A method and system for accounting-monitoring interactive verification of carbon emission factors optimization

Abstract

The application discloses a kind of accounting-monitoring interactive verification carbon emission factor optimization method and system, method includes: the mechanical equipment is divided into core equipment and auxiliary equipment;On core equipment, vehicle-mounted diagnosis terminal configured to collect mechanical equipment operating state data and exhaust sensor configured to monitor mechanical equipment direct carbon emission data are installed;For auxiliary equipment, complete daily fuel consumption record;Respectively calculated to obtain the accounting value of carbon emission data corresponding to core equipment and auxiliary equipment and the monitoring value of carbon emission data;And the ratio of monitoring value and accounting value is obtained to be used for data checking and abnormal diagnosis value;According to value, time attenuation factor and operation scene, optimize emission factor by dynamic model every week.The precision of carbon emission management is improved.

Description

Technical Field

[0001] This invention belongs to the field of carbon emission monitoring technology, and more specifically, relates to a method and system for optimizing carbon emission factors through accounting-monitoring interactive verification. Background Technology

[0002] Existing carbon emission accounting methods mostly rely on fixed emission factors, which are carbon emission coefficients per unit of energy consumption preset based on equipment model or fuel type, and then combined with energy consumption to calculate total emissions. This model is simple to operate and was widely used in the early stages of carbon emission accounting, especially suitable for scenarios with limited data collection capabilities and relatively basic accounting needs.

[0003] However, this method has revealed many limitations in practical applications. On the one hand, the fixed emission factor cannot adapt to the performance degradation of equipment after long-term use. As operating time increases, problems such as wear of internal engine components and decreased lubrication efficiency gradually emerge, directly leading to incomplete fuel combustion and a widening deviation between actual carbon emissions and theoretical values. However, the fixed emission factor always uses the initial calibration value when the equipment leaves the factory, completely ignoring the impact of performance degradation, causing the calculated results to deviate from the actual emission levels.

[0004] On the other hand, the impact of different operating scenarios on emissions is seriously overlooked. The energy efficiency of the same equipment varies significantly under different operating conditions, such as plains and mountains, high and low temperatures, and no-load and full-load conditions. For example, low-temperature environments increase engine preheating energy consumption, and high-altitude areas are affected by air pressure changes, which affect combustion efficiency. These scenario factors can cause fluctuations in the carbon emission coefficient per unit of energy consumption, and fixed factors cannot dynamically respond to such changes, making it difficult to cover diverse actual operating scenarios.

[0005] Some improvement schemes attempt to introduce manual correction coefficients, obtaining actual emission data through periodic sampling and adjusting fixed factors. However, this method relies on manual operation, which not only consumes a lot of manpower but also has a long sampling cycle and an update frequency far lower than the rate of change in operating conditions, failing to reflect high-frequency emission fluctuations in a timely manner. Furthermore, traditional correction methods often employ simple linear adjustments, failing to consider the cumulative effect over time. The emission increase caused by equipment aging is not a uniform change but rather exhibits a non-linear growth over time. Especially when equipment enters the middle and late stages of wear and tear, the emission increase may experience abrupt changes, and linear adjustments are extremely poor at capturing such situations, resulting in a severely delayed response. Summary of the Invention

[0006] This invention aims to address the problem that fixed emission factors in existing carbon emission accounting cannot adapt to equipment aging and scenario differences. It constructs a dynamic correction mechanism to address the differences in characteristics between auxiliary and core equipment. By overlaying time decay factors and scenario correction terms, the emission factors are iteratively optimized weekly, achieving high-frequency dynamic adjustments, improving accounting accuracy, providing precise data support for enterprise emission reduction strategies and environmental supervision, and contributing to the achievement of carbon neutrality goals.

[0007] To address the aforementioned deficiencies or improvement needs of existing technologies, as a first aspect of this invention, the present invention provides a carbon emission factor optimization method based on accounting-monitoring interactive verification, comprising:

[0008] S1. Based on the work schedule, the mechanical equipment is divided into core equipment and auxiliary equipment; on the core equipment, an on-board diagnostic terminal configured to collect data on the operating status of the mechanical equipment and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment are installed; for the auxiliary equipment, the daily fuel consumption is recorded through a mobile APP.

[0009] S2. Calculate the carbon emission values ​​corresponding to the core equipment and auxiliary equipment respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value;

[0010] S3. According to Value, time decay factor And the emission factors are optimized weekly through dynamic models for the operational scenarios.

[0011] Furthermore, the mobile APP in S1 includes a work check-in module, a work completion check-in module, a check-in reminder module, and a mechanical equipment operation parameter recording module;

[0012] The work check-in module is configured for administrators to select the work type and take a photo of the fuel tank markings when the equipment is started.

[0013] The end-of-work check-in module is configured for administrators to take photos of the fuel tank markings after the equipment has finished work. The system automatically calculates the daily fuel consumption based on the two photos of the fuel tank markings and the pre-entered fuel tank cross-sectional area parameters.

[0014] The check-in reminder module is configured to automatically complete workday check-in and check-out reminders, that is, to automatically send check-in reminders to the administrator no later than 10 minutes before the start of the workday and no later than 10 minutes after the end of the workday.

[0015] The mechanical equipment operation parameter recording module is configured to record data including equipment service life, cumulative operation time, most recent maintenance interval, operation type code, operation load rate, operation area temperature, and operation area altitude.

[0016] Furthermore, the specific method for determining the carbon emission data of the core equipment in S2 is as follows:

[0017] The calculated value of carbon emissions from core equipment based on the collected operational status data. Monitoring values ​​of carbon emission data from core equipment are obtained through terminal sensors. The on-board diagnostic terminal is an OBD terminal.

[0018] Carbon emission accounting value of core equipment The calculation formula is:

[0019] ,

[0020] in: The cumulative fuel consumption of the core equipment for the day is calculated by summing the instantaneous fuel consumption collected by the OBD terminal at fixed time intervals. It is the sum of the products of instantaneous fuel consumption in each time period and the duration of the corresponding time period; The initial emission factor of the equipment is the unit fuel carbon emission coefficient preset during the preparation stage based on the engine model of the equipment; The type of task is determined by the corresponding workday schedule;

[0021] Carbon emission monitoring values ​​of core equipment The calculation formula is:

[0022] ,

[0023] in, Indicates the exhaust gas at a certain moment The volume concentration is collected at high frequency by the exhaust gas sensor; It is carbon element and The mass conversion factor is used to convert the carbon content corresponding to the volume concentration to... The quality; The exhaust flow rate at that moment, combined with engine speed. Intake volume The parameters, including those included, are calculated in real time, and the calculation formula is as follows:

[0024] ,

[0025] in, The engine intake airflow is collected by the OBD terminal through the intake manifold pressure sensor and air flow meter, directly reflecting the volume of air entering the engine per unit time. Engine volumetric efficiency is a basic parameter preset during the preparation stage based on the engine model, and is dynamically adjusted with engine speed. The engine speed is collected in real time by the OBD terminal through the crankshaft position sensor; The excess air coefficient is calculated by detecting the oxygen concentration in the exhaust gas using an oxygen sensor.

[0026] Furthermore, the specific method for determining the carbon emission data of the auxiliary equipment in S2 is as follows:

[0027] For auxiliary equipment, the carbon emission data is calculated by recording fuel consumption through a mobile app. The corresponding monitoring values ​​are obtained through empirical formulas. ;

[0028] Carbon emission accounting value of auxiliary equipment The calculation formula is:

[0029] ,

[0030] in, The fuel consumption recorded by the auxiliary equipment on that day is calculated by the system based on a photo of the fuel tank scale uploaded by the administrator via the APP. The initial emission factor for this type of auxiliary equipment is the unit fuel carbon emission coefficient preset based on the equipment model during the preparation stage; Adjust the job type coefficient, which is automatically matched based on the job type selected by the administrator in the APP;

[0031] Carbon emission monitoring values ​​of auxiliary equipment The calculation formula is derived from empirical formulas and is as follows:

[0032] ,

[0033] in, This is an empirical correction factor.

[0034] Furthermore, the method for determining the empirical correction coefficient is as follows:

[0035] The empirical correction coefficient is the model prediction bias rate obtained through model learning. That is, by using historical auxiliary equipment feature data Deviation rate from corresponding historical data After training, a mathematical model is obtained. Then, using this mathematical model and new auxiliary equipment feature data, the bias rate is predicted to obtain the model prediction bias rate. ;

[0036] Let the set of characteristic variables of the auxiliary equipment be: ,in: The service life of the equipment; This refers to the cumulative work time; This is the most recent maintenance interval; Encode the job type; Average load factor; The average daily temperature of the work area; The elevation of the work area;

[0037] Label value, i.e., historical actual deviation rate for:

[0038] ,

[0039] in, The actual carbon emissions are based on manual sampling. This is the accounting value for the corresponding period;

[0040] Using gradient boosting trees as a regression model, its output prediction bias rate for:

[0041] ,

[0042] in, For the gradient boosting tree model function, by The decision trees are stacked together; during training, the loss function is determined with the goal of minimizing the mean square error between the predicted value and the label value.

[0043] For new task scenarios, the trained model is used to predict the bias rate. :

[0044] ,

[0045] in, This is the set of feature variables corresponding to the new work scenario. It is the optimal set of parameters that minimizes the loss function after training.

[0046] Furthermore, the gradient boosting tree model function is specifically as follows:

[0047] ,

[0048] in, Output the initial decision tree; For the first The output function of the tree is determined by its structure parameters. Decide; For the first The weight of each tree; This is a set of model parameters, including the initial tree, the weights of each tree, and structural parameters.

[0049] Furthermore, the specific calculation method for the loss function is as follows:

[0050] ,

[0051] in, This represents the total number of training samples; For the first Feature vectors of each sample; For the first The label value corresponding to each sample.

[0052] Furthermore, in S3 according to Value, time decay factor The specific method for optimizing emission factors weekly using a dynamic model in the operational scenario is as follows:

[0053] Calculate separately for different work scenarios Value and optimize emission factors; weekly statistics are categorized by scenario. The value is then adjusted using the scene weight allocation formula:

[0054] ,

[0055] ,

[0056] in, For the scene The initial emission factor, For the scene Weekly average value, For the scene The percentage of homework time, ; This is a time decay factor used to reflect the decay of the influence of historical data over time. The number of valid data days included in the current week; For the scene No. The sky value; It is the last day of the week; Total number of scenes;

[0057] Weekly calculation When the value is reached, through , for The standard deviation of the values ​​defines a reasonable range, and outliers that exceed the range are eliminated to avoid extreme data affecting the direction of correction.

[0058] As a second aspect of the present invention, the present invention provides a carbon emission factor optimization system for accounting-monitoring interactive verification, comprising:

[0059] The equipment classification and data acquisition unit is used to classify mechanical equipment into core equipment and auxiliary equipment according to the work schedule; on the core equipment, an on-board diagnostic terminal configured to collect mechanical equipment operating status data and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment are installed; for the auxiliary equipment, the daily fuel consumption is recorded through a mobile APP.

[0060] The accounting monitoring ratio calculation unit is used to calculate the accounting values ​​for the carbon emission data corresponding to the core equipment and auxiliary equipment, respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value;

[0061] The emission factor optimization unit is used to optimize the emission factor based on the following: Value, time decay factor And the emission factors are optimized weekly through dynamic models for the operational scenarios.

[0062] As a third aspect of the invention, the invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of any step of the aforementioned carbon emission factor optimization method of accounting-monitoring interactive verification.

[0063] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0064] 1. The carbon emission factor optimization method of the present invention, which involves calculation and monitoring interaction verification, utilizes precise equipment classification and multi-source data collection. Based on usage frequency, mechanical equipment is clearly divided into core and auxiliary equipment. For core equipment, operational status data is collected using an on-board diagnostic terminal, and direct carbon emission data is monitored using exhaust gas sensors, ensuring the accuracy and real-time nature of carbon emission data collection for core equipment. For auxiliary equipment, a mobile app allows operators to conveniently record daily fuel consumption, constructing a comprehensive data collection system adaptable to different equipment types. This lays a solid data foundation for subsequent carbon emission calculation and monitoring, making carbon emission data sources more extensive and collection more efficient, and ensuring initial data quality.

[0065] 2. The carbon emission factor optimization method of the present invention, which involves accounting and monitoring interaction verification, generates carbon emission factors through scientific accounting and monitoring. Values. Calculate the carbon emission data for both core and auxiliary equipment. Compared with monitoring values Then, by comparing the two, we can obtain... The calculated values ​​are derived from equipment operating patterns and relevant standards and specifications, conforming to theoretical emission logic; the monitoring values ​​rely on actual data collection from terminal sensors and manual records, reflecting the true emission situation. As a key indicator for data verification and anomaly diagnosis, the value can intuitively present the deviation relationship between accounting and monitoring data, provide a quantitative basis for subsequent optimization, accurately identify data anomalies, make the comparison and verification of carbon emission data more scientific, and help to discover problems in the data collection and accounting process in a timely manner.

[0066] 3. The carbon emission factor optimization method of the present invention, which involves accounting and monitoring cross-verification, ensures accuracy through dynamic optimization and adjustment of factors. Based on... Value, time decay factor In addition to the operational scenarios, the emission factors are optimized weekly using dynamic models to adapt to changes in equipment operating status and scenarios, and to respond promptly to fluctuations in actual carbon emissions. This ensures that the emission factors remain consistent with reality over long-term use, guaranteeing the accuracy and reliability of carbon emission accounting, and promoting the dynamic optimization of carbon emission factors with data and scenarios, thereby improving the precision of carbon emission management. Attached Figure Description

[0067] Figure 1 This is a flowchart of the carbon emission factor optimization method for accounting-monitoring interactive verification according to an embodiment of the present invention;

[0068] Figure 2 This is a schematic diagram of the carbon emission factor optimization method according to an embodiment of the present invention.

[0069] Figure 3 This is a system unit diagram of an embodiment of the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0071] Example 1

[0072] Please refer to Figure 1 This embodiment 1 provides a carbon emission factor optimization method based on accounting-monitoring cross-validation, including:

[0073] S1. Based on the work schedule, the mechanical equipment is divided into core equipment and auxiliary equipment; on the core equipment, an on-board diagnostic terminal configured to collect data on the operating status of the mechanical equipment and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment are installed; for the auxiliary equipment, the daily fuel consumption is recorded through a mobile APP.

[0074] S2. Calculate the carbon emission values ​​corresponding to the core equipment and auxiliary equipment respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value;

[0075] S3. According to Value, time decay factor And the emission factors are optimized weekly through dynamic models for the operational scenarios.

[0076] This embodiment 1 further elaborates on the above steps.

[0077] (1) Equipment classification and collection

[0078] This embodiment constructs a carbon emission data acquisition system by clearly defining equipment classification standards and differentiated data collection methods. Based on work scheduling plans, mechanical equipment is classified. First, various work tasks are identified, clarifying the requirements of different tasks on equipment performance, usage frequency, and monitoring needs. Equipment that plays a crucial role in key production links, is indispensable in the work process, and requires high-precision monitoring and management is classified as core equipment, directly impacting the carbon emissions and efficiency of the main production line. Auxiliary equipment, on the other hand, provides support for core operations and operates in more dispersed and diverse scenarios, such as assisting with handling and temporary resupply. Through this classification, carbon emission accounting, monitoring, and optimization mechanisms can be constructed separately for the different characteristics of core and auxiliary equipment. This allows equipment management to adapt to the production operation logic, focusing on core equipment to ensure the stability of the main line while also considering auxiliary equipment to achieve comprehensive carbon emission control. This lays a solid foundation for overall carbon emission management, making the collection and analysis of carbon emission data for various types of equipment more closely aligned with actual work scenarios, improving management accuracy and efficiency.

[0079] For core equipment, an on-board diagnostic (OBD) terminal and exhaust gas sensors are deployed for data acquisition. The OBD terminal connects to the electronic control unit of the construction machinery, reading operating parameters such as engine speed, fuel consumption rate, and load status in real time. After encryption, these parameters are uploaded to the cloud periodically, with the transmission frequency set to an interval of no more than 15 minutes. The exhaust gas sensors are installed in the exhaust emission pipeline, employing a high-precision gas detection module based on the principle of infrared absorption spectroscopy to monitor... By measuring pollutant concentrations, raw carbon emission data can be obtained directly.

[0080] The auxiliary equipment collects data via a mobile app. In a preferred embodiment, the mobile app includes a work check-in module, a work completion check-in module, a check-in reminder module, and a mechanical equipment operation parameter recording module.

[0081] The work check-in module is configured for administrators to select the work type (the work type is preset via a drop-down menu, including excavation, transportation, etc.) and take a photo of the fuel tank level when the equipment starts up.

[0082] The end-of-work check-in module is configured for administrators to take photos of the fuel tank markings after the equipment has finished work. The system automatically calculates the daily fuel consumption based on the two photos of the fuel tank markings and the pre-entered fuel tank cross-sectional area parameters.

[0083] The check-in reminder module is configured to automatically complete workday check-in and check-out reminders, that is, to automatically send check-in reminders to the administrator no later than 10 minutes before the start of the workday and no later than 10 minutes after the end of the workday.

[0084] The mechanical equipment operation parameter recording module is configured to record data including equipment service life, cumulative operation time, most recent maintenance interval, operation type code, operation load rate, operation area temperature, and operation area altitude.

[0085] By combining professional terminal data collection from core equipment with lightweight data collection via an auxiliary APP, carbon emission data from devices at different frequencies can be acquired.

[0086] (2) Calculation of monitoring ratio

[0087] Please refer to Figure 2 The carbon emission data corresponding to the core equipment and auxiliary equipment were calculated separately. Monitoring values ​​of carbon emission data Then, the calculation is used for data validation and anomaly diagnosis. value;

[0088] ,

[0089] The value reflects the deviation between calculated and actual emissions, providing a basis for verification and diagnosis. A significant deviation from 1 indicates potential problems with the calculation logic or sensors. This design reduces human intervention, comprehensively considers influencing factors, and provides support for carbon emission management.

[0090] For core equipment, the accounting value The fuel consumption is accumulated through the on-board terminal, and corrected for by the preset emission factor corresponding to the engine model and the type of operation; [Monitored Value] From the exhaust gas sensor Concentration is determined by combining real-time exhaust flow rate (dependent on sensor parameters) and conversion factor;

[0091] In a preferred embodiment, the specific method for determining the carbon emission data of the core equipment is as follows:

[0092] The calculated value of carbon emissions from core equipment based on the collected operational status data. Monitoring values ​​of carbon emission data from core equipment are obtained through terminal sensors. The on-board diagnostic terminal is an OBD terminal.

[0093] The formula for calculating the carbon emission accounting value E1 of core equipment is as follows:

[0094] ,

[0095] in: The cumulative fuel consumption of the core equipment for the day is calculated by summing the instantaneous fuel consumption collected by the OBD terminal at fixed time intervals. It is the sum of the products of instantaneous fuel consumption in each time period and the duration of the corresponding time period; The initial emission factor of the equipment is the unit fuel carbon emission coefficient preset during the preparation stage based on the engine model of the equipment; The type of task is determined by the corresponding workday schedule;

[0096] Carbon emission monitoring values ​​of core equipment The calculation formula is:

[0097] ,

[0098] in, Indicates the exhaust gas at a certain moment The volume concentration is collected at high frequency by the exhaust gas sensor; It is carbon element and The mass conversion factor is used to convert the carbon content corresponding to the volume concentration to... The quality; The exhaust flow rate at that moment, combined with engine speed. Intake volume The parameters, including those included, are calculated in real time, and the calculation formula is as follows:

[0099] ,

[0100] in, The engine intake airflow is collected by the OBD terminal through the intake manifold pressure sensor and air flow meter, directly reflecting the volume of air entering the engine per unit time. Engine volumetric efficiency is a basic parameter preset during the preparation stage based on the engine model, and is dynamically adjusted with engine speed. The engine speed is collected in real time by the OBD terminal through the crankshaft position sensor; The excess air coefficient is calculated by detecting the oxygen concentration in the exhaust gas using an oxygen sensor.

[0101] The auxiliary equipment's calculated values ​​are derived from fuel consumption recorded by the app, preset emission factors, and work type correction coefficients. Monitored values ​​are derived from the calculated values ​​and empirical correction coefficients. The calculation of these values ​​is tailored to the characteristics of the auxiliary equipment: fuel consumption is calculated based on the fuel tank's graduation difference, adapting to its dispersed and low-frequency use without requiring additional hardware; emission factors are consistent with those of the core equipment, ensuring data benchmark consistency; and the work type correction coefficients are calibrated based on historical data, adapting to diverse work scenarios and ensuring the calculated values ​​reasonably reflect emission levels. The empirical correction coefficients for the monitored values ​​are predicted by a gradient boosting tree model. The model inputs characteristic variables such as the equipment's service life and uses historical actual deviation rates as training data. In the absence of real-time monitoring hardware, accuracy is improved by correlating historical data with operating conditions.

[0102] In a preferred embodiment, the specific method for determining the carbon emission data of the auxiliary equipment is as follows:

[0103] For auxiliary equipment, the carbon emission data is calculated by recording fuel consumption through a mobile app. The corresponding monitoring values ​​are obtained through empirical formulas. ;

[0104] Carbon emission accounting value of auxiliary equipment The calculation formula is:

[0105] ,

[0106] in, The fuel consumption recorded by the auxiliary equipment on that day is calculated by the system based on the fuel tank scale photo uploaded by the administrator through the APP (that is, the difference between the fuel volume corresponding to the fuel tank scale on that day and the fuel volume on the scale of the previous day). The initial emission factor for this type of auxiliary equipment is a unit fuel carbon emission coefficient preset during the preparation stage based on the equipment model (such as forklift, loader) (consistent with the benchmark value of the same type of fuel as the core equipment, such as diesel corresponding to the same basic coefficient). The adjustment factor for the job type is automatically matched based on the job type selected by the administrator in the APP (such as short-distance transportation or cargo loading and unloading). Different job types correspond to different adjustment ratios, which are calibrated based on historical data.

[0107] The carbon emission monitoring value E2 for auxiliary equipment is derived through an empirical formula, which is as follows:

[0108] ,

[0109] in, This is an empirical correction factor.

[0110] In a preferred embodiment, the empirical correction coefficient is determined as follows:

[0111] The empirical correction coefficient is the model prediction bias rate obtained through model learning. That is, by using historical auxiliary equipment feature data Deviation rate from corresponding historical data After training, a mathematical model is obtained. Then, using this mathematical model and new auxiliary equipment feature data, the bias rate is predicted to obtain the model prediction bias rate. ;

[0112] Let the set of characteristic variables of the auxiliary equipment be: ,in: The service life of the equipment; This refers to the cumulative work time; This is the most recent maintenance interval; Encode the job type; Average load factor; The average daily temperature of the work area; The elevation of the work area;

[0113] Label value, i.e., historical actual deviation rate for:

[0114] ,

[0115] in, The actual carbon emissions are based on manual sampling. This is the accounting value for the corresponding period;

[0116] Using gradient boosting trees as a regression model, its output prediction bias rate for:

[0117] ,

[0118] in, For the gradient boosting tree model function, by The decision trees are stacked together; during training, the loss function is determined with the goal of minimizing the mean square error between the predicted value and the label value.

[0119] For new task scenarios, the trained model is used to predict the bias rate. :

[0120] ,

[0121] in, This is the set of feature variables corresponding to the new work scenario. It is the optimal set of parameters that minimizes the loss function after training.

[0122] In a preferred embodiment, the gradient boosting tree model function is specifically as follows:

[0123] ,

[0124] in, Output the initial decision tree; For the first The output function of the tree is determined by its structure parameters. Decide; For the first The weight of each tree; This is a set of model parameters, including the initial tree, the weights of each tree, and structural parameters.

[0125] In a preferred embodiment, the loss function is specifically calculated as follows:

[0126] ,

[0127] in, This represents the total number of training samples; For the first Feature vectors of each sample; For the first The label value corresponding to each sample.

[0128] This approach comprehensively considers equipment characteristics and operating condition differences, accurately monitors core equipment, and efficiently manages auxiliary equipment. It ensures data quality while controlling costs, providing reliable support for enterprises' carbon emission accounting, emission reduction strategy formulation, and industry supervision, and achieving full equipment coverage and refined management.

[0129] (3) Emission factor optimization

[0130] Furthermore, the weekly optimization of emission factors is conducted according to operational scenarios: first, scenarios are eliminated. Outliers exceeding the mean plus or minus two standard deviations are considered, and the weekly average is calculated by combining these outliers with a time decay factor. Values ​​(recent data has a greater impact), and finally based on the initial emission factor and average. The value and the percentage of operation time (scenario weight) are used to derive new emission factors through a dynamic model.

[0131] This approach makes optimization more realistic: removing outliers avoids extreme data from misleading the correction direction; classifying by scenario ensures that the emission characteristics of different operating conditions are accurately captured; the time decay factor highlights the reference value of recent data, enabling emission factors to respond promptly to changes in equipment status; and scenario weighting allows corrections of high-proportion scenarios to have a greater impact on the overall results, which is consistent with the actual emission structure.

[0132] In a preferred embodiment, according to Value, time decay factor The specific method for optimizing emission factors weekly using a dynamic model in the operational scenario is as follows:

[0133] Calculate separately for different work scenarios Value and optimize emission factors; weekly statistics are categorized by scenario. The value is then adjusted using the scene weight allocation formula:

[0134] ,

[0135] ,

[0136] in, For the scene The initial emission factor, For the scene Weekly average value, For the scene The percentage of homework time, ; This is the time decay factor, used to reflect the decay of the influence of historical data over time (values ​​range from 0 to 1, such as 0.8); The number of valid data days included in the current week; For the scene No. The sky value; It is the last day of the week; Total number of scenes;

[0137] Weekly calculation When the value is reached, through , for The standard deviation of the values ​​defines a reasonable range, and outliers that exceed the range are eliminated to avoid extreme data affecting the direction of correction.

[0138] It ensures both the real-time nature of emission factors and the reliability of data, providing an accurate benchmark for carbon emission accounting and supporting enterprises in dynamically monitoring their emission status and formulating effective emission reduction strategies.

[0139] Example 2

[0140] Please refer to Figure 3 This embodiment 2 provides a carbon emission factor optimization system with accounting-monitoring interactive verification, including:

[0141] The equipment classification and data acquisition unit is used to classify mechanical equipment into core equipment and auxiliary equipment according to usage frequency; on the core equipment, an on-board diagnostic terminal configured to collect mechanical equipment operating status data and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment are installed; for auxiliary equipment, daily fuel consumption is recorded through a mobile APP.

[0142] The accounting monitoring ratio calculation unit is used to calculate the accounting values ​​for the carbon emission data corresponding to the core equipment and auxiliary equipment, respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value;

[0143] The emission factor optimization unit is used to optimize the emission factor based on the following: Value, time decay factor Furthermore, the emission factor is optimized weekly using a dynamic model for the operational scenario; and the time decay factor is adjusted monthly by verifying data consistency. This serves as a baseline to ensure the accuracy of emission factors.

[0144] Example 3

[0145] This embodiment 3 also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement any step of a carbon emission factor optimization method involving accounting-monitoring interaction verification.

[0146] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0147] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0148] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing carbon emission factors through accounting-monitoring cross-validation, characterized in that, include: S1. Based on the work schedule, the mechanical equipment is divided into core equipment and auxiliary equipment; The core equipment is equipped with an on-board diagnostic terminal configured to collect data on the operating status of the mechanical equipment, and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment; for auxiliary equipment, daily fuel consumption is recorded via a mobile app. S2. Calculate the carbon emission values ​​corresponding to the core equipment and auxiliary equipment respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value; S3. According to Value, time decay factor And the emission factors are optimized weekly through dynamic models for different work scenarios; The specific method for determining the carbon emission data of auxiliary equipment is as follows: For auxiliary equipment, the carbon emission data is calculated by recording fuel consumption through a mobile app. The corresponding monitoring values ​​are obtained through empirical formulas. ; Carbon emission accounting value of auxiliary equipment The calculation formula is: ， in, The fuel consumption recorded by the auxiliary equipment on that day is calculated by the system based on a photo of the fuel tank scale uploaded by the administrator via the APP. The initial emission factor for this type of auxiliary equipment is the unit fuel carbon emission coefficient preset based on the equipment model during the preparation stage; Adjust the job type coefficient, which is automatically matched based on the job type selected by the administrator in the APP; The carbon emission monitoring value E2 for auxiliary equipment is derived through an empirical formula, which is as follows: ， in, This is an empirical correction factor; The method for determining the empirical correction coefficient is as follows: The empirical correction coefficient is the model prediction bias rate obtained through model learning. Specifically, by using historical auxiliary equipment feature data Deviation rate from corresponding historical data After training, a mathematical model is obtained. Then, using this mathematical model and new auxiliary equipment feature data, the bias rate is predicted to obtain the model prediction bias rate. ; Let the set of characteristic variables of the auxiliary equipment be: ,in: The service life of the equipment; This refers to the cumulative work time; This is the most recent maintenance interval; Encode the job type; Average load factor; The average daily temperature of the work area; The elevation of the work area; Label value, i.e., historical actual deviation rate for: ， in, The actual carbon emissions are based on manual sampling. This is the accounting value for the corresponding period; Using gradient boosting trees as a regression model, its output prediction bias rate for: ， in, For the gradient boosting tree model function, by The decision trees are stacked together; during training, the loss function is determined with the goal of minimizing the mean square error between the predicted value and the label value. For new task scenarios, the trained model is used to predict the bias rate. : ， in, This is the set of feature variables corresponding to the new work scenario. It is the optimal set of parameters that minimizes the loss function after training.

2. The carbon emission factor optimization method based on accounting-monitoring interactive verification according to claim 1, characterized in that, The mobile APP in S1 includes a work check-in module, a work completion check-in module, a check-in reminder module, and a mechanical equipment operation parameter recording module; The work check-in module is configured for administrators to select the work type and take a photo of the fuel tank markings when the equipment is started. The end-of-work check-in module is configured for administrators to take photos of the fuel tank markings after the equipment has finished work. The system automatically calculates the daily fuel consumption based on the two photos of the fuel tank markings and the pre-entered fuel tank cross-sectional area parameters. The check-in reminder module is configured to automatically complete workday check-in and check-out reminders, and automatically send check-in reminders to the administrator no later than 10 minutes before the start of the workday and no later than 10 minutes after the end of the workday. The mechanical equipment operation parameter recording module is configured to record data including equipment service life, cumulative operation time, most recent maintenance interval, operation type code, operation load rate, operation area temperature, and operation area altitude.

3. The carbon emission factor optimization method based on accounting-monitoring interactive verification according to claim 1, characterized in that, The specific method for determining the carbon emission data of the core equipment in S2 is as follows: The calculated value of carbon emissions from core equipment based on the collected operational status data. Monitoring values ​​of carbon emission data from core equipment are obtained through terminal sensors. The on-board diagnostic terminal is an OBD terminal. The formula for calculating the carbon emission accounting value E1 of core equipment is as follows: ， in: The cumulative fuel consumption of the core equipment for the day is calculated by summing the instantaneous fuel consumption collected by the OBD terminal at fixed time intervals. Specifically, it refers to the sum of the instantaneous fuel consumption for each time period and the corresponding time period duration; The initial emission factor of the equipment is the unit fuel carbon emission coefficient preset during the preparation stage based on the engine model of the equipment; This is a correction factor for the job type, determined through the corresponding workday job schedule; Carbon emission monitoring values ​​of core equipment The calculation formula is: ， in, It represents the volume concentration of CO2 in the exhaust gas at a certain moment, which is collected by the exhaust gas sensor at a high frequency. It is the mass conversion factor between carbon and CO2, used to convert the carbon content corresponding to the volume concentration into the mass of CO2. The exhaust flow rate at that moment, combined with engine speed. Intake volume The parameters, including those included, are calculated in real time, and the calculation formula is as follows: ， in, The engine intake airflow is collected by the OBD terminal through the intake manifold pressure sensor and air flow meter, directly reflecting the volume of air entering the engine per unit time. Engine volumetric efficiency is a basic parameter preset during the preparation stage based on the engine model, and is dynamically adjusted with engine speed. The engine speed is collected in real time by the OBD terminal through the crankshaft position sensor; The excess air coefficient is calculated by detecting the oxygen concentration in the exhaust gas using an oxygen sensor.

4. The carbon emission factor optimization method based on accounting-monitoring interactive verification according to claim 1, characterized in that, The gradient boosting tree model function is specifically as follows: ， in, Output the initial decision tree; For the first The output function of the tree is determined by its structure parameters. Decide; For the first The weight of each tree; This is a set of model parameters, including the initial tree, the weights of each tree, and structural parameters.

5. The carbon emission factor optimization method based on accounting-monitoring interactive verification according to claim 1, characterized in that, The specific calculation method for the loss function is as follows: ， in, This represents the total number of training samples; For the first Feature vectors of each sample; For the first The label value corresponding to each sample.

6. The carbon emission factor optimization method based on accounting-monitoring interactive verification according to claim 1, characterized in that, According to the S3 Value, time decay factor The specific method for optimizing emission factors weekly using a dynamic model in the operational scenario is as follows: Calculate separately for different work scenarios Value and optimize emission factors; weekly statistics are categorized by scenario. The value is then adjusted using the scene weight allocation formula: ， ， in, For the scene The initial emission factor, For the scene Weekly average value, For the scene The percentage of homework time, ; This is a time decay factor used to reflect the decay of the influence of historical data over time. The number of valid data days included in the current week; For the scene No. The sky value; It is the last day of the week; Total number of scenes; Weekly calculation When the value is reached, through , for The standard deviation of the values ​​defines a reasonable range, and outliers that exceed the range are eliminated to avoid extreme data affecting the direction of correction.

7. A carbon emission factor optimization system based on accounting-monitoring cross-validation, characterized in that, include: The equipment classification and data acquisition unit is used to classify mechanical equipment into core equipment and auxiliary equipment according to the work arrangement plan; The core equipment is equipped with an on-board diagnostic terminal configured to collect data on the operating status of the mechanical equipment, and an exhaust gas sensor configured to monitor the direct carbon emission data of the mechanical equipment; for auxiliary equipment, daily fuel consumption is recorded via a mobile app. The accounting monitoring ratio calculation unit is used to calculate the accounting values ​​for the carbon emission data corresponding to the core equipment and auxiliary equipment, respectively. Monitoring values ​​of carbon emission data And calculate the monitoring values. With accounting value ratio Obtain data for validation and anomaly diagnosis. value; The emission factor optimization unit is used to optimize the emission factor based on the following: Value, time decay factor And the emission factors are optimized weekly through dynamic models for different work scenarios; The specific method for determining the carbon emission data of auxiliary equipment is as follows: For auxiliary equipment, the carbon emission data is calculated by recording fuel consumption through a mobile app. The corresponding monitoring values ​​are obtained through empirical formulas. ; Carbon emission accounting value of auxiliary equipment The calculation formula is: ， in, The fuel consumption recorded by the auxiliary equipment on that day is calculated by the system based on a photo of the fuel tank scale uploaded by the administrator via the APP. The initial emission factor for this type of auxiliary equipment is the unit fuel carbon emission coefficient preset based on the equipment model during the preparation stage; Adjust the job type coefficient, which is automatically matched based on the job type selected by the administrator in the APP; The carbon emission monitoring value E2 for auxiliary equipment is derived through an empirical formula, which is as follows: ， in, This is an empirical correction factor; The method for determining the empirical correction coefficient is as follows: The empirical correction coefficient is the model prediction bias rate obtained through model learning. Specifically, by using historical auxiliary equipment feature data Deviation rate from corresponding historical data After training, a mathematical model is obtained. Then, using this mathematical model and new auxiliary equipment feature data, the bias rate is predicted to obtain the model prediction bias rate. ; Let the set of characteristic variables of the auxiliary equipment be: ,in: The service life of the equipment; This refers to the cumulative work time; This is the most recent maintenance interval; Encode the job type; Average load factor; The average daily temperature of the work area; The elevation of the work area; Label value, i.e., historical actual deviation rate for: ， in, The actual carbon emissions are based on manual sampling. This is the accounting value for the corresponding period; Using gradient boosting trees as a regression model, its output prediction bias rate for: ， in, For the gradient boosting tree model function, by The decision trees are stacked together; during training, the loss function is determined with the goal of minimizing the mean square error between the predicted value and the label value. For new task scenarios, the trained model is used to predict the bias rate. : ， in, This is the set of feature variables corresponding to the new work scenario. It is the optimal set of parameters that minimizes the loss function after training.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor as a carbon emission factor optimization method based on accounting-monitoring interactive verification as described in any one of claims 1-6.

Images