A panoramic monitoring method and system for carbon emission sensitivity of power systems

Abstract

This invention relates to the field of carbon emission monitoring technology, and more particularly to a panoramic monitoring method and system for carbon emission sensitivity of power systems. The method includes the following steps: acquiring corresponding SCADA data, EMS data, and fuel management data, and performing spatiotemporal alignment processing to obtain a multi-source spatiotemporally aligned dataset of the power system; acquiring the corresponding unit fuel types within the power system and performing indirect carbon emission prediction to obtain the corresponding indirect comprehensive carbon emission intensity of the power system; determining the corresponding carbon emission influencing factors, performing significant impact screening and carbon emission sensitivity analysis to obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor; and performing panoramic sensitivity visualization based on the corresponding global carbon emission sensitivity distribution to generate a panoramic distribution map of carbon emission sensitivity for the power system. This invention enables a comprehensive analysis of the factors influencing carbon emissions in power systems, allowing for precise monitoring of carbon emission sensitivity.

Description

Technical Field

[0001] This invention relates to the field of carbon emission monitoring technology, and in particular to a panoramic monitoring method and system for carbon emission sensitivity based on power systems. Background Technology

[0002] With increasing global attention to climate change, achieving carbon reduction targets has become a shared goal for countries worldwide. As a major source of carbon emissions, the power system's accurate monitoring and effective management are crucial. Current research on power system carbon emissions primarily focuses on assessing carbon emissions at the generation and consumption ends. Most carbon emission inventories are based on power generation and national average carbon emission intensity, or a set of estimates based on boiler type and installed capacity, which introduces significant errors. Furthermore, existing research on consumption-end carbon emission intensity quantifies emissions by coupling production-end carbon emissions estimates with a lossless interconnected power grid model. However, inefficient transmission and distribution infrastructure in different countries can lead to power losses of up to 25%, undoubtedly increasing consumption-end carbon emission intensity. Existing methods do not adequately consider this factor, resulting in an underestimation of actual consumption-end carbon emission intensity. Moreover, existing technologies often only focus on monitoring carbon emissions from certain parts of the power system, lacking a comprehensive and in-depth analysis of the factors influencing carbon emissions across the entire power system, thus failing to achieve a panoramic monitoring of carbon emission sensitivity. Summary of the Invention

[0003] Therefore, the present invention needs to provide a panoramic monitoring method and system for carbon emission sensitivity of power systems to solve at least one of the above-mentioned technical problems, and to achieve a comprehensive analysis of the factors affecting carbon emissions of power systems, accurately monitor carbon emission sensitivity, and provide a scientific and reliable basis for carbon emission reduction decisions of power systems.

[0004] To achieve the above objectives, a panoramic monitoring method for carbon emission sensitivity based on power systems includes the following steps:

[0005] Step S1: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, a multi-source spatiotemporal aligned dataset of the power system corresponding to the time resolution and spatial granularity is obtained.

[0006] Step S2: Obtain the corresponding unit fuel type in the power system, and based on the corresponding unit fuel type in the power system, perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system to obtain the indirect comprehensive carbon emission intensity of the power system.

[0007] Step S3: Obtain the operating principle and carbon emission related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, screen the significant impact of each carbon emission impact factor to obtain the significant carbon emission impact factors of the power system.

[0008] Step S4: Based on the indirect comprehensive carbon emission intensity corresponding to the power system, perform carbon emission sensitivity analysis on the significant carbon emission influencing factors corresponding to the power system to obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor; perform panoramic sensitivity visualization of the power system based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor to generate a panoramic distribution map of carbon emission sensitivity of the power system.

[0009] Furthermore, step S1 includes the following steps:

[0010] Step S11: Obtain SCADA data corresponding to the power system, including operating status information, measured values ​​and control commands of fuel equipment of each unit in the power system;

[0011] Step S12: Obtain the EMS data corresponding to the power system, including the power grid operation energy parameter indicators within the power system, including power generation, load, voltage, and frequency;

[0012] Step S13: Obtain fuel management data corresponding to the power system, including fuel characteristics, fuel consumption, and combustion efficiency parameters of the power system.

[0013] Step S14: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing data cleaning processing on them to handle the corresponding missing values, outliers and inconsistent data, a multi-source standard dataset of the power system is obtained.

[0014] Step S15: Perform spatiotemporal alignment processing on the multi-source standard dataset of the power system to achieve data synchronization and alignment of each data source at the same time resolution and spatial granularity, and obtain the corresponding spatiotemporally aligned dataset of the multi-source power system at the time resolution and spatial granularity.

[0015] Furthermore, step S2 includes the following steps:

[0016] Step S21: Obtain the corresponding unit fuel type within the power system;

[0017] Step S22: Based on the corresponding unit fuel type in the power system, perform unit operation efficiency analysis on the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system to obtain the operation efficiency corresponding to each unit fuel type;

[0018] Step S23: Based on the corresponding unit fuel type in the power system, perform unit energy load factor assessment on the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system to obtain the energy load factor corresponding to each unit fuel type;

[0019] Step S24: Obtain the fuel consumption corresponding to each unit's fuel type from the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system, and perform fuel line loss assessment on the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, to obtain the fuel linear loss rate corresponding to each unit's fuel type.

[0020] Step S25: Based on the linear fuel loss rate corresponding to each unit's fuel type, and combined with the operating efficiency, energy load rate, and fuel consumption corresponding to each unit's fuel type, indirect carbon emission prediction is performed to obtain the indirect comprehensive carbon emission intensity corresponding to the power system. ,in This indicates the total number of fuel types for the generator set. Indicates the first The linear fuel loss rate corresponding to the fuel type of each unit. Indicates the first Fuel consumption corresponding to the fuel type of each unit Indicates the first Operating efficiency corresponding to the fuel type of each unit Indicates the first Energy load rate corresponding to fuel type of each unit.

[0021] Furthermore, step S22 includes the following steps:

[0022] Based on the corresponding unit fuel type in the power system, the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system is used to extract the unit operation data to obtain the operation status data corresponding to each unit fuel type.

[0023] Based on the operating status data corresponding to each unit's fuel type, an operational fluctuation analysis was conducted under different working loads to obtain the unit's operational fluctuation factor corresponding to each unit's fuel type under different working loads.

[0024] Obtain the corresponding operating environment factors, including weather, load demand, and standby unit status, and perform operating efficiency fitting calculations on the unit operation fluctuation factors corresponding to each unit fuel type under different working loads based on the operating environment factors, to obtain the operating efficiency corresponding to each unit fuel type.

[0025] Furthermore, step S23 includes the following steps:

[0026] Step S231: Based on the corresponding unit fuel type in the power system, divide the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system into operating energy parameters to obtain the power generation, load, voltage and frequency parameters corresponding to each unit fuel type;

[0027] Step S232: By using the power generation, load, voltage and frequency parameters corresponding to each unit's fuel type as the influencing factors of the unit's operating energy loss, and calculating the energy loss output of the corresponding unit's fuel type in the power system based on the corresponding influencing factors, the operating energy loss output of each unit's fuel type is obtained.

[0028] Step S233: Obtain the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, and perform constraint energy accumulation calculation on the operating energy loss output corresponding to each unit's fuel type based on the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, to obtain the cumulative increment of operating energy load corresponding to each unit's fuel type;

[0029] Step S234: Calculate the corresponding unit energy load rate based on the ratio between the operating energy loss output and the cumulative increase in operating energy load for each unit fuel type, so as to obtain the energy load rate corresponding to each unit fuel type.

[0030] Furthermore, the fuel line loss assessment of the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, as described in step S24, includes the following steps:

[0031] By acquiring the fuel characteristics and combustion efficiency of each unit's fuel type from the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system;

[0032] Based on the fuel characteristics corresponding to each unit's fuel type, an assessment of the fuel consumption efficiency impact of each unit's fuel type is conducted to obtain the fuel consumption efficiency distribution corresponding to each unit's fuel type.

[0033] Based on the combustion efficiency corresponding to each unit's fuel type, the fuel consumption efficiency distribution corresponding to each unit's fuel type is evaluated and calculated to obtain the fuel linear loss rate corresponding to each unit's fuel type.

[0034] Furthermore, step S3 includes the following steps:

[0035] Step S31: Obtain the operating principle and carbon emission-related data of the power system;

[0036] Step S32: Determine the corresponding carbon emission impact factors based on the operating principle of the power system and carbon emission-related data. These include the fuel type and fuel consumption of the generating units within the power system, the efficiency of the generating equipment, the loss of the generating lines, user electricity consumption habits indicators, and distributed energy access indicators. Among these, user electricity consumption habits indicators include peak-valley electricity consumption differences and the electricity consumption time distribution of different industries, while distributed energy access indicators include the power generation and access location of the distributed power source.

[0037] Step S33: Based on the indirect comprehensive carbon emission intensity corresponding to the power system and combined with the Pearson correlation coefficient, perform correlation assessment analysis on each carbon emission influencing factor to obtain the correlation coefficient between each carbon emission influencing factor and the carbon emission intensity in the power system.

[0038] Step S34: Based on the preset correlation threshold of 0.75, the correlation coefficients between each carbon emission influencing factor and the carbon emission intensity in the power system are compared and judged. If the absolute value of the correlation coefficient with the carbon emission intensity is greater than or equal to the preset correlation threshold of 0.75, the corresponding carbon emission influencing factor is selected as a factor with significant influence; otherwise, it is removed to obtain the significant carbon emission influencing factor corresponding to the power system.

[0039] Furthermore, step S4 includes the following steps:

[0040] Step S41: Randomly sample the significant carbon emission impact factors corresponding to the power system to generate the corresponding carbon emission impact factors. Each sample value;

[0041] Step S42: By matching the corresponding factors of each significant carbon emission impact factor Using the sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as output variables, and combining the Sobol calculation derivation method to conduct carbon emission sensitivity analysis, we obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor.

[0042] Step S43: Based on the global sensitivity distribution of carbon emissions corresponding to each significant carbon emission influencing factor, visualize the sensitivity of the power system in a panoramic view. Different colors or icons represent different sensitivity levels, and bar charts, line charts or heat maps are used to show the changing trend of the sensitivity corresponding to each significant carbon emission influencing factor over time, so as to generate a panoramic distribution map of carbon emission sensitivity for the power system.

[0043] Furthermore, the Sobol calculation derivation method described in step S42 is as follows:

[0044] By corresponding to the significant impact factors of carbon emissions Using sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as the output variable, a corresponding model is constructed using a linear fitting method. Specifically, the model is... ,in It is an input variable. It is an output variable;

[0045] The total variance of the model output is calculated. ,in Represents the mathematical expectation. Represents the variance, and can also be calculated from the first... Input variables The resulting variance ,in Indicates the first Input variables Seeking expectations, Indicates in fixed Given the given conditions, the variances of other input variables can be calculated, where the variances can be fixed using the Monte Carlo simulation method. The value of , and a series of other input variables were obtained by random sampling. Value, and at the same time based on a series The conditional variance is calculated based on the value.

[0046] Based on the total variance output by the model And by the Input variables The resulting variance Calculate the first Input variables Corresponding first-order sensitivity index And based on this first-order sensitivity index A quantitative calculation of global sensitivity is performed to obtain the global sensitivity distribution of carbon emissions corresponding to this significant carbon emission influencing factor. .

[0047] Furthermore, the present invention also provides a panoramic monitoring system for carbon emission sensitivity based on a power system, used to execute the panoramic monitoring method for carbon emission sensitivity based on a power system as described above. The panoramic monitoring system for carbon emission sensitivity based on a power system includes:

[0048] The power data spatiotemporal alignment module is used to integrate the SCADA data, EMS data and fuel management data corresponding to the power system, and perform spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, so as to obtain a multi-source spatiotemporal aligned dataset of the power system in terms of temporal resolution and spatial granularity.

[0049] The carbon emission indirect prediction module is used to obtain the corresponding unit fuel type in the power system, and to perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the corresponding unit fuel type in the power system, so as to obtain the indirect comprehensive carbon emission intensity of the power system.

[0050] The carbon emission impact screening module is used to obtain the operating principle and carbon emission-related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission-related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, the impact of each carbon emission impact factor is screened to obtain the significant carbon emission impact factors of the power system.

[0051] The sensitivity panoramic analysis module is used to perform carbon emission sensitivity analysis on the indirect comprehensive carbon emission intensity of the power system based on the significant carbon emission impact factors corresponding to the power system, and obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor; based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor, the module performs sensitivity panoramic visualization of the power system to generate a carbon emission sensitivity panoramic distribution map corresponding to the power system.

[0052] The beneficial effects of this invention are:

[0053] 1. The panoramic monitoring method for carbon emission sensitivity of power systems proposed in this invention has the following advantages over existing technologies: By integrating SCADA (Supervisory Control and Data Acquisition) data, EMS (Energy Management System) data, and fuel management data corresponding to the power system, and performing spatiotemporal alignment processing, a multi-source spatiotemporally aligned dataset is formed. The key to this process is that power system operation involves data from multiple aspects, and these data often exhibit inconsistencies in time and space, making direct correlation analysis difficult. By performing spatiotemporal alignment on these data, the temporal resolution and spatial granularity of the data can be ensured to be consistent, solving the problem of data source heterogeneity and improving the availability and accuracy of the data. This spatiotemporally aligned dataset not only provides more accurate and comprehensive information for subsequent analysis but also helps the power system identify potential correlation patterns, thereby providing a solid data foundation for realizing intelligent power system management. This also provides a prerequisite for accurately predicting the carbon emission intensity of the power system and optimizing its energy utilization efficiency, which helps improve the operational decision-making level and carbon emission management capabilities of the power system. Secondly, after obtaining the fuel types of the generating units within the power system, carbon emissions are indirectly predicted based on the system's fuel types using SCADA data, EMS data, and fuel management data from a multi-source spatiotemporally aligned dataset. This ultimately yields the indirect comprehensive carbon emission intensity of the power system. The core advantage of this step lies in accurately identifying and quantifying the carbon emission impact of different fuel types, thereby enabling precise prediction of the power system's carbon emission intensity. Different fuel types exhibit significant differences in carbon emission characteristics. Through reasonable data analysis and modeling, the contribution of different generating unit fuel types to the overall carbon emissions of the system can be revealed, resulting in more accurate and reliable carbon emission data. This step provides a basis for subsequent carbon emission optimization and policy formulation, and offers energy managers accurate emission data, enabling them to accurately grasp the changing trends of carbon emission intensity when optimizing power system operations. Then, by acquiring data on the operating principles of the power system and carbon emissions, carbon emission influencing factors are further determined based on this information. The significance of each carbon emission influencing factor is screened based on the indirect comprehensive carbon emission intensity, thereby obtaining the significant carbon emission influencing factors of the power system. These factors include unit operating conditions, fuel efficiency, ambient temperature, and other factors. Scientific screening of these factors identifies which factors have the most significant impact on carbon emissions, thus avoiding interference from irrelevant or secondary factors. Determining significant influencing factors allows for more precise and targeted subsequent carbon emission control measures, enabling a comprehensive and in-depth analysis of the carbon emission influencing factors of the entire power system.Finally, sensitivity analysis of significant carbon emission influencing factors based on indirect comprehensive carbon emission intensity is conducted. The core value of this analytical step lies in revealing the specific degree of influence of each factor on overall carbon emissions by systematically assessing the sensitivity of significant carbon emission influencing factors. Sensitivity analysis can help decision-makers understand which factors have a greater impact on changes in carbon emissions and can achieve panoramic monitoring of carbon emission sensitivity. It can provide managers with an intuitive carbon emission impact map, making it easy to quickly identify and adjust factors that have a greater impact on system carbon emissions. The power system can reduce carbon emissions more efficiently and achieve refined management from macro to micro levels, thereby promoting the operation of a more environmentally friendly and energy-efficient power system.

[0054] 2. The panoramic monitoring system for carbon emission sensitivity based on power systems proposed in this invention consists of a power data spatiotemporal alignment module, a carbon emission indirect prediction module, a carbon emission impact screening module, and a sensitivity panoramic analysis module. It can realize any panoramic monitoring method for carbon emission sensitivity based on power systems as described in this invention. It is used to combine the operations between computer programs running on each module to realize the panoramic monitoring method for carbon emission sensitivity based on power systems. The internal structure of the system cooperates with each other, which can greatly reduce repetitive work and manpower input, and can quickly and effectively provide a more accurate and efficient panoramic monitoring process for carbon emission sensitivity based on power systems, thereby simplifying the operation process of the panoramic monitoring system for carbon emission sensitivity based on power systems. Attached Figure Description

[0055] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0056] Figure 1 This is a schematic diagram of the steps of the panoramic monitoring method for carbon emission sensitivity based on the power system of the present invention;

[0057] Figure 2 for Figure 1 A detailed flowchart of step S1;

[0058] Figure 3 for Figure 1 A detailed flowchart of step S1. Detailed Implementation

[0059] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0060] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0061] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0062] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a panoramic monitoring method for carbon emission sensitivity based on power systems, the method comprising the following steps:

[0063] Step S1: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, a multi-source spatiotemporal aligned dataset of the power system corresponding to the time resolution and spatial granularity is obtained.

[0064] Step S2: Obtain the corresponding unit fuel type in the power system, and based on the corresponding unit fuel type in the power system, perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system to obtain the indirect comprehensive carbon emission intensity of the power system.

[0065] Step S3: Obtain the operating principle and carbon emission related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, screen the significant impact of each carbon emission impact factor to obtain the significant carbon emission impact factors of the power system.

[0066] Step S4: Based on the indirect comprehensive carbon emission intensity corresponding to the power system, perform carbon emission sensitivity analysis on the significant carbon emission influencing factors corresponding to the power system to obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor; perform panoramic sensitivity visualization of the power system based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor to generate a panoramic distribution map of carbon emission sensitivity of the power system.

[0067] In the embodiments of this invention, please refer to Figure 1 The diagram shown is a flowchart illustrating the steps of the panoramic monitoring method for carbon emission sensitivity based on a power system according to the present invention. In this example, the panoramic monitoring method for carbon emission sensitivity based on a power system includes the following steps:

[0068] Step S1: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, a multi-source spatiotemporal aligned dataset of the power system corresponding to the time resolution and spatial granularity is obtained.

[0069] In this embodiment of the invention, raw data is obtained from the SCADA system, EMS system and fuel management system of the power system by using data interface technology. The SCADA data includes the operating status information of each unit's fuel equipment recorded every 5 minutes, such as the coal feeder speed and coal mill current of the coal-fired unit; the EMS data collects the grid operating energy parameters at 15-minute intervals, such as the total regional power generation and load; the fuel management data is calculated on a daily basis as the fuel characteristics, consumption and combustion efficiency of each unit, such as the calorific value of coal and daily coal consumption of a certain unit. In the data cleaning stage, for outliers in the SCADA data, the normal range of main steam pressure is set to 16-18 MPa. If the measured value at a certain moment is 25 MPa, it is replaced with the average pressure value of 17 MPa before and after the measurement. For missing load data in the EMS data, linear interpolation is used to supplement it. After cleaning, spatiotemporal alignment is performed. In terms of time, the SCADA data is aggregated every 15 minutes, for example, the average speed of the coal feeder is taken every 5 minutes. In terms of space, the data is assigned to the corresponding region based on the geographical location of the substation and the unit. For example, the data of a power plant is assigned to its administrative region. Finally, a multi-source spatiotemporal aligned dataset of the power system is formed to ensure that the data is consistent in terms of time resolution (15 minutes) and spatial granularity (regional division).

[0070] Step S2: Obtain the corresponding unit fuel type in the power system, and based on the corresponding unit fuel type in the power system, perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system to obtain the indirect comprehensive carbon emission intensity of the power system.

[0071] In this embodiment of the invention, the fuel type of the generating units within the system is identified through the power system equipment archive, assuming it includes coal-fired units, gas-fired units, and oil-fired units. Based on this, data within a multi-source spatiotemporally aligned dataset is analyzed. Operating parameters for each fuel type are extracted from SCADA data, such as main steam temperature and flow rate for coal-fired units; power generation and load are obtained from EMS data; and fuel consumption and combustion efficiency are obtained from fuel management data. Based on the above data, the operating efficiency, energy load factor, fuel consumption, and fuel linear loss rate corresponding to each fuel type of the generating unit are statistically derived for indirect carbon emission prediction, using a formula... Calculate the indirect composite carbon emission intensity, where This indicates the total number of fuel types for the generator set. Indicates the first The linear fuel loss rate corresponding to the fuel type of each unit. Indicates the first Fuel consumption corresponding to the fuel type of each unit Indicates the first Operating efficiency corresponding to the fuel type of each unit Indicates the first The energy load rate corresponding to the fuel type of each generating unit is used to calculate the indirect comprehensive carbon emission intensity of the power system, thus providing a quantitative basis for the monitoring and management of carbon emissions in the power system.

[0072] Step S3: Obtain the operating principle and carbon emission related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, screen the significant impact of each carbon emission impact factor to obtain the significant carbon emission impact factors of the power system.

[0073] In this embodiment of the invention, the operating principle is obtained from power system design drawings and technical reports to clarify the power generation, transmission, and distribution processes; carbon emission-related data, such as different fuel compositions, transmission line lengths and materials, are collected from energy audit reports and equipment monitoring systems to determine carbon emission influencing factors, covering unit fuel type and consumption, power generation equipment efficiency, and line losses (formula L=I). 2 Rt, where I is current, R is resistance, and t is time; user electricity consumption habits (peak-valley electricity consumption ratio calculated from smart meter data); distributed energy access (recording distributed power generation and access points); and the Pearson correlation coefficient formula. The correlation between each factor and the indirect comprehensive carbon emission intensity was calculated, among which... For impact factor data, The mean, For carbon emission intensity, The average value is used, and a threshold of 0.75 is set. If the absolute value of the correlation coefficient of a certain factor is greater than or equal to this value, it is selected as a significant factor affecting carbon emissions. For example, the correlation coefficient between unit fuel consumption and carbon emission intensity is 0.85, which is a significant factor. On the other hand, a factor with a correlation coefficient of 0.6 is eliminated. Finally, the significant factors affecting carbon emissions corresponding to the power system are selected, such as unit fuel consumption and power generation equipment efficiency.

[0074] Step S4: Based on the indirect comprehensive carbon emission intensity corresponding to the power system, perform carbon emission sensitivity analysis on the significant carbon emission influencing factors corresponding to the power system to obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor; perform panoramic sensitivity visualization of the power system based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor to generate a panoramic distribution map of carbon emission sensitivity of the power system.

[0075] In this embodiment of the invention, by randomly sampling the significant factors affecting carbon emissions, assuming 100 samples are drawn for each factor, taking the unit's fuel consumption as an example, 100 days of daily consumption data are randomly selected from historical data, and the corresponding fuel consumption of the unit is... Using sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as the output variable, a corresponding model is constructed using a linear fitting method. Specifically, the model is... ,in It is an input variable. It is the output variable, and the total variance of the model output is calculated. ,in Represents the mathematical expectation. Represents the variance, and can also be calculated from the first... Input variables The resulting variance ,in Indicates the first Input variables Seeking expectations, Indicates in fixed Given the given conditions, the variances of other input variables can be calculated, where the variances can be fixed using the Monte Carlo simulation method. The value of , and a series of other input variables were obtained by random sampling. Value, and at the same time based on a series The conditional variance is calculated based on the value, and then the total variance is calculated based on the output of the model. And by the Input variables The resulting variance Calculate the first Input variables Corresponding first-order sensitivity index And based on this first-order sensitivity index A quantitative calculation of global sensitivity is performed to obtain the global sensitivity distribution of carbon emissions corresponding to this significant carbon emission influencing factor. Simultaneously, based on the global sensitivity distribution of carbon emissions corresponding to the significant carbon emission influencing factors, a panoramic visualization of sensitivity is performed. The sensitivity distribution is set as low sensitivity (blue) for 0-0.2, medium sensitivity (yellow) for 0.2-0.5, and high sensitivity (red) for 0.5-1. With monthly as the time unit, a bar chart is drawn to show the monthly sensitivity distribution of each factor, and a line chart shows the trend. A heat map is used to show the comprehensive sensitivity relationship of multiple factors in different months. The horizontal axis is the month, the vertical axis is the factor, and the color depth represents the sensitivity. Finally, a panoramic distribution map of carbon emission sensitivity is generated, which intuitively reflects the degree of influence and changing trend of each factor on carbon emission intensity.

[0076] Furthermore, step S1 includes the following steps:

[0077] Step S11: Obtain SCADA data corresponding to the power system, including operating status information, measured values ​​and control commands of fuel equipment of each unit in the power system;

[0078] Step S12: Obtain the EMS data corresponding to the power system, including the power grid operation energy parameter indicators within the power system, including power generation, load, voltage, and frequency;

[0079] Step S13: Obtain fuel management data corresponding to the power system, including fuel characteristics, fuel consumption, and combustion efficiency parameters of the power system.

[0080] Step S14: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing data cleaning processing on them to handle the corresponding missing values, outliers and inconsistent data, a multi-source standard dataset of the power system is obtained.

[0081] Step S15: Perform spatiotemporal alignment processing on the multi-source standard dataset of the power system to achieve data synchronization and alignment of each data source at the same time resolution and spatial granularity, and obtain the corresponding spatiotemporally aligned dataset of the multi-source power system at the time resolution and spatial granularity.

[0082] As an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps:

[0083] Step S11: Obtain SCADA data corresponding to the power system, including operating status information, measured values ​​and control commands of fuel equipment of each unit in the power system;

[0084] In this embodiment of the invention, the power system monitoring and data acquisition (SCADA) system acquires the operating status information, measured values, and control commands corresponding to the fuel equipment of each unit at a frequency of once per second. For example, for a coal-fired boiler of a thermal power unit, the system acquires operating status information such as equipment start-up and shutdown status (1 indicates operation, 0 indicates stop), coal feeder speed (unit: revolutions per minute), and coal mill current (unit: amperes). The measured values ​​include key parameters such as main steam pressure (unit: megapascals), temperature (unit: degrees Celsius), and flow rate (unit: tons per hour). Control commands include instructions to adjust the coal feed rate. During the continuous 24-hour acquisition process, a total of 86,400 sets of data were acquired, covering the fuel equipment information of 10 units in the system, ensuring the real-time and comprehensiveness of the data and providing basic operating data for subsequent analysis.

[0085] Step S12: Obtain the EMS data corresponding to the power system, including the power grid operation energy parameter indicators within the power system, including power generation, load, voltage, and frequency;

[0086] In this embodiment of the invention, the power grid operation energy parameters are collected from the Energy Management System (EMS) every 15 minutes. For power generation, the total power generation of each power plant every 15 minutes (in megawatt-hours) is recorded. For example, if three power plants in a certain area generate 200 MWh, 180 MWh, and 220 MWh respectively during a 15-minute period, the total power generation for that period is 600 MWh. Load data records the power load of each area (in megawatts), such as a city with a load of 350 megawatts during that period. Voltage parameters are collected from the bus voltage of each substation (in kilovolts), such as a 220 kV substation with a bus voltage of 225 kV. Frequency is recorded as the power grid operating frequency (in hertz), generally stable at around 50 hertz; if the measured value is 49.9 hertz at a certain moment, it is also collected. Through continuous collection, a complete sequence of power grid operation energy parameters is obtained, reflecting the overall operating status of the system.

[0087] Step S13: Obtain fuel management data corresponding to the power system, including fuel characteristics, fuel consumption, and combustion efficiency parameters of the power system.

[0088] In this embodiment of the invention, fuel characteristics, fuel consumption, and combustion efficiency parameters are obtained from the fuel management system. Regarding fuel characteristics, for coal, the received basis lower heating value (unit: kJ / kg) is obtained. Laboratory testing shows that a batch of coal has a heating value of 25,000 kJ / kg. Other parameters include ash content (percentage) and volatile matter content (percentage). Fuel consumption data records the fuel consumption of each unit per unit time, such as a unit consuming 80 tons of coal per hour. Combustion efficiency is calculated using the formula... Calculated by ×100%, where To effectively utilize heat, Let the total heat input from the fuel be the total heat of a certain unit. =1.8×10 9 kilojoules =2×10 9 kilojoules, then combustion efficiency =1.8×10 9 / 2×10 9 ×100%=90%, collect these data on a daily basis, and finally build a fuel management dataset.

[0089] Step S14: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing data cleaning processing on them to handle the corresponding missing values, outliers and inconsistent data, a multi-source standard dataset of the power system is obtained.

[0090] In this embodiment of the invention, SCADA data, EMS data, and fuel management data are integrated into the same database. For missing values, if the main steam temperature data of a unit is missing at a certain moment, linear interpolation is used to calculate and supplement it based on the temperature values ​​before and after the missing moment. For example, if the temperature is 530 degrees Celsius at the previous moment and 535 degrees Celsius at the next moment, and the missing moment is in the middle, then the supplemented value is 530 + (535 - 530) / 2 = 532.5 degrees Celsius. For abnormal values, the normal range of main steam pressure is set to 16-18 MPa. If a measured value is 25 MPa, it is judged as abnormal and replaced with the average value of 17 MPa in that period. Regarding inconsistent data, if there is a difference between the power generation of a unit recorded by the SCADA system and the power generation of the unit counted by the EMS system, the EMS system data is used for correction. Through these operations, data cleaning is completed, and finally a multi-source standard dataset of the power system is obtained.

[0091] Step S15: Perform spatiotemporal alignment processing on the multi-source standard dataset of the power system to achieve data synchronization and alignment of each data source at the same time resolution and spatial granularity, and obtain the corresponding spatiotemporally aligned dataset of the multi-source power system at the time resolution and spatial granularity.

[0092] In this embodiment of the invention, the multi-source standard dataset of the power system is spatiotemporally aligned, with a unified time resolution of 15 minutes. For the data collected per second by the SCADA system, the mean method is used for aggregation. For example, if there are 900 main steam pressure measurements within a 15-minute period, the average value is calculated as the pressure value for that 15-minute period. In terms of spatial granularity, the data from each data source is mapped to the corresponding spatial unit based on the regional division of the power grid and the location of the generating units. For example, all data from a power plant is assigned to the spatial unit of the geographical area where the power plant is located. In this way, the data from each data source is synchronously aligned at the same time resolution and spatial granularity, and finally, a multi-source spatiotemporally aligned dataset of the power system is obtained, which provides an accurate and consistent data foundation for subsequent panoramic monitoring and analysis of carbon emission sensitivity based on this dataset.

[0093] Furthermore, step S2 includes the following steps:

[0094] Step S21: Obtain the corresponding unit fuel type within the power system;

[0095] In this embodiment of the invention, the fuel type of the corresponding generating units within the power system is obtained through the power system's equipment file management system and real-time operation monitoring records. Assuming the power system contains 15 generating units, 8 are coal-fired units using standard coal as fuel; 4 are gas-fired units using natural gas as fuel; and 3 are oil-fired units using diesel fuel. These units are categorized and recorded according to fuel type, clearly defining the unit number and quantity corresponding to each fuel type. For example, coal-fired units are numbered G1-G8, gas-fired units G9-G12, and oil-fired units G13-G15, forming a clear list of unit fuel types, providing a foundation for subsequent data analysis based on different fuel types.

[0096] Step S22: Based on the corresponding unit fuel type in the power system, perform unit operation efficiency analysis on the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system to obtain the operation efficiency corresponding to each unit fuel type;

[0097] In this embodiment of the invention, based on a previously determined unit fuel type, the SCADA data corresponding to the multi-source spatiotemporal aligned dataset of the power system is used to analyze the unit's operating efficiency. At 15-minute intervals, for coal-fired units, operating parameters such as main steam pressure, temperature, flow rate, coal feeder speed, and coal mill current are extracted from the SCADA data. For example, the operating efficiency can be calculated using the corresponding heat rate method. (in For the input fuel heat, (This refers to the electrical energy output of the unit). Assume a coal-fired unit receives a certain amount of fuel heat within a 15-minute period. =2×10 9 kilojoules, outputting electrical energy =500×10 3 kilowatt-hour (1 kilowatt-hour = 3.6 × 10⁻⁶) 6 (J), then the heat consumption rate ≈1.11 kJ / kWh, operating efficiency ≈90.09%. For gas turbine units and oil turbine units, their respective key operating parameters are extracted, such as the gas intake volume, gas turbine speed, and exhaust temperature of gas turbine units, and the fuel injection volume, fuel pump pressure, and turbine power of oil turbine units. The corresponding efficiency calculation formulas are used (such as the power generation efficiency formula of gas turbine units based on the relationship between gas calorific value and power generation) to calculate the operating efficiency every 15 minutes. Finally, the operating efficiency of each fuel type of unit is statistically averaged to obtain the operating efficiency corresponding to each fuel type of unit. For example, the average operating efficiency of coal-fired units is 90%, gas turbine units are 93%, and oil turbine units are 88%.

[0098] Step S23: Based on the corresponding unit fuel type in the power system, perform unit energy load factor assessment on the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system to obtain the energy load factor corresponding to each unit fuel type;

[0099] In this embodiment of the invention, the energy load factor of generating units is assessed based on the EMS data corresponding to the multi-source spatiotemporal aligned dataset of the power system, using the unit fuel type as the basis. From the EMS data, the power generation, load, voltage, and frequency parameters corresponding to each unit fuel type are obtained in 15-minute time units. The energy load factor calculation formula is as follows: 100%, taking a coal-fired unit as an example, the cumulative increase in its operating energy load is known. =5.4718 megawatt-hours, operating energy loss output =5.45 megawatt-hours, substituting into the formula yields... ≈99.6%, for gas turbine units, if the cumulative increase in operating energy load is 2.1 MWh and the operating energy loss output is 2 MWh, then the energy load rate is approximately 99.6%. ≈95.2%; the cumulative increase in energy load of the fuel-fired unit was 1.8 MWh, the energy loss output was 1.7 MWh, and the energy load rate was... The energy load rate is approximately 94.4%. By calculating the energy load rate corresponding to each unit's fuel type, the energy utilization efficiency of the unit under various constraints can be directly reflected.

[0100] Step S24: Obtain the fuel consumption corresponding to each unit's fuel type from the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system, and perform fuel line loss assessment on the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, to obtain the fuel linear loss rate corresponding to each unit's fuel type.

[0101] In this embodiment of the invention, fuel consumption corresponding to each unit's fuel type is obtained from the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system, with a daily time period. For coal-fired units, the daily coal consumption in tons is recorded, such as a coal-fired unit consuming 800 tons of coal per day; for gas-fired units, the daily natural gas consumption in cubic meters is recorded, such as a gas-fired unit consuming 100,000 cubic meters of natural gas per day; and for oil-fired units, the daily diesel consumption in tons is recorded, such as an oil-fired unit consuming 150 tons of diesel per day. Simultaneously, fuel line loss is assessed using a fuel loss rate formula, and the fuel linear loss rate is... ,in For fuel linear loss rate, For combustion efficiency, To illustrate the fuel consumption efficiency distribution, taking a coal-fired unit as an example, the combustion efficiency of this coal-fired unit is known. =90%, fuel consumption efficiency distribution under high load =0.3, substituting into the formula, we get = (1-0.9) / 0.9 0.3≈0.033=3.3%. Similarly, calculate the fuel linear loss rate of gas turbine units and oil turbine units. For example, the fuel linear loss rate of gas turbine units is 8%, and that of oil turbine units is 12%. Finally, obtain the fuel linear loss rate corresponding to each fuel type of unit.

[0102] Step S25: Based on the linear fuel loss rate corresponding to each unit's fuel type, and combined with the operating efficiency, energy load rate, and fuel consumption corresponding to each unit's fuel type, indirect carbon emission prediction is performed to obtain the indirect comprehensive carbon emission intensity corresponding to the power system. ,in This indicates the total number of fuel types for the generator set. Indicates the first The linear fuel loss rate corresponding to the fuel type of each unit. Indicates the first Fuel consumption corresponding to the fuel type of each unit Indicates the first Operating efficiency corresponding to the fuel type of each unit Indicates the first Energy load rate corresponding to fuel type of each unit.

[0103] In this embodiment of the invention, carbon emissions are indirectly predicted based on the linear fuel loss rate corresponding to each unit's fuel type, combined with operating efficiency, energy load factor, and fuel consumption. It is assumed that the total number of unit fuel types in the power system is n=3 (coal, gas, and oil, respectively). Fuel linear loss rate corresponding to fuel type of each unit , No. Fuel consumption corresponding to the fuel type of each unit , No. Operating efficiency corresponding to fuel type of each unit , No. Energy load rate corresponding to fuel type of each unit Therefore, the specific formula for calculating the indirect comprehensive carbon emission intensity can be derived as follows: Ultimately, the indirect comprehensive carbon emission intensity corresponding to the power system is calculated, thus providing a quantitative basis for the monitoring and management of carbon emissions in the power system.

[0104] Furthermore, step S22 includes the following steps:

[0105] Based on the corresponding unit fuel type in the power system, the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system is used to extract the unit operation data to obtain the operation status data corresponding to each unit fuel type.

[0106] In this embodiment of the invention, SCADA data is extracted specifically based on the fuel type of the generating units by centralizing multi-source spatiotemporal aligned datasets in the power system. Assuming the power system contains three types of generating units: coal-fired, gas-fired, and oil-fired, data is filtered at 15-minute intervals. For coal-fired units, operating status data such as feeder speed, mill current, main steam pressure, temperature, and flow rate are extracted from the SCADA data every 15 minutes. For example, within a certain 15-minute interval, the feeder speed of a coal-fired unit is 30 rpm, the mill current is 80 amperes, the main steam pressure is 17 MPa, the temperature is 530 degrees Celsius, and the flow rate is 200 tons per hour. Similarly, for gas-fired units, data such as gas intake volume, gas turbine speed, and exhaust temperature are extracted; for oil-fired units, data such as fuel injection volume, fuel pump pressure, and turbine power are extracted. Through this operation, the corresponding operating status data sets for each type of generating unit (coal-fired, gas-fired, and oil-fired) are obtained, providing basic data for subsequent analysis.

[0107] Preferably, the operation fluctuation analysis of each unit under different working loads is performed based on the operating status data corresponding to each unit's fuel type to obtain the unit operation fluctuation factor corresponding to each unit's fuel type under different working loads;

[0108] In this embodiment of the invention, by analyzing the operating status data corresponding to the fuel type of each unit, the operating fluctuations under different working loads are analyzed to divide the working load into three intervals: low load (less than 40% of the rated load), medium load (40%-70% of the rated load), and high load (greater than 70% of the rated load). Taking a coal-fired unit as an example, in the low load interval, ten consecutive 15-minute main steam pressure data points for a certain unit are selected, namely 15.2 MPa, 15.5 MPa, 15.3 MPa, 15.4 MPa, 15.6 MPa, 15.1 MPa, 15.7 MPa, 15.3 MPa, 15.4 MPa, and 15.5 MPa, and the standard deviation formula is used. Calculate the operational volatility, where The main steam pressure value at each moment, The average value (calculated) =15.4 MPa), =10, substituting the data, we get... ≈0.19, this value is the operating fluctuation factor of the main steam pressure of the coal-fired unit under low load. Following the same method, the operating fluctuation factors of each key operating parameter of each unit under different working loads for each fuel type are calculated separately. By combining the fluctuation factors of multiple key parameters, the overall unit operating fluctuation factor corresponding to each unit fuel type under different working loads is finally obtained.

[0109] Preferably, the corresponding operating environment factors are obtained, including weather, load demand and standby unit status, and the operating efficiency is fitted and calculated based on the operating environment factors for the unit operation fluctuation factors corresponding to each unit fuel type under different working loads, so as to obtain the operating efficiency corresponding to each unit fuel type.

[0110] In this embodiment of the invention, operating environment factor data is acquired. Weather data is obtained from meteorological monitoring stations, including information such as temperature, humidity, and wind speed. For example, at a certain moment, the temperature is 25 degrees Celsius, the humidity is 60%, and the wind speed is 2 meters per second. Load demand data is obtained from the EMS system, showing real-time power load values, such as a regional power load of 800 megawatts at a certain moment. Standby unit status data is obtained from the power system dispatch management system, recording information such as the number, type, and available operating time of standby units. For example, if there are currently two gas-fired standby units that can be put into operation within 30 minutes, based on the operating environment factors, the operating efficiency is fitted and calculated for the unit operating volatility factors corresponding to different fuel types under different working loads, and a multiple linear regression model is used. ,in For operational efficiency, For unit operation volatility factors, These are weather influencing factors (calculated by weighting data such as temperature, humidity, and wind speed). The combined influencing factors are load demand and standby unit status. , , , For regression coefficients, For the error term, the model is trained using historical data to determine the regression coefficients. Taking a coal-fired unit under high load as an example, its operating volatility factor is known to be 0.25, the weather impact factor is 0.15, and the combined impact factor of load demand and standby unit status is 0.2. Substituting these values ​​into the model, the operating efficiency is calculated. =0.8-0.3×0.25+0.2×0.15-0.1×0.2=0.715. Using this method, the operating efficiency corresponding to each unit's fuel type can be calculated, providing a basis for power system operation optimization and carbon emission analysis.

[0111] Furthermore, step S23 includes the following steps:

[0112] Step S231: Based on the corresponding unit fuel type in the power system, divide the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system into operating energy parameters to obtain the power generation, load, voltage and frequency parameters corresponding to each unit fuel type;

[0113] In this embodiment of the invention, operating energy parameters are divided according to the fuel type of the generating units within the EMS data of the multi-source spatiotemporally aligned dataset of the power system. Assuming the power system includes coal-fired units, gas-fired units, and oil-fired units, with a statistical interval of 15 minutes, for coal-fired units, the total power generation (in megawatt-hours), load (in megawatts), average bus voltage (in kilovolts), and frequency (in hertz) of that type of unit are selected from the EMS data for each 15-minute interval. For example, in a certain 15-minute period, the total power generation of all coal-fired units is 500 megawatt-hours. The area under its jurisdiction has a load of 350 MW, an average bus voltage of 222 kV, and a stable frequency of 50 Hz. Similarly, for gas turbine units and oil turbine units, the power generation, load, voltage, and frequency parameters for the corresponding time periods are extracted. For example, in a 15-minute period, the gas turbine unit generates 120 MWh, has a load of 80 MW, a voltage of 110 kV, and a frequency of 49.9 Hz; the oil turbine unit generates 80 MWh, has a load of 60 MW, a voltage of 105 kV, and a frequency of 50.1 Hz. Through this operation, the EMS data is accurately divided, and the set of operating energy parameters corresponding to the fuel type of each unit is finally obtained.

[0114] Step S232: By using the power generation, load, voltage and frequency parameters corresponding to each unit's fuel type as the influencing factors of the unit's operating energy loss, and calculating the energy loss output of the corresponding unit's fuel type in the power system based on the corresponding influencing factors, the operating energy loss output of each unit's fuel type is obtained.

[0115] In this embodiment of the invention, the power generation, load, voltage, and frequency parameters corresponding to each unit's fuel type are used as influencing factors of unit operating energy loss. A multiple linear regression model is employed to calculate the energy loss output. Taking a coal-fired unit as an example, an energy loss calculation formula is constructed. ,in Energy loss output (unit: megawatt-hours). Electricity generation (megawatt-hours). Load (megawatts). Voltage (kilovolts). Frequency (Hertz) , , , , For regression coefficients, As the error term, regression coefficients were determined by training the system with operating data from the past year for this type of unit. =10, =0.01, =0.005, =-0.1, =0, in terms of specific operations, data from coal-fired power units are collected at 15-minute intervals over the past year, including power generation. ,load ,Voltage ,frequency The dataset, consisting of 15 minutes of data, includes the actual energy loss (obtainable through other precise measurement methods, such as statistically analyzing the difference between input and output energy over a period of time using energy metering equipment). Each 15-minute data point constitutes a sample. The entire dataset contains a large number of samples. The least squares method is then used to solve for the regression coefficients. The core objective of the least squares method is to find a set of regression coefficients that make the model's predicted energy loss value... To minimize the sum of squared errors between the model's predicted and actual energy loss values, the regression coefficients are continuously adjusted during the calculation process to determine the sum of squared errors between the model's predicted and actual values. in For the first The actual energy loss value of each sample For the first The predicted energy loss value calculated for each sample using the current regression coefficients. Given the sample size, the regression coefficients are gradually adjusted using an iterative optimization algorithm (such as gradient descent) along the direction of the fastest decrease in the sum of squared errors, until the sum of squared errors reaches its minimum or a preset convergence condition is met (such as the change in the sum of squared errors being less than a certain minimum value). The regression coefficients obtained at this point are... , , , , This refers to the final determined value. Taking actual data as an example, if the initial regression coefficients are set as a set of random values, after 1000 iterations, the sum of squared errors decreases from the initial 100 to 10, and the sum of squared errors changes very little in subsequent iterations. At this point, the regression coefficients... =10, =0.01, =0.005, =-0.1, =0 can be used for subsequent energy loss output calculations. If, within a certain 15-minute period, the coal-fired unit generates 500 MWh, the load is 350 MWh, the voltage is 222 kV, and the frequency is 50 Hz, substituting these values ​​into the formula yields... =10+0.01×500+0.005×350-0.1×222+0×50=10+5+1.75-22.2=-5.45 (take the absolute value as 5.45 MWh). Using the same method, calculate the operating energy loss output of the gas turbine and oil turbine units during each statistical time period.

[0116] Step S233: Obtain the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, and perform constraint energy accumulation calculation on the operating energy loss output corresponding to each unit's fuel type based on the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, to obtain the cumulative increment of operating energy load corresponding to each unit's fuel type;

[0117] In this embodiment of the invention, data on fuel cost constraints, environmental constraints, and load constraints of the power system are acquired. Regarding fuel cost constraints, the unit prices of coal, natural gas, and oil are determined, such as 800 yuan per ton for coal, 3 yuan per cubic meter for natural gas, and 6000 yuan per ton for oil. Environmental constraints are set at a system-wide carbon emission limit of 1000 tons of CO2 equivalent per day, limited by the carbon emission coefficients of each unit's fuel type (e.g., 0.8 tons of CO2 equivalent per 1 MWh generated by a coal-fired unit, 0.4 tons for a natural gas unit, and 0.6 tons for an oil-fired unit). Load constraints are determined based on regional electricity demand, such as a minimum guaranteed load of 800 MW for a certain period. Based on these constraints, the energy loss output of each unit's fuel type is calculated using a constraint energy accumulation method. Taking a coal-fired unit as an example, let its energy loss output for a certain 15 minutes be... The power generation is Based on carbon emission constraints, the maximum allowable energy loss is calculated in relation to carbon emissions. If this limit is exceeded, adjustments are made according to the constraints, using the formula... ,in For the cumulative increment of operating energy load, The fuel cost for that period. The set fuel cost ceiling, This represents the amount of carbon emissions from this unit that impose environmental constraints. To constrain the total amount of environmental resources, To actually bear the load, To ensure the minimum guaranteed load, the cumulative increment of operating energy load for gas turbine units and oil turbine units is calculated using this method.

[0118] Step S234: Calculate the corresponding unit energy load rate based on the ratio between the operating energy loss output and the cumulative increase in operating energy load for each unit fuel type, so as to obtain the energy load rate corresponding to each unit fuel type.

[0119] In this embodiment of the invention, the unit energy load rate is calculated based on the operating energy loss output and cumulative increment of operating energy load corresponding to each unit's fuel type, using the following formula: 100%, taking a coal-fired unit as an example, the cumulative increase in its operating energy load is known. =5.4718 megawatt-hours, operating energy loss output =5.45 megawatt-hours, substituting into the formula yields... ≈99.6%, for gas turbine units, if the cumulative increase in operating energy load is 2.1 MWh and the operating energy loss output is 2 MWh, then the energy load rate is approximately 99.6%. ≈95.2%; the cumulative increase in energy load of the fuel-fired unit was 1.8 MWh, the energy loss output was 1.7 MWh, and the energy load rate was... The energy load rate of each unit is approximately 94.4%. By calculating the energy load rate corresponding to each unit's fuel type, the energy utilization efficiency of the unit under various constraints is intuitively reflected, providing key indicator data for panoramic monitoring of carbon emission sensitivity and operation optimization of the power system.

[0120] Furthermore, the fuel line loss assessment of the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, as described in step S24, includes the following steps:

[0121] By acquiring the fuel characteristics and combustion efficiency of each unit's fuel type from the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system;

[0122] In this embodiment of the invention, fuel characteristics and combustion efficiency are extracted from the fuel management data of the multi-source spatiotemporally aligned dataset of the power system, according to the fuel type of the generating unit. Assuming the power system includes coal-fired units, gas-fired units, and oil-fired units, with daily statistical intervals, for coal-fired units, fuel characteristic parameters such as the received basis lower heating value (kJ / kg), ash content (percentage), and volatile matter content (percentage) of each batch of coal are obtained from the fuel management data. For example, a batch of coal has an received basis lower heating value of 24,000 kJ / kg, an ash content of 20%, and a volatile matter content of 25%. Simultaneously, the combustion efficiency of the coal-fired unit is obtained through thermal efficiency test data, calculated using the following formula: ,in To effectively utilize heat, The total heat input of the fuel is given, assuming the coal-fired unit... =1.8×10 9 kilojoules =2×10 9 kilojoules, then combustion efficiency =1.8×10 9 / 2×10 9 kilojoules × 100% = 90%. Similarly, for gas turbine units and oil turbine units, the corresponding fuel characteristic parameters such as the lower heating value and composition ratio of the fuel are extracted, as well as the combustion efficiency calculated by the same formula. For example, the combustion efficiency of the gas turbine unit is 92% and the combustion efficiency of the oil turbine unit is 88% on a certain day. Through this operation, the fuel characteristics and combustion efficiency set corresponding to each fuel type of the unit are finally obtained.

[0123] Preferably, the fuel consumption of each unit is evaluated based on the fuel characteristics corresponding to each unit's fuel type to obtain the fuel consumption efficiency distribution corresponding to each unit's fuel type.

[0124] In this embodiment of the invention, the fuel consumption efficiency impact is assessed based on the fuel characteristics corresponding to the fuel type of each unit. Taking a coal-fired unit as an example, a fuel consumption efficiency assessment model is constructed, taking into account the received lower heating value. Factors such as ash content A and volatile matter content V are calculated using the following formula: ,in This is a fuel consumption efficiency indicator. For a specific batch of coal, =24000 kJ / kg =20%, =25%, substituting into the formula, we get =2.4, calculate the fuel consumption of the coal-fired unit under different loads, such as 800 tons of coal per day under high load (greater than 70% of rated load), 600 tons per day under medium load (40%-70% of rated load), and 400 tons per day under low load (less than 40% of rated load). Combine this with fuel consumption efficiency indicators to calculate the fuel consumption efficiency distribution under different loads, using the formula... ,in Fuel consumption (unit: tons) under high load, =2.4 / 800 100 = 0.3; Under medium load, =2.4 / 600 100 = 0.4; Under low load, =2.4 / 400 100 = 0.6. Using this method, the fuel consumption efficiency distribution of gas turbine units and oil turbine units under different loads is calculated respectively.

[0125] Preferably, the fuel linear loss assessment is performed on the fuel consumption efficiency distribution corresponding to the fuel type of each unit based on the combustion efficiency corresponding to the fuel type of each unit, so as to obtain the fuel linear loss rate corresponding to the fuel type of each unit.

[0126] In this embodiment of the invention, fuel consumption efficiency distribution is assessed and calculated based on the combustion efficiency corresponding to each unit's fuel type. Taking a coal-fired unit as an example, a formula for calculating the fuel linear loss rate is constructed. ,in For fuel linear loss rate, For combustion efficiency, Given the fuel consumption efficiency distribution, the combustion efficiency of this coal-fired unit is known. =90%, fuel consumption efficiency distribution under high load =0.3, substituting into the formula, we get = (1-0.9) / 0.9 0.3 ≈ 0.033. Similarly, calculate the linear fuel loss rate of the coal-fired unit under medium and low loads. Under medium load... ≈0.044, under low load ≈0.067, for gas turbine units, if the combustion efficiency is... =92%, and the fuel consumption efficiency distribution under a certain load is 0.5, then the linear fuel loss rate is... ≈0.043; Combustion efficiency of fuel-fired power units =88%, and the fuel consumption efficiency distribution under a certain load is 0.45, then the linear fuel loss rate is... The linear fuel loss rate of each unit under different loads was calculated to be approximately 0.061, providing key indicator data for fuel management and carbon emission analysis of the power system.

[0127] Furthermore, step S3 includes the following steps:

[0128] Step S31: Obtain the operating principle and carbon emission-related data of the power system;

[0129] Step S32: Determine the corresponding carbon emission impact factors based on the operating principle of the power system and carbon emission-related data. These include the fuel type and fuel consumption of the generating units within the power system, the efficiency of the generating equipment, the loss of the generating lines, user electricity consumption habits indicators, and distributed energy access indicators. Among these, user electricity consumption habits indicators include peak-valley electricity consumption differences and the electricity consumption time distribution of different industries, while distributed energy access indicators include the power generation and access location of the distributed power source.

[0130] Step S33: Based on the indirect comprehensive carbon emission intensity corresponding to the power system and combined with the Pearson correlation coefficient, perform correlation assessment analysis on each carbon emission influencing factor to obtain the correlation coefficient between each carbon emission influencing factor and the carbon emission intensity in the power system.

[0131] Step S34: Based on the preset correlation threshold of 0.75, the correlation coefficients between each carbon emission influencing factor and the carbon emission intensity in the power system are compared and judged. If the absolute value of the correlation coefficient with the carbon emission intensity is greater than or equal to the preset correlation threshold of 0.75, the corresponding carbon emission influencing factor is selected as a factor with significant influence; otherwise, it is removed to obtain the significant carbon emission influencing factor corresponding to the power system.

[0132] As an embodiment of the present invention, reference is made to... Figure 3 As shown, Figure 1 A detailed flowchart of step S3 is shown below. In this embodiment, step S3 includes the following steps:

[0133] Step S31: Obtain the operating principle and carbon emission-related data of the power system;

[0134] In this embodiment of the invention, the operating principle and carbon emission-related data of the power system are obtained from the power system's design documents, technical manuals, and operation monitoring system. Regarding the operating principle, the power system's power generation methods (such as thermal power, hydropower, wind power, etc.), transmission network structure (including the layout and parameters of substations and transmission lines), and power distribution methods (such as by region, by user type, etc.) are clearly defined. The carbon emission-related data includes fuel consumption records of each power generation device (such as coal consumption of coal-fired units, natural gas consumption of gas-fired units, etc.), efficiency data of the power generation devices (such as generator conversion efficiency, turbine thermal efficiency, etc.), loss statistics of power generation lines (calculated through parameters such as line resistance and current), and user electricity consumption habit data (through...). Smart meters record users' electricity consumption time and volume, analyzing peak-valley electricity consumption differences and the distribution of electricity consumption time across different industries. They also collect distributed energy access data (including the type of distributed power source, its generation capacity, and access location). For example, data collected on a power system might show that coal-fired units consume 1 million tons of coal annually, gas-fired units consume 50 million cubic meters of natural gas annually, the average conversion efficiency of generators is 90%, the resistance of a transmission line is 0.1 ohms, and the average annual current is 1000 amperes. These data can be used to calculate the annual power loss of the line. Furthermore, analysis of user electricity consumption data reveals a peak-valley electricity consumption difference of 30%, that industrial users primarily consume electricity during the daytime on weekdays, and that commercial users have relatively balanced electricity demand between day and night.

[0135] Step S32: Determine the corresponding carbon emission impact factors based on the operating principle of the power system and carbon emission-related data. These include the fuel type and fuel consumption of the generating units within the power system, the efficiency of the generating equipment, the loss of the generating lines, user electricity consumption habits indicators, and distributed energy access indicators. Among these, user electricity consumption habits indicators include peak-valley electricity consumption differences and the electricity consumption time distribution of different industries, while distributed energy access indicators include the power generation and access location of the distributed power source.

[0136] In this embodiment of the invention, the corresponding carbon emission impact factors are determined based on the acquired operating principles and carbon emission-related data. For the unit fuel type and fuel consumption, the units with different fuel types (such as coal, gas, and oil) and their corresponding fuel consumption quantities are directly extracted from the collected data. The efficiency of the power generation equipment is calculated based on the technical parameters of the equipment and actual operating data. For example, the thermal efficiency of a certain coal-fired unit is 38%, and the conversion efficiency of a certain wind turbine is 40%. The power generation line loss is calculated based on parameters such as the line resistance, current, and operating time. The formula is: Line loss power = current squared × resistance × operating time. In the user electricity consumption habit index, the peak-valley electricity consumption difference is obtained by calculating the ratio of the difference in electricity consumption during peak and off-peak periods to the total electricity consumption. The electricity consumption time distribution of different industries is obtained by statistical analysis of the electricity consumption time data of users in each industry. In the distributed energy access index, the power generation corresponding to the distributed power source is directly obtained from the monitoring data of the distributed energy system, and the access location is determined through the geographic information system (GIS) or the network topology map of the power system. For example, a distributed solar power station has an annual power generation of 1 million kilowatt-hours and is located in the suburbs of a city. Through these steps, the carbon emission impact factors of the power system were comprehensively determined.

[0137] Step S33: Based on the indirect comprehensive carbon emission intensity corresponding to the power system and combined with the Pearson correlation coefficient, perform correlation assessment analysis on each carbon emission influencing factor to obtain the correlation coefficient between each carbon emission influencing factor and the carbon emission intensity in the power system.

[0138] In this embodiment of the invention, the correlation assessment of each carbon emission influencing factor is performed based on the indirect comprehensive carbon emission intensity corresponding to the power system and the Pearson correlation coefficient. The formula for calculating the Pearson correlation coefficient is as follows: ,in, and These are the observed values ​​of the two variables, and These are the averages of the two variables, respectively. To determine the number of observations, taking unit fuel consumption and carbon emission intensity as examples, assuming 12 months of unit fuel consumption and corresponding carbon emission intensity data were collected, fuel consumption is used as variable x and carbon emission intensity as variable y. The Pearson correlation coefficient between the two is calculated using the formula. Similarly, the correlation coefficients between power generation equipment efficiency, power line losses, user electricity consumption habits, and distributed energy access indicators with carbon emission intensity are calculated. For example, the correlation coefficient between unit fuel consumption and carbon emission intensity is calculated to be 0.85; the correlation coefficient between power generation equipment efficiency and carbon emission intensity is -0.78; the correlation coefficient between power line losses and carbon emission intensity is 0.65; the correlation coefficient between peak-valley electricity consumption differences and carbon emission intensity is 0.55; the correlation coefficient between electricity consumption time distribution in different industries and carbon emission intensity is 0.45; the correlation coefficient between distributed power generation and carbon emission intensity is -0.60; and the correlation coefficient between distributed power access location and carbon emission intensity is 0.35.

[0139] Step S34: Based on the preset correlation threshold of 0.75, the correlation coefficients between each carbon emission influencing factor and the carbon emission intensity in the power system are compared and judged. If the absolute value of the correlation coefficient with the carbon emission intensity is greater than or equal to the preset correlation threshold of 0.75, the corresponding carbon emission influencing factor is selected as a factor with significant influence; otherwise, it is removed to obtain the significant carbon emission influencing factor corresponding to the power system.

[0140] In this embodiment of the invention, the correlation coefficients between various carbon emission influencing factors and carbon emission intensity within the power system are compared and judged according to a preset correlation threshold of 0.75. The absolute value of each calculated correlation coefficient is compared with 0.75. If it is greater than or equal to 0.75, the corresponding carbon emission influencing factor is selected as a significant factor; otherwise, it is discarded. For example, the absolute value of the correlation coefficient between unit fuel consumption and carbon emission intensity is 0.85, which is greater than 0.75, so unit fuel consumption is selected as a significant factor; the absolute value of the correlation coefficient between power generation equipment efficiency and carbon emission intensity is 0.78, which is also greater than 0.75, so power generation equipment efficiency is also selected as a significant factor; while the absolute value of the correlation coefficient between power generation line loss and carbon emission intensity is 0.65, which is less than 0.75, so power generation line loss is discarded; the absolute value of the correlation coefficient between peak-valley electricity consumption difference and carbon emission intensity is 0.55, which is less than 0.75, so peak-valley electricity consumption difference is discarded; and the correlation coefficient between different industries' electricity consumption time distribution... The absolute value of the correlation coefficient between distributed generation and carbon emission intensity is 0.45, which is less than 0.75, so the electricity consumption time distribution of different industries was excluded. The absolute value of the correlation coefficient between distributed generation and carbon emission intensity is 0.60, which is less than 0.75, so distributed generation was excluded. The absolute value of the correlation coefficient between distributed generation access location and carbon emission intensity is 0.35, which is less than 0.75, so distributed generation access location was excluded. Finally, the significant influencing factors of carbon emissions for this power system are unit fuel consumption and power generation equipment efficiency. These significant influencing factors will provide important decision-making basis for the carbon emission control and optimization of the power system.

[0141] Furthermore, step S4 includes the following steps:

[0142] Step S41: Randomly sample the significant carbon emission impact factors corresponding to the power system to generate the corresponding carbon emission impact factors. Each sample value;

[0143] In this embodiment of the invention, random sampling is performed on identified significant factors affecting carbon emissions from the power system, such as unit fuel consumption and power generation equipment efficiency, and a sampling quantity is set. =100. A simple random sampling method is used to extract samples from historical data. Taking unit fuel consumption as an example, assuming that the historical data covers the daily consumption records of the past 3 years (a total of 1095 days), a random number generator is used to generate 100 unique random numbers in the range of 1-1095. These 100 numbers correspond to the 100 days of fuel consumption data, forming 100 sample values ​​of unit fuel consumption. For power generation equipment efficiency, 100 sample values ​​are also extracted from the daily operating efficiency records of the past 3 years based on random numbers. For example, in the 100 samples of a certain coal-fired unit, the fuel consumption values ​​are 500 tons, 520 tons, 480 tons, etc., and the power generation equipment efficiency values ​​are 35%, 37%, 36%, etc. In this way, 100 sample values ​​are generated for each significant carbon emission influencing factor, providing a data foundation for subsequent carbon emission sensitivity analysis.

[0144] Step S42: By matching the corresponding factors of each significant carbon emission impact factor Using the sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as output variables, and combining the Sobol calculation derivation method to conduct carbon emission sensitivity analysis, we obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor.

[0145] In this embodiment of the invention, the carbon emission significant impact factor is corresponding to Using sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as the output variable, a corresponding model is constructed using a linear fitting method. Specifically, the model is... ,in It is an input variable. It is the output variable, and the total variance of the model output is calculated. ,in Represents the mathematical expectation. Represents the variance, and can also be calculated from the first... Input variables The resulting variance ,in Indicates the first Input variables Seeking expectations, Indicates in fixed Given the given conditions, the variances of other input variables can be calculated, where the variances can be fixed using the Monte Carlo simulation method. The value of , and a series of other input variables were obtained by random sampling. Value, and at the same time based on a series The conditional variance is calculated based on the value, and then the total variance is calculated based on the output of the model. And by the Input variables The resulting variance Calculate the first Input variables Corresponding first-order sensitivity index And based on this first-order sensitivity index A quantitative calculation of global sensitivity is performed to obtain the global sensitivity distribution of carbon emissions corresponding to this significant carbon emission influencing factor. Based on the above derivation process, all significant carbon emission influencing factors are calculated in the same way, and finally the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor is obtained.

[0146] Step S43: Based on the global sensitivity distribution of carbon emissions corresponding to each significant carbon emission influencing factor, visualize the sensitivity of the power system in a panoramic view. Different colors or icons represent different sensitivity levels, and bar charts, line charts or heat maps are used to show the changing trend of the sensitivity corresponding to each significant carbon emission influencing factor over time, so as to generate a panoramic distribution map of carbon emission sensitivity for the power system.

[0147] In this embodiment of the invention, a panoramic visualization of the power system's sensitivity is achieved by analyzing the global sensitivity distribution of carbon emissions corresponding to each significant carbon emission influencing factor. By setting sensitivity level classification standards, the Sobol index is divided into low sensitivity levels (0-0.2, represented by blue), medium sensitivity levels (0.2-0.5, represented by yellow), and high sensitivity levels (0.5-1, represented by red). For significant influencing factors such as unit fuel consumption and power generation equipment efficiency, their sensitivity trends over time are plotted. The Sobol index is calculated from sample data within each month, and a bar chart is used to display the sensitivity values ​​of each influencing factor in different months, with the bar height representing the Sobol index magnitude. A line chart is also used to connect the sensitivity values ​​of each month, clearly presenting the changing trends. Furthermore, a heatmap is used to display the comprehensive sensitivity relationship of multiple significant influencing factors over different time periods. The horizontal axis represents time (months), and the vertical axis represents different significant carbon emission influencing factors. The depth of color represents the sensitivity level. For example, the heatmap shows that in summer months, the sensitivity of unit fuel consumption is darker, indicating a high sensitivity level, while the sensitivity of power generation equipment efficiency is relatively lighter, indicating a medium sensitivity level. By combining multiple charts and using color differentiation, a panoramic distribution map of carbon emission sensitivity for the power system is finally generated, providing an intuitive decision-making basis for power system carbon emission management.

[0148] Furthermore, the present invention also provides a panoramic monitoring system for carbon emission sensitivity based on a power system, used to execute the panoramic monitoring method for carbon emission sensitivity based on a power system as described above. The panoramic monitoring system for carbon emission sensitivity based on a power system includes:

[0149] The power data spatiotemporal alignment module is used to integrate the SCADA data, EMS data and fuel management data corresponding to the power system, and perform spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, so as to obtain a multi-source spatiotemporal aligned dataset of the power system in terms of temporal resolution and spatial granularity.

[0150] The carbon emission indirect prediction module is used to obtain the corresponding unit fuel type in the power system, and to perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the corresponding unit fuel type in the power system, so as to obtain the indirect comprehensive carbon emission intensity of the power system.

[0151] The carbon emission impact screening module is used to obtain the operating principle and carbon emission-related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission-related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, the impact of each carbon emission impact factor is screened to obtain the significant carbon emission impact factors of the power system.

[0152] The sensitivity panoramic analysis module is used to perform carbon emission sensitivity analysis on the indirect comprehensive carbon emission intensity of the power system based on the significant carbon emission impact factors corresponding to the power system, and obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor; based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor, the module performs sensitivity panoramic visualization of the power system to generate a carbon emission sensitivity panoramic distribution map corresponding to the power system.

[0153] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0154] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A panoramic monitoring method for carbon emission sensitivity based on power systems, characterized in that, Includes the following steps: Step S1: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, a multi-source spatiotemporal aligned dataset of the power system corresponding to the time resolution and spatial granularity is obtained. Step S2: Obtain the corresponding unit fuel type within the power system, and based on the corresponding unit fuel type within the power system, indirectly predict carbon emissions from the corresponding SCADA data, EMS data, and fuel management data in the multi-source spatiotemporal aligned dataset of the power system to obtain the corresponding indirect comprehensive carbon emission intensity of the power system; wherein, step S2 includes the following steps: Step S21: Obtain the corresponding unit fuel type within the power system; Step S22: Based on the corresponding unit fuel type in the power system, perform unit operation efficiency analysis on the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system to obtain the operation efficiency corresponding to each unit fuel type; Step S23: Based on the corresponding unit fuel type in the power system, perform unit energy load factor assessment on the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system to obtain the energy load factor corresponding to each unit fuel type; Step S24: Obtain the fuel consumption corresponding to each unit's fuel type from the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system, and perform fuel line loss assessment on the corresponding fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, to obtain the fuel linear loss rate corresponding to each unit's fuel type. Step S25: Based on the linear fuel loss rate corresponding to each unit's fuel type, and combined with the operating efficiency, energy load rate, and fuel consumption corresponding to each unit's fuel type, indirect carbon emission prediction is performed to obtain the indirect comprehensive carbon emission intensity corresponding to the power system. ,in This indicates the total number of fuel types for the generator set. Indicates the first The linear fuel loss rate corresponding to the fuel type of each unit. Indicates the first Fuel consumption corresponding to the fuel type of each unit Indicates the first Operating efficiency corresponding to the fuel type of each unit Indicates the first The energy load rate corresponding to the fuel type of each unit; wherein, the linear fuel loss rate is specifically... ,in For combustion efficiency, For fuel consumption efficiency distribution, the operating efficiency is specifically as follows: ,in For the input fuel heat, The energy load factor is the electrical energy output by the unit. 100%, of which The cumulative increment of operating energy load, This refers to the output of operating energy loss; Step S3: Obtain the operating principle and carbon emission related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, screen the significant impact of each carbon emission impact factor to obtain the significant carbon emission impact factors of the power system. Step S4: Based on the indirect comprehensive carbon emission intensity corresponding to the power system, perform carbon emission sensitivity analysis on the significant carbon emission influencing factors corresponding to the power system to obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor; perform panoramic sensitivity visualization of the power system based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor to generate a panoramic distribution map of carbon emission sensitivity of the power system.

2. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Obtain SCADA data corresponding to the power system, including operating status information, measured values ​​and control commands of fuel equipment of each unit in the power system; Step S12: Obtain the EMS data corresponding to the power system, including the power grid operation energy parameter indicators within the power system, including power generation, load, voltage, and frequency; Step S13: Obtain fuel management data corresponding to the power system, including fuel characteristics, fuel consumption, and combustion efficiency parameters of the power system. Step S14: By integrating the SCADA data, EMS data and fuel management data corresponding to the power system, and performing data cleaning processing on them to handle the corresponding missing values, outliers and inconsistent data, a multi-source standard dataset of the power system is obtained. Step S15: Perform spatiotemporal alignment processing on the multi-source standard dataset of the power system to achieve data synchronization and alignment of each data source at the same time resolution and spatial granularity, and obtain the corresponding spatiotemporally aligned dataset of the multi-source power system at the time resolution and spatial granularity.

3. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, Step S22 includes the following steps: Based on the corresponding unit fuel type in the power system, the corresponding SCADA data in the multi-source spatiotemporal aligned dataset of the power system is used to extract the unit operation data to obtain the operation status data corresponding to each unit fuel type. Based on the operating status data corresponding to each unit's fuel type, an operational fluctuation analysis was conducted under different working loads to obtain the unit's operational fluctuation factor corresponding to each unit's fuel type under different working loads. Obtain the corresponding operating environment factors, including weather, load demand, and standby unit status, and perform operating efficiency fitting calculations on the unit operation fluctuation factors corresponding to each unit fuel type under different working loads based on the operating environment factors, to obtain the operating efficiency corresponding to each unit fuel type.

4. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, Step S23 includes the following steps: Step S231: Based on the corresponding unit fuel type in the power system, divide the corresponding EMS data in the multi-source spatiotemporal aligned dataset of the power system into operating energy parameters to obtain the power generation, load, voltage and frequency parameters corresponding to each unit fuel type; Step S232: By using the power generation, load, voltage and frequency parameters corresponding to each unit's fuel type as the influencing factors of the unit's operating energy loss, and calculating the energy loss output of the corresponding unit's fuel type in the power system based on the corresponding influencing factors, the operating energy loss output of each unit's fuel type is obtained. Step S233: Obtain the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, and perform constraint energy accumulation calculation on the operating energy loss output corresponding to each unit's fuel type based on the fuel cost constraints, environmental constraints, and load constraints corresponding to the power system, to obtain the cumulative increment of operating energy load corresponding to each unit's fuel type; Step S234: Calculate the corresponding unit energy load rate based on the ratio between the operating energy loss output and the cumulative increase in operating energy load for each unit fuel type, so as to obtain the energy load rate corresponding to each unit fuel type.

5. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, The fuel line loss assessment in step S24, which assesses the fuel management data within the multi-source spatiotemporal aligned dataset of the power system based on the fuel consumption corresponding to each unit's fuel type, includes the following steps: By acquiring the fuel characteristics and combustion efficiency of each unit's fuel type from the corresponding fuel management data within the multi-source spatiotemporal aligned dataset of the power system; Based on the fuel characteristics corresponding to each unit's fuel type, an assessment of the fuel consumption efficiency impact of each unit's fuel type is conducted to obtain the fuel consumption efficiency distribution corresponding to each unit's fuel type. Based on the combustion efficiency corresponding to each unit's fuel type, the fuel consumption efficiency distribution corresponding to each unit's fuel type is evaluated and calculated to obtain the fuel linear loss rate corresponding to each unit's fuel type.

6. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Obtain the operating principle and carbon emission-related data of the power system; Step S32: Determine the corresponding carbon emission impact factors based on the operating principle of the power system and carbon emission-related data. These include the fuel type and fuel consumption of the generating units within the power system, the efficiency of the generating equipment, the loss of the generating lines, user electricity consumption habits indicators, and distributed energy access indicators. Among these, user electricity consumption habits indicators include peak-valley electricity consumption differences and the electricity consumption time distribution of different industries, while distributed energy access indicators include the power generation and access location of the distributed power source. Step S33: Based on the indirect comprehensive carbon emission intensity corresponding to the power system and combined with the Pearson correlation coefficient, perform correlation assessment analysis on each carbon emission influencing factor to obtain the correlation coefficient between each carbon emission influencing factor and the carbon emission intensity in the power system. Step S34: Based on the preset correlation threshold of 0.75, the correlation coefficients between each carbon emission influencing factor and the carbon emission intensity in the power system are compared and judged. If the absolute value of the correlation coefficient with the carbon emission intensity is greater than or equal to the preset correlation threshold of 0.75, the corresponding carbon emission influencing factor is selected as a factor with significant influence; otherwise, it is removed to obtain the significant carbon emission influencing factor corresponding to the power system.

7. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Randomly sample the significant carbon emission impact factors corresponding to the power system to generate the corresponding carbon emission impact factors. Each sample value; Step S42: By matching the corresponding factors of each significant carbon emission impact factor Using the sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as output variables, and combining the Sobol calculation derivation method to conduct carbon emission sensitivity analysis, we obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission influencing factor. Step S43: Based on the global sensitivity distribution of carbon emissions corresponding to each significant carbon emission influencing factor, visualize the sensitivity of the power system in a panoramic view. Different colors or icons represent different sensitivity levels, and bar charts, line charts or heat maps are used to show the changing trend of the sensitivity corresponding to each significant carbon emission influencing factor over time, so as to generate a panoramic distribution map of carbon emission sensitivity for the power system.

8. The panoramic monitoring method for carbon emission sensitivity based on power systems according to claim 7, characterized in that, The Sobol calculation derivation method described in step S42 is as follows: By corresponding to the significant impact factors of carbon emissions Using sample values ​​as input variables and the indirect comprehensive carbon emission intensity corresponding to the power system as the output variable, a corresponding model is constructed using a linear fitting method. Specifically, the model is... ,in It is an input variable. It is an output variable; The total variance of the model output is calculated. ,in Represents the mathematical expectation. Represents the variance, and can also be calculated from the first... Input variables The resulting variance ,in Indicates the first Input variables Seeking expectations, Indicates in fixed Given the given conditions, the variances of other input variables can be calculated, where the variances can be fixed using the Monte Carlo simulation method. The value of , and a series of other input variables were obtained by random sampling. Value, and at the same time based on a series The conditional variance is calculated based on the value. Based on the total variance output by the model And by the Input variables The resulting variance Calculate the first Input variables The corresponding first-order sensitivity index And based on this first-order sensitivity index A quantitative calculation of global sensitivity is performed to obtain the global sensitivity distribution of carbon emissions corresponding to this significant carbon emission influencing factor. .

9. A panoramic monitoring system for carbon emission sensitivity based on a power system, characterized in that, For executing the panoramic monitoring method for carbon emission sensitivity based on a power system as described in claim 1, the panoramic monitoring system for carbon emission sensitivity based on a power system comprises: The power data spatiotemporal alignment module is used to integrate the SCADA data, EMS data and fuel management data corresponding to the power system, and perform spatiotemporal alignment processing on the SCADA data, EMS data and fuel management data corresponding to the power system, so as to obtain a multi-source spatiotemporal aligned dataset of the power system in terms of temporal resolution and spatial granularity. The carbon emission indirect prediction module is used to obtain the corresponding unit fuel type in the power system, and to perform indirect carbon emission prediction on the corresponding SCADA data, EMS data and fuel management data in the multi-source spatiotemporal aligned dataset of the power system based on the corresponding unit fuel type in the power system, so as to obtain the indirect comprehensive carbon emission intensity of the power system. The carbon emission impact screening module is used to obtain the operating principle and carbon emission-related data of the power system, and determine the corresponding carbon emission impact factors based on the operating principle and carbon emission-related data of the power system; based on the indirect comprehensive carbon emission intensity of the power system, the impact of each carbon emission impact factor is screened to obtain the significant carbon emission impact factors of the power system. The sensitivity panoramic analysis module is used to perform carbon emission sensitivity analysis on the indirect comprehensive carbon emission intensity of the power system based on the significant carbon emission impact factors corresponding to the power system, and obtain the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor; based on the global carbon emission sensitivity distribution corresponding to each significant carbon emission impact factor, the module performs sensitivity panoramic visualization of the power system to generate a carbon emission sensitivity panoramic distribution map corresponding to the power system.

Images