Dynamic tracking and early warning method and system for intelligent carbon emission

Abstract

The invention discloses an intelligent dynamic tracking and early warning method and system for carbon emission, and relates to the technical field of carbon emission management.The method comprises the steps that a digital twinborn model of a target area is constructed, grid division is conducted based on monitoring precision, and a first layer of fixed monitoring points are arranged; planning a mobile monitoring path in combination with a carbon emission attention fitting result, and configuring a second layer of mobile monitoring points; collecting data through two layers of monitoring points, and establishing a carbon emission data network; performing airflow simulation based on a digital twinborn model, performing traceability inversion in combination with a data network, and positioning an emission source; and finally, equipment-level positioning and carbon emission tracking early warning are realized through equipment matching analysis. The technical problems that in the prior art, the carbon emission monitoring coverage range is limited, the traceability is insufficient, and accurate positioning and real-time early warning of the carbon emission source are difficult to achieve are solved, and the technical effects of high-precision traceability positioning and real-time traceability early warning of the carbon emission source are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission management, and particularly to a method and system for dynamically tracking and warning intelligent carbon emissions. Background Art

[0002] The demand for enterprise carbon emission supervision is increasing day by day. The traditional data collection method relying on fixed monitoring equipment can no longer meet the dynamic control requirements of the current complex urban emission environment. The current monitoring means have problems such as insufficient distribution density of monitoring points, many coverage blind spots, slow abnormal response, lack of emission path tracking and equipment traceability, etc., resulting in the inability to detect, accurately locate and effectively handle emission problems in a timely manner. Especially in industrial parks, high-density emission areas or non-point source scenarios, emission behaviors have significant spatial diffusivity and temporal dynamics, and it is difficult to achieve full coverage and accurate identification only by static monitoring. Summary of the Invention

[0003] The present application provides a method and system for dynamically tracking and warning intelligent carbon emissions, which are used to solve the technical problems in the prior art that the coverage of carbon emission monitoring is limited, the traceability ability is insufficient, and it is difficult to achieve accurate positioning and real-time warning of carbon emission sources.

[0004] In the first aspect of the present application, a method for dynamically tracking and warning intelligent carbon emissions is provided. The method includes: performing digital twin modeling of a target area, and after configuring grid granularity according to monitoring accuracy, performing grid segmentation of the target area, and distributing first-level fixed monitoring points according to the grid segmentation result; performing carbon emission concern fitting of the target area by using the digital twin model, establishing a mobile monitoring path according to the carbon emission concern fitting result, and configuring second-level mobile monitoring points through the mobile monitoring path; after monitoring the target area by the first-level fixed monitoring points and the second-level mobile monitoring points, establishing a carbon emission data network; performing airflow simulation in the target area based on the digital twin model, performing traceability inversion according to the airflow simulation result and the carbon emission data network, and establishing a traceability inversion result; performing equipment matching analysis in the target area according to the traceability inversion result, establishing an equipment positioning result, and establishing a carbon emission tracking warning according to the equipment positioning result.

[0005] In the second aspect of the present application, a dynamic tracking and early warning system for intelligent carbon emissions is provided. The system includes: a fixed monitoring point distribution module, which is used to perform digital twin modeling of the target area, perform grid segmentation of the target area after configuring the grid granularity according to the monitoring accuracy, and distribute the first-level fixed monitoring points according to the grid segmentation results; a mobile monitoring point configuration module, which is used to perform carbon emission concern fitting of the target area by using the digital twin model, establish a mobile monitoring path according to the carbon emission concern fitting result, and configure the second-level mobile monitoring points through the mobile monitoring path; a carbon emission data network establishment module, which is used to establish a carbon emission data network after the target area is monitored by the first-level fixed monitoring points and the second-level mobile monitoring points; a traceability inversion module, which is used to perform airflow simulation in the target area based on the digital twin model, perform traceability inversion according to the airflow simulation result and the carbon emission data network, and establish a traceability inversion result, including: inputting the carbon emission data network into a feature extraction channel, performing feature extraction, and establishing a feature extraction result, including: a: reading the carbon concentration values at consecutive time nodes for each position point, calculating the time derivative of the carbon concentration value, and establishing a local time series gradient feature; b: constructing a spatial gradient field by spatial interpolation of the local time series gradient features of all position points, and establishing a spatial gradient distribution feature; c: performing gradient change feature extraction according to the local time series gradient feature and the spatial gradient distribution feature, and establishing a key gradient feature, and outputting the key gradient feature as the feature extraction result; performing traceability inversion according to the feature extraction result and the airflow simulation result; a carbon emission tracking and early warning module, which is used to perform equipment matching analysis in the target area according to the traceability inversion result, establish an equipment positioning result, and establish a carbon emission tracking and early warning according to the equipment positioning result.

[0006] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The dynamic tracking and early warning method and system for intelligent carbon emissions provided in the present application relate to the technical field of carbon emission management. By constructing a virtual environment of the target area through digital twin modeling, dividing grids according to the monitoring accuracy and arranging fixed monitoring points, generating a mobile monitoring path in combination with carbon emission concern fitting, forming a carbon emission data network covering the whole area, then performing traceability inversion based on airflow simulation and the data network, and matching equipment calibration data to achieve source positioning, finally realizing dynamic tracking and early warning of carbon emissions. It solves the technical problems in the prior art that the coverage of carbon emission monitoring is limited, the traceability ability is insufficient, and it is difficult to accurately locate and real-time early warn carbon emission sources, and realizes the technical effect of high-precision traceability positioning and real-time tracking and early warning of carbon emission sources by constructing a multi-level monitoring network and introducing an airflow simulation and emission feature matching mechanism. Description of the Drawings

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0008] Figure 1 It is a schematic flowchart of the dynamic tracking and early warning method for intelligent carbon emissions provided by the embodiments of the present application; Figure 2 It is a schematic structural diagram of the dynamic tracking and early warning system for intelligent carbon emissions provided by the embodiments of the present application.

[0009] Explanation of reference numerals: Fixed monitoring point distribution module 11, mobile monitoring point configuration module 12, carbon emission data network establishment module 13, traceability inversion module 14, carbon emission tracking and early warning module 15. Specific embodiments

[0010] The present application provides a dynamic tracking and early warning method and system for intelligent carbon emissions, which is used to solve the technical problems in the prior art that the coverage of carbon emission monitoring is limited and the traceability ability is insufficient, and it is difficult to achieve accurate positioning and real-time early warning of carbon emission sources.

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0012] It should be noted that the terms "first", "second", etc. in the description and accompanying drawings of the present application are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products, or devices.

[0013] Embodiment 1, as Figure 1 shown, the present application provides a dynamic tracking and early warning method for intelligent carbon emissions, and the method includes: P10: After performing digital twin modeling of the target area and configuring the grid granularity according to the monitoring accuracy, perform grid segmentation of the target area, and distribute the first-level fixed monitoring points according to the grid segmentation results.

[0014] Specifically, first carry out digital twin modeling of the target area, that is, based on the basic information such as the building layout, terrain and landform, meteorological characteristics, existing monitoring data and carbon emission source types of the target area, through a combination of physical modeling and data-driven models, construct a virtual area model with high simulation accuracy and dynamic response capabilities. This digital twin model supports the continuous access and feedback of real-time monitoring data, can dynamically map the spatio-temporal evolution characteristics of carbon emission behaviors in the target area, and provides basic support for subsequent monitoring path planning, emission fitting and inversion.

[0015] After completing the digital twin modeling, it is necessary to perform grid granularity configuration operations according to the requirements of the carbon concentration perception accuracy for the actual monitoring task. Grid granularity refers to the fineness of dividing grids in the digital twin model, which directly affects the accuracy of subsequent monitoring data and the computational complexity. This operation can determine the division size and shape of the grid by analyzing factors such as the spatial scale of the target area, the expected monitoring accuracy, and the distribution density of carbon emission sources. For example, equilateral regular grids (such as squares or hexagons) or irregular adaptive grids are used. The specific granularity size is positively correlated with the regional complexity - for example, in densely populated, traffic-intensive or carbon source-intensive areas, smaller granularity should be used to enhance the monitoring resolution; while in open or areas with stable carbon emission behaviors, larger granularity can be used to save resources.

[0016] After determining the granularity, further perform grid segmentation, that is, divide the target area into several spatial units according to the set grid granularity. This operation can be assisted by a GIS (Geographic Information System) to achieve automatic alignment and adaptation with the regional physical landform. After segmentation, select appropriate monitoring positions based on grid centers, boundary intersection points, and typical scenario points (such as main road intersections, industrial equipment concentration areas), and distribute the first-level fixed monitoring points. These monitoring points are immovable sensing terminals deployed in the area, usually with the function of real-time detection of carbon gas concentrations such as carbon dioxide and methane, and at the same time integrated with communication modules (such as LoRa, NB-IoT or 5G modules) and positioning modules (such as GPS) for data backhaul and device status synchronization.

[0017] In addition, to improve the coverage and timeliness of monitoring data, the layout of fixed monitoring points can be optimized by combining grid positions with the meteorological simulation results of the digital twin model. For example, the distribution density of monitoring points can be increased in the downwind direction or airflow convergence areas to enhance the sensitivity of source identification. Through the above operations, the first-level fixed monitoring network is constructed, forming a static and distributed carbon emission information collection infrastructure for the target area, providing basic data support and a spatial reference framework for subsequent high-precision and dynamic carbon emission tracking and early warning.

[0018] P20: Use the digital twin model to perform carbon emission concern fitting for the target area, establish a mobile monitoring path based on the carbon emission concern fitting result, and configure the second-level mobile monitoring points through the mobile monitoring path.

[0019] Furthermore, step P20 of the embodiment of the present application further includes: P21: Read the fixed monitoring data set of the first-level fixed monitoring points; P22: Perform carbon emission clustering on the fixed monitoring data set to generate a carbon emission clustering result; P23: Execute the monitoring coverage weakness analysis in the carbon emission clustering result to establish a real-time weakness concern; P24: Generate a real-time location concern based on the carbon emission clustering result; P25: Establish the mobile monitoring path after compensating according to the real-time weakness concern, the real-time location concern, and the carbon emission concern fitting result.

[0020] It should be understood that based on the existing digital twin model, carbon emission concern fitting for the target area is carried out. This fitting process takes real-time and dynamically updated environmental parameters as inputs, including the carbon emission concentration of fixed monitoring points, regional wind speed and direction, temperature and humidity, terrain structure, building occlusion characteristics, and potential emission source lists, etc. By modeling the spatio-temporal evolution trend of carbon concentration in the area, the abnormally concentrated carbon emission areas and their boundaries are identified, and a dynamic spatial concern map of carbon emission hotspots is established. Based on this fitting result, an efficient mobile monitoring mechanism can be further constructed to flexibly supplement the measurement of areas that cannot be fully covered or timely responded to by the fixed monitoring network, thereby improving the response range and time resolution of the overall monitoring system.

[0021] During the specific execution process, first read and organize the fixed monitoring data set uploaded by the first-level fixed monitoring points. This data set covers the carbon gas concentration collected by each monitoring point within a preset time window (such as , Environmental meteorological parameters (such as wind speed, wind direction, temperature, humidity, etc.), along with corresponding spatial coordinates and timestamps, are used to form a multi-dimensional monitoring data set with geographical tags and time series characteristics. Subsequently, clustering analysis operations are performed on this data set. Algorithms such as K-Means, hierarchical clustering, and DBSCAN are used to divide the target area into several sub-regions with similar carbon emission levels based on spatial adjacency and carbon concentration similarity, generating a carbon emission clustering result. This clustering result not only reveals the local characteristics and trend boundaries of carbon emissions within the region but also provides data support for subsequent identification of monitoring blind spots and areas of concern.

[0022] Furthermore, an analysis of the weakness of monitoring coverage is carried out for each sub-region, that is, the clustering boundary range is superimposed and compared with the coverage range of the current fixed monitoring points, and the relationship between the monitoring point density and the carbon concentration change rate within each sub-region is calculated. If the carbon concentration fluctuates greatly while the monitoring points are sparse in a certain clustering area, then this area is marked as a weak monitoring coverage area, denoted as the real-time weak concern area. At the same time, representative coordinate points are extracted at the clustering center, the peak position of the carbon concentration, and the frequently fluctuating boundaries to construct real-time position concern points.

[0023] Furthermore, to establish a mobile monitoring path that meets the requirements of supplementary measurement, based on the generated carbon emission concern fitting result, the real-time weak concern area and the real-time position concern points are integrated, and a path is constructed through a path planning algorithm. During the path generation process, factors such as monitoring priority, spatial coverage efficiency, path length, monitoring frequency, geographical accessibility (such as whether it is a no-fly zone for drones or a ground obstacle area) are comprehensively considered for path optimization to obtain an optimized mobile monitoring path map. This path is used to guide mobile monitoring devices (such as drones, inspection robots, or portable devices) to cruise, stay, or loop for sampling in sequence within a specified time period, and a second layer of mobile monitoring points is configured at key points of the path or in areas with high-frequency changes to collect high-frequency and high-spatial-resolution carbon emission data, realizing the dynamic reinforcement of the fixed monitoring system.

[0024] Furthermore, step P25 of the embodiment of the present application further includes: P25-1: Compensate the carbon emission concern fitting result using the real-time weak concern and the real-time position concern to establish a collection position set, and the collection position set is marked with position importance and position time limit concern; P25-2: Configure the maximum number of mobile monitoring devices; P25-3: Within the range of the maximum number of devices, execute the mobile monitoring devices to perform collection adaptation optimization for the collection position set to generate a collection adaptation optimization result; P25-4: Establish the mobile monitoring path according to the collection adaptation optimization result.

[0025] Optionally, on the basis of performing mobile monitoring path planning, a fusion compensation mechanism for multi-source attention results and a path adaptation mechanism under resource constraints can be introduced to achieve priority coverage and efficient sampling scheduling for key carbon emission areas under the constraint of limited monitoring device resources, and improve the executability of path generation and the overall operation efficiency of the system.

[0026] First, using the previously generated real-time weak attention areas and real-time location focus points, perform spatial and temporal compensation processing on the carbon emission attention fitting results obtained through the digital twin model to establish a set of collection locations. This set of collection locations refers to a group of spatial points that are identified by the system as needing key supplementary measurement or dynamic coverage, and each location is assigned a location importance identifier and a location time limit attention identifier. Among them, the location importance is used to characterize the contribution of the sampling point to the inversion of the carbon emission state of the entire region, and is related to the historical concentration volatility, clustering weight, or downstream influencing factors of this location; the location time limit attention indicates that the data collection at this location is time-sensitive, such as rapidly changing emission hotspots or areas greatly affected by meteorological interference, and needs to be completed within a specific time window to ensure data validity and timeliness.

[0027] Subsequently, according to the current schedulable monitoring resource status, configure the limit number of mobile monitoring devices, that is, set the total number of mobile monitoring terminals that can participate in task scheduling simultaneously during this monitoring cycle. This limit number can be dynamically adjusted according to the remaining battery power, operation radius, task priority, or scheduling strategy of the device, and is input into the subsequent path optimization model as a constraint condition.

[0028] Under the premise of limited device quantity, perform a collection adaptation optimization process on the set of collection locations, that is, under the constraint condition of the limit number of devices, use multi-objective optimization algorithms such as heuristic search algorithms (such as ), integer linear programming models, genetic algorithms, particle swarm optimization, etc., to jointly solve the allocation relationship between each mobile device and the collection points, the path coverage path, the energy consumption cost, and the time window matching degree, etc., to generate a collection adaptation optimization result. This result clearly indicates the list of points to be covered, the sampling order, and the path trajectory of each device during the task cycle, and while satisfying the path length limit and device capacity limit, maximally achieve the coverage of high-weight collection points.

[0029] Finally, based on the acquisition adaptation optimization results, one or more actually executable mobile monitoring paths are established and output. This path not only meets the objective requirements such as the shortest path, optimal time, or full coverage, but also has deployability characteristics such as high device capability matching and strong data timeliness. This path will be sent to each mobile terminal device in real time and can be dynamically adjusted or adaptively optimized in combination with real-time feedback data during the execution process to build a mobile carbon emission monitoring mechanism with high efficiency, low latency, and wide coverage, and further improve the collaborative coverage ability of the fixed + mobile double-layer network.

[0030] Furthermore, step P25-3 of the embodiment of the present application further includes: P25-31: Establish a penalty for the number of mobile monitoring devices enabled; P25-32: Within the limit device number range, perform path planning for the coverage acquisition location set, and evaluate the path fitness through the location importance, location time limit attention flag, and number enabled penalty, and establish a fitness evaluation result mapped to the path; P25-33: Screen the path through the fitness evaluation result to establish the acquisition adaptation optimization result.

[0031] Specifically, the process of establishing the mobile monitoring path can be further refined. First, establish a penalty mechanism for the number of mobile monitoring devices enabled, that is, set a penalty coefficient for the number of devices actually scheduled and used to form a constraint function for the device enabling behavior. This penalty mechanism aims to avoid waste of energy consumption and increased scheduling complexity caused by excessive use of device resources. Based on the ratio of the number of enabled devices to the limit device capacity, and combined with factors such as device operating cost, battery remaining capacity, and deployment risk, a non-linear or piecewise penalty function is set as one of the important parameters for subsequent path evaluation.

[0032] Next, within the determined limit device number range, perform multi-objective path planning calculations on the path set covering the current acquisition location set. The path planning process uses heuristic search algorithms (such as 、 genetic algorithm, simulated annealing, etc.), based on the shortest path cost function, and combines the location importance and location time limit attention flag of the sampling points covered on each path as the path priority weight. On this basis, establish a path fitness evaluation index system, which comprehensively considers the following three core factors: First, the weighted sum of high-importance points covered within the path; Second, whether the path sampling time window meets the time limit requirements of all locations; Third, whether the number of devices required for the path brings a high penalty value. The above factors jointly construct a fitness evaluation function to quantitatively score each candidate path and form a one-to-one mapping result between the path and the fitness, that is, the fitness evaluation result mapped to the path.

[0033] Finally, based on the above fitness evaluation results, perform path screening and optimal retention operations on all candidate path sets to select several paths with the highest global fitness under the current constraints as the output of the final acquisition adaptation optimization results. Exemplarily, a fixed threshold method (retaining paths with fitness greater than the threshold), a Pareto optimal method (retaining multi-objective balanced optimal solutions), or a dynamic hierarchical selection mechanism (prioritizing the retention of paths covering high-priority sampling points) can be used to ensure that the output paths have a high task completion rate, a reasonable resource consumption level, and good execution stability.

[0034] P30: After the target area is monitored by the first-layer fixed monitoring points and the second-layer mobile monitoring points, a carbon emission data network is established.

[0035] Optionally, during the monitoring of the target area, the first-layer fixed monitoring points and the second-layer mobile monitoring points work together to collect carbon emission data. The fixed monitoring points are distributed according to a pre-set grid and continuously and stably collect carbon emission data at their locations, including but not limited to carbon emission concentration, emission rate, monitoring timestamp, and geographical location information. The mobile monitoring points move flexibly within the target area according to the optimized mobile monitoring path to supplement the monitoring of key positions and weak monitoring areas, also collecting data such as carbon emission concentration and emission rate, and recording the monitoring time and specific location.

[0036] The collected data will be transmitted to the data processing center in real time through wireless communication technologies (such as 4G, 5G, NB-IoT, etc.) or wired communication methods (such as Ethernet). At the data processing center, the data from the fixed monitoring points and the mobile monitoring points are first preprocessed, including operations such as data cleaning, format unification, and outlier removal. The purpose of data cleaning is to remove incorrect data caused by factors such as equipment failures and signal interference; format unification is to convert the data collected by different monitoring points into a unified format for subsequent processing; outlier removal can avoid the impact of extreme data on the overall analysis results.

[0037] The preprocessed data will be integrated into the carbon emission data network. The carbon emission data network is a time-series data distribution set that records the carbon emission data of each monitoring point in the target area at different time points with time as the axis. Through data fusion technology, the data from the fixed monitoring points and the mobile monitoring points are fused to form a comprehensive and dynamic carbon emission data view. In the data fusion process, algorithms such as weighted average method and Kalman filter can be used to fuse the data from different monitoring points to improve the accuracy and reliability of the data. For example, for data at the same location within a similar time, different weights can be assigned according to the accuracy and reliability of the monitoring points for weighted averaging to obtain more accurate carbon emission data.

[0038] Meanwhile, the carbon emission data network will conduct spatio-temporal analysis on the data. In the spatial dimension, using Geographic Information System (GIS) technology, the carbon emission data is combined with geographical information to generate a carbon emission distribution map, visually showing the spatial distribution characteristics of carbon emissions in the target area. In the time dimension, the carbon emission data of the same monitoring point at different times is analyzed to draw a carbon emission time series curve, reflecting the dynamic change trend of carbon emissions. Through spatio-temporal analysis, high-value areas and high-emission periods of carbon emissions can be identified, providing data support for subsequent carbon emission tracing, equipment matching analysis, and early warning.

[0039] Furthermore, after establishing the carbon emission data network, the embodiment of the present application further includes step P30a, and step P30a further includes: P31a: Conduct carbon emission early warning analysis on the carbon emission data network and establish an early warning associated area; P32a: Report carbon emission anomalies according to the early warning associated area.

[0040] In a possible embodiment of the present application, after completing the construction of the carbon emission data network, step P30a can be continued to implement carbon emission early warning analysis and active reporting of abnormal information based on the data network, thereby extending the monitoring result from the static data layer to the dynamic response mechanism, and constructing an intelligent environmental supervision system with the capabilities of independent perception and active warning.

[0041] First, conduct carbon emission early warning analysis on the established carbon emission data network. Taking the real-time monitoring data in the network as input, combined with historical monitoring data, regional emission background models, meteorological impact models, and set emission risk thresholds, trend identification and risk judgment are carried out on the carbon emission status of each node and its adjacent areas. The early warning analysis not only focuses on the absolute over-standard situation of carbon concentration, but also includes multi-dimensional indicators such as its short-term change rate, spatial diffusion trend, and deviation degree from the airflow simulation model. Real-time identification of emission anomalies is achieved by setting a multi-level threshold model (such as general early warning, strong early warning, emergency early warning) or using a machine learning model (such as time series anomaly detection model, LSTM prediction model).

[0042] After completing the early warning analysis, an early warning associated area is further established. This area not only covers the core location where abnormal emission concentrations occur, but also deduces the possible extended range affected by carbon emissions by predicting the spatio-temporal diffusion path of carbon emissions and combining environmental factors such as meteorological data (wind direction, wind speed), building occlusion structure, and terrain undulation, thereby forming a dynamic associated block centered on the core anomaly point. For example, Geographic Information System (GIS) technology can be used to generate a carbon emission distribution map in combination with carbon emission data, and high-emission areas can be visually identified through the map.

[0043] Subsequently, based on the construction result of the above warning correlation area, the reporting operation of abnormal carbon emissions is completed. This reporting process supports multiple triggering mechanisms, including periodic reporting, real-time triggering when the threshold is exceeded, reporting based on sudden change rate detection, etc. The reported content is standardized to generate an event object, including but not limited to: abnormal occurrence time, abnormal type, warning level, influence range, core emission point number, monitoring data summary, deviation rate from the historical average value, etc., and is pushed to the supervision platform, industrial equipment control system or third-party environmental platform through the platform interface. When necessary, this abnormal report can also be linked to other modules within the system (such as the path reconstruction module, inversion analysis module) to achieve subsequent automated response processing.

[0044] P40: Based on the digital twin model, conduct airflow simulation within the target area, and perform source tracing and inversion according to the airflow simulation results and the carbon emission data network to establish the source tracing and inversion results.

[0045] Furthermore, step P40 of the embodiment of the present application further includes: P41: Input the carbon emission data network into the feature extraction channel, perform feature extraction, and establish the feature extraction results, including: a: Read the carbon concentration values at continuous time nodes for each position point, calculate the time derivative of the carbon concentration values, and establish local temporal gradient features; b: Construct a spatial gradient field through spatial interpolation of the local temporal gradient features of all position points to establish spatial gradient distribution features; c: Extract gradient change features based on the local temporal gradient features and the spatial gradient distribution features to establish key gradient features, and output the key gradient features as the feature extraction results; P42: Perform source tracing and inversion based on the feature extraction results and the airflow simulation results.

[0046] It should be understood that based on the digital twin model, a high-precision airflow simulation is carried out on the air flow situation within the target area, and the simulation results are combined with the established carbon emission data network to perform carbon emission source tracing and inversion operations to determine the possible sources of abnormal carbon emission concentrations and their diffusion paths, so as to achieve reverse tracking of pollution results.

[0047] First, based on the building geometry, terrain undulation, thermal environment field and real-time meteorological parameters (such as wind speed, wind direction, temperature and humidity, etc.) in the digital twin model, use the computational fluid dynamics (CFD) simulation algorithm or the wind field simulation method based on the lattice particle model to perform three-dimensional spatial modeling of the airflow state within the target area, and generate a wind speed vector field and an airflow direction field with resolution. The airflow simulation results provide dynamic conditions for the subsequent inversion calculation of carbon concentration changes and movement trends.

[0048] Furthermore, input the constructed carbon emission data network into the feature extraction channel and perform a series of spatio-temporal gradient feature extraction operations. Exemplarily, for each monitoring location point, read its carbon concentration measurement values at multiple consecutive time nodes, calculate the concentration change rate of this point in the time dimension through differential or smoothed derivative algorithms, and form the local temporal gradient feature of this point, which reflects the rapid growth or decay trend of the carbon concentration at this point. Then, perform spatial interpolation (such as Kriging interpolation, inverse distance weighted method, etc.) on the temporal gradient values of all location points in the spatial layer to form a continuous spatial gradient field, that is, the spatial distribution map of the carbon concentration change rate in different regions, and further construct the overall spatial gradient distribution feature.

[0049] On this basis, fuse the above two types of gradient information, analyze the synergy and abnormality of the gradient changes in the spatio-temporal dimension, and identify the key gradient features representing the emission anomaly propagation source or possible source. This process can combine gradient aggregation point detection, streamline backtracking analysis or high-order derivative feature analysis methods to achieve the dynamic extraction of the local emission peak diffusion center, and output this result as the feature extraction result.

[0050] Finally, use the above key gradient feature results and the airflow simulation output results for joint modeling and perform carbon emission source tracing inversion. In this process, compare the vector direction of the gradient change direction of the carbon concentration with the wind field simulation result, and establish a set of inversion equations in combination with the concentration change rate. Infer the possible starting point and path of the carbon emission anomaly through the reverse path backtracking method (such as reverse particle tracking, source tracing model regularization solution method, etc.), and output it as a structured source tracing inversion result, including core fields such as origin coordinates, inversion path, time window, credibility score, etc. The source tracing inversion result will provide a scientific basis for subsequent equipment matching analysis and carbon emission tracking and warning, and help achieve the effective management and control of carbon emissions in the target area.

[0051] Furthermore, the embodiment of the present application further includes step P40a, and step P40a further includes: P41a: Establish an inversion adaptation mode database for instantaneous emissions, continuous emissions, and intermittent emissions; P42a: When performing source tracing inversion, use the inversion adaptation mode database for carbon emission data network matching, and perform source tracing inversion according to the inversion adaptation mode matching result and the airflow simulation result.

[0052] Optionally, to improve the accuracy and adaptability of carbon emission source tracing inversion, after completing the basic inversion based on airflow simulation and gradient feature extraction, an emission behavior pattern recognition and inversion adaptation mechanism can be introduced. By constructing an emission feature pattern database and matching strategy, the inversion accuracy under complex emission types can be improved and the source discrimination ability in the case of multi-source emissions can be achieved.

[0053] First, construct an inversion adaptation mode database for instantaneous emissions, continuous emissions, and intermittent emissions. This database classifies and models a large number of historical emission event samples, summarizes the regular characteristics of different types of emission events in terms of temporal variation of carbon concentration, spatial diffusion path, gradient feature structure, etc., and establishes corresponding inversion adaptation templates respectively. Among them, the instantaneous emission mode corresponds to the scenario of short-term and large-scale release of carbon gas, and its characteristic is that the concentration rises rapidly in a very short time and has a clear diffusion wavefront; the continuous emission mode is applicable to the scenario of long-term and stable release of carbon gas by industrial equipment or pipelines, manifested as a high concentration level but gentle change, and relatively stable spatio-temporal gradient; while the intermittent emission mode is mostly seen in periodically operating equipment, and its characteristic is regular concentration fluctuations within a certain time period, with periodic temporal characteristics and local spatial aggregation. Each emission mode in the database is associated with corresponding information such as typical concentration change curves, spatial gradient forms, propagation time windows, and wind field perturbation response models, which are used for matching and adaptation in the subsequent source tracing process.

[0054] After the database construction is completed, that is, during the process of carbon emission source tracing inversion, first match and identify the actual observed data in the currently constructed carbon emission data network with various emission modes in the inversion adaptation mode database. This matching process can comprehensively judge the emission mode that best suits the current observation situation based on multi-dimensional indicators such as temporal trend similarity (such as DTW dynamic time warping), gradient distribution consistency measurement (such as structural similarity index), and wind field response behavior consistency comparison. After identifying the most suitable emission mode, introduce the typical diffusion characteristics, response delay parameters, and reverse path form included in this mode as supplementary constraint conditions into the inversion calculation model.

[0055] Subsequently, jointly model the above inversion adaptation mode matching results with the existing airflow simulation results, key gradient characteristics, and path inversion equations, and re-perform high-precision source tracing inversion through methods such as difference compensation, path adjustment, and credibility weighting. Compared with traditional methods, this joint inversion result can more effectively correct the systematic deviation in the inversion path caused by emission mode differences, and improve the recognition ability of multi-type emission sources and the source location accuracy.

[0056] P50: Perform equipment matching analysis in the target area according to the source tracing inversion result, establish equipment location results, and establish carbon emission tracking and early warning according to the equipment location results.

[0057] Furthermore, step P50 of the embodiment of the present application further includes: P51: Obtain the carbon emission data of the equipment calibration working conditions in the target area; P52: Use the obtained result of traceability inversion to perform adaptation and matching of the carbon emission data of the equipment calibration working conditions; P53: If the adaptation and matching result meets the first constraint condition, after obtaining the enterprise's authorization and permission, read the working data of the corresponding equipment, and perform matching verification based on the working data to establish the equipment positioning result.

[0058] Specifically, based on the result of traceability inversion obtained in the previous steps, further carry out equipment matching analysis and positioning identification in the target area to achieve accurate traceability of carbon emission sources and identification of responsible entities, and accordingly establish a targeted carbon emission tracking and early warning mechanism.

[0059] First, obtain the carbon emission data of the calibration working conditions of all registered or filed emission-related equipment in the target area. This data includes the carbon emission baseline value, emission rate characteristic curve, carbon emission factor per unit energy consumption, emission cycle and timing characteristics corresponding to the operation mode, etc. of each piece of equipment. The data sources can include the enterprise's self-monitoring system, environmental protection department's filing materials, performance parameters provided by equipment manufacturers, etc. The data format needs to match the unified standard of the system for subsequent automated comparison and analysis.

[0060] Next, according to the generated result of traceability inversion, including abnormal source point coordinates, spatio-temporal diffusion trajectory, emission intensity characteristics, key gradient indicators, etc., perform the adaptation and matching operation on the calibration working condition data of each piece of equipment. This matching process is implemented based on a multi-dimensional comparison strategy. For example: calculate the relative error of the emission intensity amplitude and the standard emission level; analyze the shape similarity of the time series concentration change characteristics and the working condition emission curve (relevant coefficients, DTW, structural similarity, etc. can be used); and combine indicators such as the geographical distance between the equipment and the inversion result in space and the overlap degree of the wind field path to judge the spatial adaptability.

[0061] After the preliminary matching is completed, if the matching result of a certain piece of equipment meets the preset first constraint condition (such as: the shape similarity exceeds the threshold, the spatio-temporal matching rate is higher than the set value, the relative error of the emission intensity is less than the specified limit, etc.), it is determined that the equipment is a possible emission source, and on the premise of obtaining the enterprise's authorization and permission, further retrieve the real-time or historical working data of the equipment. This working data can include the operation time period, load status, start-stop records, production process parameters, etc. of the equipment. Compare this data with the emission time series information obtained by inversion, and carry out matching verification analysis from the perspectives of the logical consistency between the working load and the emission intensity, the coincidence degree between the operation cycle and the abnormal time point, etc. If the verification result further meets the set multi-level matching conditions, it is confirmed that the equipment is the direct responsible equipment for this carbon emission event, and an equipment positioning result with equipment number, positioning coordinates, emission behavior characteristics and matching credibility is generated.

[0062] Based on the device positioning result, immediately generate and publish carbon emission tracking warning information. This warning information not only includes the abnormal location and time period, but also can be associated with content such as the responsible device identifier, emission intensity, warning level, and disposal suggestions, forming an intelligent warning output that is operationally significant for enterprises or regulatory units, and supporting rapid response in subsequent emission control or automated intervention and other links.

[0063] Furthermore, step P52 of the embodiment of the present application further includes: P52-1: If the adaptation matching result cannot meet the first constraint condition, then establish a joint device emission constraint; P52-2: Perform joint device adaptation matching on the source inversion result according to the joint device emission constraint to establish a device positioning result.

[0064] In a possible embodiment of the present application, to cope with the complex situation where carbon emission behaviors may be caused by the collaborative or superimposed emissions of multiple devices, a joint emission adaptation mechanism in the case where the matching of a single device fails can be introduced to enhance the system's adaptability and positioning accuracy in a multi-source emission environment.

[0065] First, judge the preliminary matching result. If the adaptation matching result of a single device cannot meet the first constraint condition (that is, it cannot simultaneously meet the system-set thresholds in terms of emission intensity, timing characteristics, spatial matching degree, etc.), it is presumed that the current emission behavior may be caused by the joint emission or alternating emission of two or more devices. At this time, instead of directly terminating the device positioning process, enter the joint matching analysis path and establish a joint device emission constraint model accordingly. This constraint model takes the device combination as a unit and defines the theoretical carbon concentration synthesis result within the target time and space range after the superimposed emissions of multiple devices. The specific constraint parameters include: the maximum allowable emission intensity of the participating devices, the emission time overlap relationship, the spatial diffusion impact overlap area, the cooperation probability between devices (which can be based on historical joint operation records), and the distance weight function near the inversion source point, etc.

[0066] Subsequently, based on the above joint device emission constraint, perform joint device adaptation matching analysis on the source inversion result. During this analysis process, perform combined iteration on the emission condition data of all devices or device combinations in the target area, simulate the synthetic impact of their joint emissions on the local spatio-temporal carbon concentration change, and perform a full-dimensional comparison with key gradient features, path diffusion patterns, airflow disturbance trajectories, and other elements in the source inversion result. Through a multi-objective fitting function, evaluate the minimum residual between the concentration simulation value and the actual measurement value of each combination, and score the morphological consistency and change trend conformity of the fitting curve, and finally determine the optimal device combination as the suspected joint emission source.

[0067] When a certain combination meets the set joint adaptation constraint conditions (such as the minimum sum of squared residuals, the fitting similarity being higher than a threshold, a high confidence score, etc.), output this combination as the final device positioning result, and clearly mark that this emission behavior is jointly generated by multiple devices, and attach detailed information such as the joint emission intensity ratio, the contribution score of each device, and the matching confidence level.

[0068] By introducing the above steps, a dynamic adaptation and intelligent collaborative reasoning mechanism for the case of single emission source matching failure is realized, significantly improving the problem closed-loop ability, source identification comprehensiveness, and supervision accuracy in actual complex carbon emission scenarios.

[0069] In summary, the embodiments of the present application at least have the following technical effects: The present application constructs a virtual environment of the target area through digital twin modeling, divides grids according to the monitoring accuracy to deploy fixed monitoring points, combines carbon emission concern fitting to generate mobile monitoring paths, forms a carbon emission data network covering the whole area, and then performs traceability inversion based on airflow simulation and data network, matches the device calibration data to achieve source positioning, and finally realizes dynamic tracking and early warning of carbon emissions. It achieves the technical effect of realizing high-precision traceability positioning and real-time tracking and early warning of carbon emission sources by constructing a multi-level monitoring network, introducing airflow simulation and emission feature matching mechanisms.

[0070] Embodiment 2, based on the same inventive concept as the dynamic tracking and early warning method of intelligent carbon emissions in the foregoing embodiment, as Figure 2 shown, the present application provides a dynamic tracking and early warning system for intelligent carbon emissions. The system in the embodiments of the present application and the method embodiments are based on the same inventive concept. Among them, the system includes: A fixed monitoring point distribution module 11, which is used to perform digital twin modeling of the target area, execute grid segmentation of the target area after configuring the grid granularity according to the monitoring accuracy, and distribute the first-level fixed monitoring points according to the grid segmentation result.

[0071] A mobile monitoring point configuration module 12, which is used to perform carbon emission concern fitting of the target area by using the digital twin model, establish a mobile monitoring path according to the carbon emission concern fitting result, and configure the second-level mobile monitoring points through the mobile monitoring path.

[0072] A carbon emission data network establishment module 13, which is used to establish a carbon emission data network after monitoring the target area by the first-level fixed monitoring points and the second-level mobile monitoring points.

[0073] The traceability inversion module 14 is used to perform airflow simulation in the target area based on the digital twin model, perform traceability inversion according to the airflow simulation results and the carbon emission data network, and establish traceability inversion results.

[0074] The carbon emission tracking and warning module 15 is used to perform equipment matching analysis in the target area according to the traceability inversion results, establish equipment positioning results, and establish carbon emission tracking and warning according to the equipment positioning results.

[0075] Furthermore, the mobile monitoring point configuration module 12 is further used to perform the following steps: Read the fixed monitoring data set of the first-level fixed monitoring points; perform carbon emission clustering on the fixed monitoring data set to generate carbon emission clustering results; perform weak monitoring coverage analysis in the carbon emission clustering results to establish real-time weak focus; generate real-time location focus according to the carbon emission clustering results; after compensating according to the real-time weak focus, the real-time location focus, and the carbon emission focus fitting results, establish the mobile monitoring path.

[0076] Furthermore, the mobile monitoring point configuration module 12 is further used to perform the following steps: Use the real-time weak focus and the real-time location focus to compensate the carbon emission focus fitting results to establish a set of acquisition locations, and the set of acquisition locations is marked with location importance and location time limit focus; configure the limit number of mobile monitoring devices; within the limit number of devices, perform acquisition adaptation optimization of the set of acquisition locations by the mobile monitoring devices to generate acquisition adaptation optimization results; establish the mobile monitoring path according to the acquisition adaptation optimization results.

[0077] Furthermore, the mobile monitoring point configuration module 12 is further used to perform the following steps: Establish a penalty for enabling the number of mobile monitoring devices; within the limit number of devices, perform path planning for covering the set of acquisition locations, and perform path fitness evaluation through location importance, location time limit focus identification, and number enabling penalty to establish a fitness evaluation result mapped to the path; perform path screening through the fitness evaluation result to establish acquisition adaptation optimization results.

[0078] Furthermore, the carbon emission data network establishment module 13 is further used to perform the following steps: Perform carbon emission warning analysis on the carbon emission data network and establish a warning associated area; report carbon emission anomalies according to the warning associated area.

[0079] Furthermore, the traceability inversion module 14 is further used to perform the following steps: Input the carbon emission data network into the feature extraction channel, perform feature extraction, and establish a feature extraction result, including: a: Read the carbon concentration values at consecutive time nodes for each location point, calculate the time derivative of the carbon concentration values, and establish local temporal gradient features; b: Construct a spatial gradient field by spatially interpolating the local temporal gradient features of all location points, and establish spatial gradient distribution features; c: Extract gradient change features based on the local temporal gradient features and the spatial gradient distribution features, establish key gradient features, and output the key gradient features as the feature extraction result; perform source tracing inversion based on the feature extraction result and the airflow simulation result.

[0080] Furthermore, the source tracing inversion module 14 is further configured to perform the following steps: Establish an inversion adaptation mode database for instantaneous emissions, continuous emissions, and intermittent emissions; when performing source tracing inversion, use the inversion adaptation mode database to match the carbon emission data network, and perform source tracing inversion based on the inversion adaptation mode matching result and the airflow simulation result.

[0081] Furthermore, the carbon emission tracking and warning module 15 is further configured to perform the following steps: Obtain the carbon emission data of the equipment calibration working conditions in the target area; perform adaptation matching on the carbon emission data of the equipment calibration working conditions by using the source tracing inversion result; if the adaptation matching result meets the first constraint condition, after obtaining the enterprise authorization and permission, read the working data of the corresponding equipment, and perform matching verification based on the working data to establish an equipment positioning result.

[0082] Furthermore, the carbon emission tracking and warning module 15 is further configured to perform the following steps: If the adaptation matching result does not meet the first constraint condition, establish a joint equipment emission constraint; perform joint equipment adaptation matching on the source tracing inversion result according to the joint equipment emission constraint to establish an equipment positioning result.

[0083] It should be noted that the above sequence of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above describes specific embodiments of this specification. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0084] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0085] This specification and the drawings are merely illustrative of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, provided that these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications therein.

Claims

1. A dynamic tracking and warning method for intelligent carbon emissions, characterized in that, The method includes: Performing digital twin modeling of the target area, configuring the grid granularity according to the monitoring accuracy, then performing grid segmentation of the target area, and distributing the first-level fixed monitoring points according to the grid segmentation results; Using the digital twin model to perform carbon emission concern fitting of the target area, establishing a mobile monitoring path according to the carbon emission concern fitting results, and configuring the second-level mobile monitoring points through the mobile monitoring path; After monitoring the target area by the first-level fixed monitoring points and the second-level mobile monitoring points, establishing a carbon emission data network; Performing airflow simulation within the target area based on the digital twin model, and performing traceability inversion according to the airflow simulation results and the carbon emission data network to establish a traceability inversion result, including: Inputting the carbon emission data network into a feature extraction channel, performing feature extraction, and establishing a feature extraction result, including: a: Reading the carbon concentration values at consecutive time nodes for each location point, calculating the time derivative of the carbon concentration values, and establishing local temporal gradient features; b: Constructing a spatial gradient field through spatial interpolation of the local temporal gradient features of all location points, and establishing spatial gradient distribution features; c: Performing gradient change feature extraction according to the local temporal gradient features and the spatial gradient distribution features, establishing key gradient features, and outputting the key gradient features as the feature extraction result; Performing traceability inversion according to the feature extraction result and the airflow simulation result; Performing device matching analysis within the target area according to the traceability inversion result, establishing a device positioning result, and establishing a carbon emission tracking warning according to the device positioning result.

2. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 1, characterized in that The establishing of the mobile monitoring path according to the carbon emission concern fitting results and configuring the second-level mobile monitoring points through the mobile monitoring path includes: Reading the fixed monitoring data set of the first-level fixed monitoring points; Performing carbon emission clustering on the fixed monitoring data set to generate a carbon emission clustering result; Performing monitoring coverage weakness analysis in the carbon emission clustering result to establish real-time weakness concerns; Generating real-time location concerns according to the carbon emission clustering result; After compensating according to the real-time weakness concerns, the real-time location concerns, and the carbon emission concern fitting results, establishing the mobile monitoring path.

3. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 2, characterized in that, The establishing of the mobile monitoring path after compensating according to the real-time weakness concerns, the real-time location concerns, and the carbon emission concern fitting results includes: Compensating the carbon emission concern fitting results using the real-time weakness concerns and the real-time location concerns to establish a set of collection locations, and the set of collection locations is marked with location importance and location time limit concerns; Configuring the limit number of mobile monitoring devices; Within the limit number of devices, performing collection adaptation optimization of the set of collection locations by the mobile monitoring devices to generate a collection adaptation optimization result; Establishing the mobile monitoring path according to the collection adaptation optimization result.

4. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 3, characterized in that, The performing of collection adaptation optimization of the set of collection locations by the mobile monitoring devices within the limit number of devices to generate a collection adaptation optimization result includes: Establishing a penalty for the number of mobile monitoring devices enabled; Within the range of the limit number of devices, perform path planning to cover the set of acquisition locations, and evaluate the path fitness by focusing on identification, quantity enabling penalties based on location importance and location time limits, and establish the fitness evaluation results mapped to the path. Perform path screening based on the fitness evaluation results to establish the acquisition adaptation optimization results.

5. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 1, characterized in that Perform device matching analysis within the target area based on the traceability inversion results, and establish device positioning results, including: Obtain the carbon emission data of the device calibration working conditions within the target area; Use the traceability inversion results to perform adaptive matching of the carbon emission data of the device calibration working conditions; If the adaptive matching result meets the first constraint condition, after obtaining the enterprise authorization and permission, read the working data of the corresponding device, and perform matching verification based on the working data to establish the device positioning results.

6. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 5, characterized in that The use of the traceability inversion results to perform adaptive matching of the carbon emission data of the device calibration working conditions includes: If the adaptive matching result does not meet the first constraint condition, establish a joint device emission constraint; Perform joint device adaptive matching of the traceability inversion results according to the joint device emission constraint to establish the device positioning results.

7. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 1, characterized in that The traceability inversion based on the airflow simulation results and the carbon emission data network further includes: Establish an inversion adaptation mode database for instantaneous emission, continuous emission, and intermittent emission; When performing traceability inversion, use the inversion adaptation mode database to perform carbon emission data network matching, and perform traceability inversion based on the inversion adaptation mode matching results and the airflow simulation results.

8. The dynamic tracking and early warning method for intelligent carbon emissions according to claim 1, wherein, After establishing the carbon emission data network, it further includes: Perform carbon emission early warning analysis on the carbon emission data network, and establish an early warning associated area; Report carbon emission anomalies according to the early warning associated area.

9. The dynamic tracking and early warning system for intelligent carbon emissions, characterized in that, The system includes: A fixed monitoring point distribution module, which is used to perform digital twin modeling of the target area, perform grid segmentation of the target area after configuring the grid granularity according to the monitoring accuracy, and distribute the first-level fixed monitoring points according to the grid segmentation results; A mobile monitoring point configuration module, which is used to perform carbon emission concern fitting of the target area using the digital twin model, establish a mobile monitoring path according to the carbon emission concern fitting results, and configure the second-level mobile monitoring points through the mobile monitoring path; A carbon emission data network establishment module, which is used to establish a carbon emission data network after monitoring the target area by the first-level fixed monitoring points and the second-level mobile monitoring points; A traceability inversion module, which is used to perform airflow simulation in the target area based on the digital twin model, and perform traceability inversion according to the airflow simulation results and the carbon emission data network to establish traceability inversion results, including: inputting the carbon emission data network into a feature extraction channel, performing feature extraction, and establishing feature extraction results, including: a: reading carbon concentration values at continuous time nodes for each location point, calculating the time derivative of the carbon concentration values, and establishing local temporal gradient features; b: constructing a spatial gradient field through spatial interpolation of the local temporal gradient features of all location points, and establishing spatial gradient distribution features; c: extracting gradient change features according to the local temporal gradient features and the spatial gradient distribution features, establishing key gradient features, and outputting the key gradient features as the feature extraction results; performing traceability inversion according to the feature extraction results and the airflow simulation results; A carbon emission tracking and warning module, which is used to perform equipment matching analysis in the target area according to the traceability inversion results, establish equipment positioning results, and establish carbon emission tracking and warnings according to the equipment positioning results.

Images