between day type and load. The energy output-dominant type electric carbon offset interval set includes wind power output, photovoltaic output, thermal power output, load change direction, energy share change, fluctuation amplitude, abnormal fluctuation intervals, and fluctuation classification. The energy-driven type electric carbon coupling transfer structure table includes the matching of load increase intervals with wind power-dominant paths, the matching of load decrease intervals with thermal power-dominant paths, the matching relationship of energy-dominant paths, the optimized path performance, and the coupling characteristics between load and energy paths.

[0026] In this embodiment of the invention, step S200 extracts meteorological data, daily types, and load combinations for each time period from the joint dataset of power and external influences, clusters combinations with consistent load change directions, marks the rise and fall intervals and frequencies, filters typical patterns, and forms a load trend segmentation identifier set, including: The temperature, humidity, day type, and power load values ​​for each time period are extracted from the combined information set of power and external influences. A set of combinations of temperature, humidity, day type, and load values ​​for each time period is constructed, the changing trend of each combination is identified, and a set of load change combinations is generated. In one feasible approach, temperature, humidity, day type, and power load values ​​are extracted point-by-point from a combined information set of power and external influences to construct a set of combinations of temperature, humidity, day type, and load values ​​for each time period. This process treats each 15-minute interval data point as an independent combination unit. For example, for the time period "Day 6, 08:15:00", the extracted combination would be {timestamp: "Day 6, 08:15:00", temperature: 18.5℃, humidity: 68%, day type: 1, power load: 1650MW}. To identify the changing trend of each combination, the power load value of the current combination is compared with the previous time period "08:00:00". The load value of 1620MW is compared with the load value of 1650-1620=+30MW. Since the change is positive, the trend of this combination is marked as "rising". This calculation and marking operation is applied to each time point in the information set. For example, for the time period "18:30:00", the load value is 1950MW, while the previous time period "18:15:00" was 1980MW. The load change is 1950-1980=-30MW. The trend of this combination is marked as "falling". If the change is 0, it is marked as "stable". All data points are traversed and the trend marking is completed to generate a load change combination set.

[0027] In this embodiment of the invention, in step S200, based on the load change combination set, directional analysis is performed on each combination to identify the combination cluster with the same load change direction, and the load rise or fall interval in each cluster is marked. The frequency of each interval is calculated to obtain the load change direction cluster. In one feasible approach, based on a set of load change combinations, directional analysis is performed on each combination. The core of this analysis is to categorize combinations based on the recorded "increasing" or "decreasing" trend labels, identifying clusters of combinations with consistent load change directions. A cluster is defined as a sequence of three or more consecutive time points (i.e., at least 45 minutes) where the load change direction remains consistent. For example, if the load change trend for 10 consecutive time points from "07:00:00" to "09:15:00" on a given day is all marked as "increasing," this constitutes a cluster of combinations with increasing load. The load interval within this cluster is then labeled as an "increasing interval," and its start and end times ("07:00:00-09:15") are recorded. At the same time, this operation is performed on all sequences in the dataset that meet the conditions. For example, if another sequence is identified as having a load change trend of "decreasing" from "19:00:00" to "21:30:00", it is marked as a "decreasing interval". After completing the identification and marking of all intervals, the total number of times each specific pattern interval (such as "weekday morning peak rising interval" and "nighttime trough declining interval") appears in the entire dataset is calculated as its frequency. For example, in a week of data, the "weekday 07:00-09:30 rising interval" appears 5 times, so its frequency is 5. All identified clusters, corresponding interval labels and occurrence frequencies are summarized to obtain the load change direction cluster.

[0028] In this embodiment of the invention, in step S200, typical load change patterns are selected based on the load change direction cluster, mapped to the target time interval, and a load trend segmentation structure identifier set is generated.

[0029] In one feasible approach, representative load change patterns are selected based on the frequency of occurrence and the magnitude of load change in each pattern within the load change direction cluster. The selection criteria for representativeness are set as follows: the frequency of occurrence is not less than 4 times (the frequency of occurrence on weekdays is higher than 80% in a week's data) and the absolute value of the average load change every 15 minutes within the interval is not less than 20MW. For example, the aforementioned "rising interval from 07:00 to 09:30 on weekdays" occurs 5 times, which meets the condition. Then, the total load increment within the interval is calculated. Assume that the total increments of the 5 occurrences are 300MW, 320MW, 310MW, 290MW, and 330MW, respectively. The average total increment is 310MW, and the interval duration is 2.5 hours (10 15-minute intervals). Therefore, the average change every 15 minutes is 310 / 10 = 31MW, which is greater than 20MW. Thus, this pattern is considered representative. If an "afternoon steady decline interval" is identified, which occurs 6 times, but the average load change every 15 minutes is only -5MW, which is below the 20MW threshold, then this pattern will not be screened. All representative patterns that pass the screening, such as the "weekday morning peak rise pattern", "weekday evening peak rise pattern", and "nighttime load decline pattern", are accurately mapped to their most frequently occurring time intervals. For example, the "weekday morning peak rise pattern" is mapped to the "07:00-09:30" time period to form a structured identifier and generate a load trend segmented structure identifier set.

[0030] In this embodiment of the invention, step S300 involves acquiring multi-energy output data, matching the multi-energy output data with the direction of load changes, identifying abnormal periods based on energy proportion fluctuations, and generating a set of energy-dominant type electric carbon offset intervals according to the dominant energy source. Acquire actual power output data from wind, solar, and thermal power, compare it with the direction of power load change, perform unified time alignment on the data of different energy sources based on timestamps, match the directional relationship between power output of each energy source and load change, and generate an energy and load matching dataset. In one feasible approach, actual power output data for wind, solar, and thermal power during the research period are obtained from the power dispatch system and the new energy monitoring center, respectively. The data format is {timestamp, wind power output (MW), solar power output (MW), thermal power output (MW)}. This data is then compared with the power load change direction obtained in the preceding steps. Based on the timestamp, the output data of these three energy types undergoes the same time alignment processing as the load data, unified to a 15-minute time resolution to ensure a one-to-one correspondence between data points. Then, the directional relationship between the power output of each energy source and the load change is matched. Specifically, at each time point, the direction of power load change (increasing / decreasing / stable) is extracted, and the change in output power of wind, solar, and thermal power is simultaneously calculated. For example, in the "first" time point... At 08:30:00 on the sixth day, the load change direction is "increasing," with a change of +35MW. At the same time, the wind power output changes by +15MW, the photovoltaic output changes by +5MW, and the thermal power output changes by +15MW. Since the output changes of wind power, photovoltaic, and thermal power are all positive, consistent with the "increasing" direction of the load, the matching relationship at this time point is recorded as {Load: Increasing, Wind Power: Same Direction, Photovoltaic: Same Direction, Thermal Power: Same Direction}. If at 19:30:00 the load change direction is "decreasing," with a change of -40MW, the thermal power output changes by -50MW, and the wind power changes by +10MW, it is recorded as {Load: Decreasing, Thermal Power: Same Direction, Wind Power: Reverse Direction}. The entire dataset is traversed to complete the matching, generating an energy and load matching dataset.

[0031] In this embodiment of the invention, in step S300, based on the energy and load matching dataset, the changes in the proportion of wind power, photovoltaic power and thermal power energy in each time period are analyzed, the relationship between the fluctuation of energy proportion and the direction of load change is identified, and the fluctuation amplitude in each time period is compared and judged in combination with the set fluctuation threshold, and abnormal fluctuations exceeding the threshold are screened out to generate an abnormal energy fluctuation dataset. In one feasible approach, based on an energy and load matching dataset, the changes in the proportion of wind, solar, and thermal power in total power generation within each 15-minute time period are analyzed to identify the relationship between energy proportion fluctuations and load change directions. For example, at "Day 6, 12:00:00", the total load is 2000MW, assuming the total power generation is also 2000MW, of which wind power accounts for 150MW (7.5%), solar power 450MW (22.5%), and thermal power 1400MW (70%). In the next time period, "12:15:00", the total load rises to 2020MW, wind power becomes 155MW (7.67%), solar power becomes 480MW (23.76%), and thermal power becomes 1385MW (68.57%). The changes in the proportion of each energy source are then calculated: the proportion of wind power increases... With an increase of 0.17%, the proportion of photovoltaic power increased by 1.26%, while the proportion of thermal power decreased by 1.43%. Next, based on the set fluctuation thresholds, the fluctuation amplitude of each time period was compared and judged. The fluctuation thresholds were set based on the statistical analysis of the daily standard deviation of the proportion of each energy source in historical data, and the 85th percentile was selected as the threshold. After calculation, the threshold for the change in the proportion of wind power was set at 2%, photovoltaic at 3%, and thermal power at 2.5%. In the above example, the change in the proportion of photovoltaic power of 1.26% did not exceed the threshold of 3%, and the change in the proportion of thermal power of -1.43% did not exceed the threshold of 2.5%. Therefore, this period was not marked as abnormal fluctuation. If the change in the proportion of thermal power reaches -3.0% in another period, it will be screened as abnormal fluctuation because its absolute value exceeds the threshold of 2.5%. All the screened records are summarized to generate an energy fluctuation abnormal dataset.

[0032] In this embodiment of the invention, in step S300, the filtered time period data is classified according to the energy fluctuation anomaly dataset, the energy-dominated electric carbon offset intervals are identified and marked, and the different intervals are analyzed and corresponding classification rules are established to generate a set of energy output-dominated electric carbon offset intervals.

[0033] In one feasible approach, based on an energy fluctuation anomaly dataset, each selected time period is categorized to identify and label the electrical carbon shift intervals dominated by a single energy source such as wind, solar, or thermal power. The classification rules are as follows: within a time period marked as an anomalous fluctuation, the absolute value of the change in the proportion of each energy source is calculated, and the energy type with the largest absolute value is determined as the dominant energy source for that time period. For example, in the time period "Day 6, 18:00:00," the dataset is marked as an anomalous fluctuation. Calculations show that the change in the proportion of wind power is -0.5%, the change in the proportion of solar power is -4.5% (due to sunset), and the change in the proportion of thermal power is +5.0%. The absolute value of the change is |5.0%|>|-4.5%|>|-0.5%|. Therefore, thermal power is identified as the dominant energy source during this period, and this 15-minute period is marked as the "thermal power-dominated carbon shift interval". This is because the carbon emission intensity of thermal power is much higher than that of photovoltaic power, and its significant increase in proportion directly leads to the shift in the average carbon intensity of the power grid. Similarly, if the absolute value of the change in the proportion of wind power is the largest and exceeds the threshold during the early morning period, then this interval is marked as the "wind power-dominated carbon shift interval". This classification and labeling process is performed on all data points in the energy fluctuation anomaly dataset, and corresponding classification rules are established to generate a set of energy output-dominated carbon shift intervals. Table 2. Example of energy output-dominated carbon offset interval sets.

[0034] As shown in Table 2, this table lists some key records of the concentrated range of energy output-dominant carbon offset, demonstrating how to identify the dominant energy type for a specific period based on fluctuation anomaly determination.

[0035] In this embodiment of the invention, step S400, which involves matching the concentrated load rise and fall intervals of the load trend segmentation identifier with the dominant paths of the concentrated energy-dominant electric carbon shift intervals, optimizing the path parameters, and generating an electric carbon coupling transfer structure table, includes: Based on the load rise interval and load fall interval in the load trend segmentation structure identifier set, and combined with the wind energy dominant path in the energy output-dominant type carbon offset interval set, the load rise interval and the wind energy dominant path are matched to obtain the wind energy and load rise matching dataset. In one feasible approach, load increase intervals provided by a load trend segmentation structure identifier set, such as "weekday morning peak increase interval (07:00-09:30)," are combined with wind-dominant paths identified from the energy output-dominant carbon offset interval set, i.e., the set of time points marked as "wind-dominant." A temporal match is then performed between these two. Specifically, each 15-minute time point within "07:00-09:30" is iterated, and the label of that time point in the energy output-dominant carbon offset interval set is checked. For example, if the load is increasing at "07:15:00" and this time point is marked as "wind-dominant," a successful match is recorded. The interaction between the two is then analyzed, revealing that in this matched instance, the load... The load increased by 30MW, while wind power output increased by 25MW, and other power sources such as thermal power only supplemented by 5MW. This shows that the load increase was mainly met by the increase in wind power output. All time points in the entire "07:00-09:30" interval were matched and checked, and the proportion of time points dominated by wind power was calculated. If the proportion exceeded a preset matching degree threshold, which was determined by statistically analyzing historical matching data and taking the 70th percentile, it was set to 60%. Assuming that there were 7 time points dominated by wind power in this interval (out of a total of 10 time points, accounting for 70%), it was considered that there was a strong matching relationship between the "weekday morning peak rise interval" and the "wind power dominant path". All such successfully matched intervals and their detailed data were compiled and archived to obtain the wind power and load rise matching dataset.

[0036] In this embodiment of the invention, in step S400, the load decline interval and the thermal power dominant path are matched according to the load decline interval in the load trend segmentation structure identifier set and the thermal power dominant path in the energy output dominant type electric carbon offset interval set to obtain a thermal power and load decline matching dataset. In one feasible approach, load decline intervals are clearly defined in a segmented load trend structure, such as the "nighttime load decline interval (22:00-04:00)," and the thermal power-dominated path is defined in a set of energy output-dominant carbon offset intervals. This involves matching the load decline intervals with the thermal power-dominant path for all time points marked as "thermal power-dominant." The matching process examines 15-minute time points within the "22:00-04:00" interval, searching for the dominant energy source label in the energy output-dominant carbon offset interval set. For example, at "23:00:00," the load is declining, and this time point is marked as "thermal power-dominant," constituting a matching point. Further analysis of the interaction between the two reveals that at this time... At this point, the load decreased by 25MW, while the output of thermal power was reduced by 30MW. The output of new energy sources such as wind power remained relatively stable or increased slightly, indicating that the load reduction was mainly achieved by reducing the output of thermal power. This is a typical low-carbon dispatch behavior. This matching analysis was performed on the entire "22:00-04:00" interval, and the proportion of time points dominated by thermal power in the interval was calculated. 60% was used as the matching degree threshold. If the calculated proportion of time points dominated by thermal power in the interval reached 85%, it was determined that there was a strong matching relationship between the "nighttime load reduction interval" and the "thermal power-dominated path". All analysis and matching results, including time interval, load change details, thermal power output change details, and matching degree information, were integrated to obtain the thermal power and load reduction matching dataset.

[0037] In this embodiment of the invention, in step S400, based on the matching datasets of wind power and load increase and thermal power and load decrease, the matching relationships of each energy-dominant path are sorted out and optimized, the optimized path is constructed, and an energy-driven type electric carbon coupling transfer structure table is generated.

[0038] In one feasible approach, based on a wind power and load increase matching dataset and a thermal power and load decrease matching dataset, the matching relationships of various energy-dominant paths are systematically organized and optimized. This process summarizes the matching relationships from the two datasets to form an initial list containing multiple energy-driven paths, such as {Path 1: Morning peak increase - Wind power dominance}, {Path 2: Nighttime decrease - Thermal power dominance}, {Path 3: Evening peak increase - Thermal power dominance}, etc. Each path is optimized, with the optimization criteria being to improve the path's universality and stability. For all instances belonging to the same path (e.g., 5 instances of "morning peak increase - wind power dominance" within a week), the start and end times, load... Statistical analysis was performed on key parameters such as rate of change and energy share, calculating their mean and standard deviation. Outliers with parameters deviating from the mean by more than two standard deviations were removed. The typical characteristics of the path were redefined using the mean of the parameters of the remaining instances. For example, the optimized "morning peak rise - wind power dominance" path was defined as a time interval of "07:15-09:15", with an average load growth rate of 3% / 15min, during which the wind power generation share increased from an average of 15% to 25%. A series of optimized paths with clearly defined characteristics were constructed in this way, and these paths and their quantitative characteristic parameters were organized into a structured table to generate an energy-driven type electro-carbon coupling transfer structure table, as shown in Table 3. Table 3. Energy Driven Types and Electro-Carbon Coupling Transfer Structure

[0039] As shown in Table 3, the table clearly lists the driving paths formed by the coupling of different load change trends and the dominant energy source, and provides a quantitative description of the key parameters of each path.

[0040] In this embodiment of the invention, step S400 further includes: based on the energy driving path in the electric-carbon coupling transfer structure table, marking the dominant periods of wind power, photovoltaic power and thermal power in the daily load cycle, calculating the consistency of load fluctuation direction with external factors, classifying coupling states, and generating electric-carbon coupling state relationship results. The results of the electricity-carbon coupling state relationship include wind-dominated periods, photovoltaic-dominated periods, thermal power-dominated periods, consistency between load fluctuations and external factor fluctuations, coupling state categories, state characteristic summaries, and energy-dominated period identifiers within daily load cycles.

[0041] In this embodiment of the invention, obtaining the result of the electric-carbon coupling state relationship in step S400 includes: Based on the energy drive path in the energy drive type electro-carbon coupling transfer structure table, the time periods dominated by wind power, photovoltaic power and thermal power are marked within the daily load cycle. The dominant energy for each time period is distinguished, and the dominant energy for each time period is clearly identified by combining the time dimension, generating a dominant energy time period identifier set. Referring to Table 3, in one feasible approach, based on the energy drive paths defined in the energy drive type electro-carbon coupling transfer structure table, the system automatically marks the periods dominated by wind, solar, or thermal power within the daily load cycle. Specifically, the system uses 96 15-minute time points throughout the day as scanning targets and matches them one by one. For example, for the period from "07:15" to "09:15" on a certain workday, if the system detects that the load change trend and wind power share change are within 10% of the characteristics of drive path number P01 in the structure table (average load change rate +3.1%, wind power share change +10.2%), then the system will mark "0..." The period from 7:15 to 9:15 is marked as "wind-dominated" and the dominant energy source is clearly identified as wind. Similarly, if the actual operating data of the period from 10:00 to 14:00 matches the characteristics of the P02 path (photovoltaic-dominated), then this period is marked as "photovoltaic-dominated". For time periods not covered in the structure table, or time periods where the actual operating data deviates from the characteristics of all known paths by more than 10%, they are marked as "hybrid-driven" or "atypical period". By performing this matching and marking operation on all time periods of the day, and combining the time dimension, the dominant energy source of each time period is clearly identified, generating a dominant energy time period identifier set.

[0042] In this embodiment of the invention, in step S400, based on the dominant energy period identifier set, the load fluctuation direction and the fluctuation consistency of external factors within each period are calculated to obtain the quantitative value of the fluctuation consistency of each period and generate a fluctuation consistency dataset. In one feasible approach, based on a dominant energy period identifier set, the consistency between the direction of power load fluctuation and the fluctuation of external influencing factors (here, temperature is selected as a representative) within each marked period is calculated, and the load change for each 15-minute period is obtained. and temperature change Then, correlation analysis is performed by comparing the fluctuation trends of the two. Specifically, the direction of the two changes is quantified. If, then its direction value is +1, if The direction value is -1, if The direction value is 0, for Perform the same operation to obtain its direction value. Then, multiply the two direction values ​​to obtain the fluctuation consistency quantification value for that period. The result of this value is 1, -1, or 0, where 1 indicates fluctuation in the same direction (both increasing or decreasing), -1 indicates fluctuation in the opposite direction, and 0 indicates that at least one quantity has not changed. For example, in the period "08:00-08:15", the dominant energy source is wind power, and the load increases from 1700MW to 1730MW. =+30MW, direction=+1), the temperature increased from 19.1℃ to 19.3℃ ( =+0.2℃, direction=+1), then the fluctuation consistency value for this period is (+1)*(+1)=1. Apply this calculation to all identified periods to obtain the quantified value of fluctuation consistency for each period and generate a fluctuation consistency dataset.

[0043] In this embodiment of the invention, in step S400, the coupling state of each time period is classified according to the fluctuation consistency dataset, and the characteristics of different coupling states are identified by combining the fluctuation consistency value. The state characteristics of each time period are then summarized to generate the electric-carbon coupling state relationship result.

[0044] In one feasible approach, the coupling state of each identified dominant energy period is classified based on the quantized values ​​in the volatility consistency dataset. The classification rule combines the volatility consistency value and the dominant energy type, as follows: If the volatility consistency value is 1 and the dominant energy is wind or solar, the state is classified as "renewable energy forward coupling"; if the volatility consistency value is 1 and the dominant energy is thermal power, the state is classified as "conventional energy forward coupling"; if the volatility consistency value is -1 and the dominant energy is wind or solar, the state is classified as "renewable energy reverse coupling"; if the volatility consistency value is -1 and the dominant energy is thermal power, the state is classified as "conventional energy reverse coupling"; if the volatility consistency value is 0, it is uniformly classified as "decoupling state". For example, during the aforementioned period of "08:00-08:15", the dominant energy source is wind power, and the fluctuation consistency value is 1. Therefore, its coupling state is classified as "positive coupling of renewable energy", indicating that the growth trend of load is consistent with the rising trend of temperature, and is mainly met by clean energy. In another period of "18:30-18:45", the dominant energy source is thermal power, the load is still rising, but the temperature begins to drop, resulting in a fluctuation consistency value of -1. Its state is classified as "reverse coupling of conventional energy". The state characteristics of all periods of the day are summarized and statistically analyzed. For example, it is found that the "positive coupling of renewable energy" state lasts for 6 hours and the "positive coupling of conventional energy" state lasts for 5 hours, etc., to form the result of the electricity-carbon coupling state relationship of the power grid operation on that day.

[0045] Example 2: The above example is an illustrative scheme of a multi-dimensional evaluation method for an electro-carbon coupling system. It should be noted that the technical solution of this multi-dimensional evaluation system for an electro-carbon coupling system belongs to the same concept as the technical solution of the aforementioned multi-dimensional evaluation method for an electro-carbon coupling system. Details not described in detail in this example can be found in the description of the technical solution of the aforementioned multi-dimensional evaluation method for an electro-carbon coupling system.

[0046] This embodiment presents a multi-dimensional evaluation system for an electro-carbon coupling system, comprising: The multi-source data fusion and preprocessing module is used to collect multi-source data from the power area, preprocess the multi-source data, and construct a joint dataset of power and external influences. The load trend pattern filtering module is used to extract meteorological data, daily type and load combination for each time period in the joint data set of power and external influences, cluster combinations with the same load change direction, mark the rise and fall intervals and frequencies, filter typical patterns, and form a load trend segmentation identifier set. The energy-dominant type carbon offset identification module is used to acquire multi-energy output data, match the multi-energy output data with the direction of load change, identify abnormal periods based on energy proportion fluctuations, and generate a set of energy-dominant type carbon offset intervals according to the dominant energy. The electric carbon coupling path optimization and structure generation module is used to match the concentrated load rise and fall intervals of load trend segmentation with the dominant paths of energy-dominant electric carbon offset intervals, optimize path parameters, and generate an electric carbon coupling transfer structure table.

[0047] This embodiment also provides an electronic device applicable to multi-dimensional evaluation methods for electro-carbon coupling systems, including: The memory and processor are used to store computer-executable instructions and execute computer-executable instructions to implement the multi-dimensional evaluation method for the electric-carbon coupling system proposed in the above embodiments.

[0048] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the multi-dimensional evaluation method for the electro-carbon coupling system proposed in the above embodiments.

[0049] The storage medium proposed in this embodiment and the multi-dimensional evaluation method for realizing the electro-carbon coupling system proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0050] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0051] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-dimensional evaluation method for an electro-carbon coupling system, characterized in that, include: Collect multi-source data of the power area, preprocess the multi-source data, and construct a joint dataset of power and external influences; Extract meteorological data, daily data types, and load combinations for each time period from the combined data set of power and external influences; cluster combinations with consistent load change directions; mark the rise and fall intervals and frequencies; filter typical patterns; and form a load trend segmentation identifier set. Acquire multi-energy output data, match multi-energy output data with load change direction, identify abnormal periods based on energy proportion fluctuations, and generate energy-dominant type carbon offset interval sets according to dominant energy. The load trend segmentation identifiers are matched with the dominant paths of the energy-dominant electric carbon offset intervals, the path parameters are optimized, and an electric carbon coupling transfer structure table is generated.

2. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 1, characterized in that, The construction of the joint dataset of power and external influences includes: acquiring the regional power load sequence, temperature record value, air humidity value, working and non-working labels in the public calendar, and real-time output data of wind and photovoltaic power; detecting the timestamps of each data item; comparing differences by calling the timestamp field; adjusting the time axis format of the data and unifying it to the same time base; and generating a unified time series value. Based on the unified time series values, the spliced ​​power load, temperature, humidity, day type and wind and photovoltaic output data are compared field by field to determine missing values ​​in the fields, and values ​​from adjacent time points are called for interpolation. Abnormal fields that fail to match are deleted to generate valid spliced ​​interval values. Based on the effective splicing interval values, the values ​​of power load, temperature, humidity, day type and wind and photovoltaic output are compared time-by-time to filter out the complete time period set without anomalies, thus obtaining the joint information set of power and external influences.

3. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 2, characterized in that, Extract meteorological data, daily data types, and load combinations for each time period from the joint dataset of power and external impacts; cluster combinations with consistent load change directions; label the rise and fall intervals and frequencies; filter typical patterns; and form a load trend segmentation identifier set, including: The temperature, humidity, day type, and power load value for each time period are extracted from the combined information set of power and external influences. A set of combinations of temperature, humidity, day type, and load value for each time period is constructed, the changing trend of each combination is identified, and a set of load change combinations is generated. Based on the load change combination set, directional analysis is performed on each combination to identify the combination clusters with the same load change direction, and the load increase or decrease intervals in each cluster are marked. The frequency of each interval is calculated to obtain the load change direction clusters. Based on the load change direction cluster, typical load change patterns are selected, mapped to the target time interval, and a load trend segmentation structure identifier set is generated.

4. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 3, characterized in that, Acquire multi-energy output data, match the multi-energy output data with the direction of load changes, identify abnormal periods based on energy share fluctuations, and generate a set of energy-dominant type electricity carbon offset intervals according to the dominant energy source, including: Acquire actual power output data from wind, solar, and thermal power, compare it with the direction of power load change, perform unified time alignment on the data of different energy sources based on timestamps, match the directional relationship between power output of each energy source and load change, and generate an energy and load matching dataset. Based on the energy and load matching dataset, the changes in the proportion of wind, photovoltaic and thermal power energy in each time period are analyzed, the relationship between energy proportion fluctuation and load change direction is identified, and the fluctuation amplitude in each time period is compared and judged in combination with the set fluctuation threshold. Abnormal fluctuations exceeding the threshold are screened out, and an energy fluctuation abnormal dataset is generated. Based on the energy fluctuation anomaly dataset, the filtered time period data are classified, energy-dominated carbon offset intervals are identified and marked, and analysis is performed on different intervals to establish corresponding classification rules, thereby generating a set of energy output-dominated carbon offset intervals.

5. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 4, characterized in that, Matching the concentrated load rise and fall intervals segmented by the load trend with the dominant paths of the energy-dominant electric carbon offset intervals, optimizing path parameters, and generating an electric carbon coupling transfer structure table includes: Based on the load rise interval and load fall interval in the load trend segmentation structure identifier set, and combined with the wind energy dominant path in the energy output-dominant type carbon offset interval set, the load rise interval and the wind energy dominant path are matched to obtain the wind energy and load rise matching dataset. Based on the load decline intervals in the load trend segmentation structure identifier set and the thermal power dominant path in the energy output-dominant type electric carbon offset interval set, the load decline intervals and thermal power dominant paths are matched to obtain the thermal power and load decline matching dataset. Based on the aforementioned wind power and load increase matching dataset and thermal power and load decrease matching dataset, the matching relationships of each energy-dominant path are sorted out and optimized, the optimized path is constructed, and an energy-driven type electric carbon coupling transfer structure table is generated.

6. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 5, characterized in that, Also includes: Based on the energy drive paths in the aforementioned electricity-carbon coupling transfer structure table, the dominant periods of wind power, photovoltaic power, and thermal power are marked within the daily load cycle. The consistency between the load fluctuation direction and external factors is calculated, the coupling states are classified, and the electricity-carbon coupling state relationship results are generated. The results of the electricity-carbon coupling state relationship include wind-dominated periods, photovoltaic-dominated periods, thermal power-dominated periods, consistency between load fluctuations and external factor fluctuations, coupling state categories, state characteristic summaries, and energy-dominated period identifiers within daily load cycles.

7. The multi-dimensional evaluation method for an electro-carbon coupling system as described in claim 6, characterized in that, The acquisition of the electro-carbon coupling state relationship results includes: Based on the energy drive path in the energy drive type electro-carbon coupling transfer structure table, the time periods dominated by wind power, photovoltaic power and thermal power are marked within the daily load cycle. The dominant energy for each time period is distinguished, and the dominant energy for each time period is clearly identified by combining the time dimension, generating a dominant energy time period identifier set. Based on the dominant energy period identifier set, the load fluctuation direction and the fluctuation consistency of external factors within each period are calculated to obtain the quantitative value of fluctuation consistency for each period and generate a fluctuation consistency dataset. Based on the fluctuation consistency dataset, the coupling states of each time period are classified, and the characteristics of different coupling states are identified by combining the fluctuation consistency values. The state characteristics of each time period are then summarized to generate the electric-carbon coupling state relationship results.

8. A multi-dimensional evaluation system for an electro-carbon coupling system, applied to the method described in any one of claims 1-7, characterized in that, include: The multi-source data fusion and preprocessing module is used to collect multi-source data from the power area and preprocess the multi-source data to construct a joint dataset of power and external influences. The load trend pattern filtering module is used to extract the weather, daily type and load combination of each time period in the power and external influences joint dataset, cluster combinations with the same load change direction, mark the rise and fall intervals and frequencies, filter typical patterns, and form a load trend segmentation identifier set. The energy-dominant type carbon offset identification module is used to acquire multi-energy output data, match the multi-energy output data with the direction of load change, identify abnormal periods based on energy proportion fluctuations, and generate a set of energy-dominant type carbon offset intervals according to the dominant energy. The electric carbon coupling path optimization and structure generation module is used to match the load trend segmentation identification concentrated load rise and fall intervals with the dominant paths concentrated in the energy-dominant electric carbon offset intervals, optimize path parameters, and generate an electric carbon coupling transfer structure table.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the multi-dimensional evaluation method for the electro-carbon coupling system according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the multi-dimensional evaluation method for the electro-carbon coupling system according to any one of claims 1 to 7.

Images