Real-time dynamic interaction and regulation platform for carbon flux based on sky tower ground well observation

Abstract

This invention discloses a real-time dynamic interactive and control platform for carbon flux based on "sky-tower-ground-well" observations. It includes a comprehensive "sky-tower-ground-well" observation and data acquisition module, a control module, and an interactive visualization and decision support module. The comprehensive "sky-tower-ground-well" observation and data acquisition module includes a satellite remote sensing module, a UAV remote sensing module, an eddy covariance flux tower, a ground observation module, and a downhole observation module. The control module, based on the data acquired by the data acquisition module, provides instructions and dynamic control mechanisms for regulating carbon emissions and / or carbon sinks. The interactive visualization and decision support module provides support for intelligent decision-making in the carbon emission and carbon sink management process through an intuitive, real-time graphical interface. This invention enables real-time observation of carbon emissions and carbon sinks in coal mining areas and automatically adjusts management measures through a dynamic control mechanism, significantly improving the accuracy, efficiency, and flexibility of carbon flux observation and control.

Description

Technical Field

[0001] This invention relates to the field of data interaction, control, and visualization technology, and in particular to a real-time dynamic interaction and control platform for carbon flux based on "sky tower ground well" observations. Background Technology

[0002] In the development and management of coal mining areas, the dynamic changes in carbon emissions and carbon sinks are crucial to balancing environmental protection and resource utilization. In recent years, with the increasing awareness of environmental protection, although some integrated monitoring technologies have been developed, such as combined sky-to-ground, sky-to-ground-well, and sky-to-to-ground methods, long-term, real-time, and high-precision carbon emission and carbon sink monitoring is still not possible. As a result, most existing carbon emission and carbon sink monitoring platforms remain at the level of data collection and analysis, lacking real-time and dynamic control functions. They cannot make immediate adjustments to carbon emissions and carbon sinks based on actual data, let alone meet the dynamic optimization needs of carbon emissions and carbon sinks under different environmental conditions.

[0003] Specifically, existing carbon emission and carbon sink monitoring platforms have the following shortcomings:

[0004] Insufficient decision support: Existing carbon emission and carbon sink monitoring platforms mainly rely on a single data source, making it difficult to provide a comprehensive analysis of carbon emission and carbon sink status. They lack intelligent decision support and cannot effectively respond to sudden changes in carbon emissions and carbon sinks.

[0005] The visualization and interactive functions are inadequate: Although the existing carbon emission and carbon sink monitoring platforms provide a visual display of the carbon emission and carbon sink monitoring results, they are limited to static data charts and trend analysis, lacking interactivity and dynamism, and unable to achieve real-time interaction with users.

[0006] Lack of dynamic response mechanism: Existing carbon emission and carbon sink monitoring platforms lack intelligent dynamic control mechanisms for carbon emissions and carbon sinks.

[0007] Therefore, there is an urgent need for a comprehensive platform that can achieve multi-source data fusion, real-time dynamic control, intelligent decision support, and interactive visualization to adapt to the complexity and dynamism of carbon emission and carbon sink monitoring in coal mining areas. Summary of the Invention

[0008] To overcome the aforementioned shortcomings, this invention provides a real-time dynamic interaction and control platform for carbon flux based on "sky-tower-ground-well" observations. Utilizing the integration of multi-source heterogeneous data and dynamic control algorithms, it provides real-time, intelligent, and interactive monitoring and decision support. This invention aims to address the problems of data silos, slow response, lack of dynamic control, and insufficient decision support in existing carbon emission and carbon sink management technologies. "Sky-tower-ground-well" refers to the collective term for multi-source data collected from the sky, tower, and ground levels.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0010] The real-time dynamic interaction and control platform for carbon flux based on "Sky Tower Ground Well" observations includes a comprehensive "Sky Tower Ground Well" observation and data acquisition module, a control module, and an interactive visualization and decision support module.

[0011] The data acquisition module includes a satellite remote sensing module, an UAV remote sensing module, a vorticity covariance flux tower, a ground observation module, and a downhole observation module.

[0012] The control module adjusts carbon emissions and / or carbon sinks in real time based on the data acquired by the data acquisition module, thereby reducing carbon emissions and enhancing carbon sink capacity.

[0013] The interactive visualization and decision support module provides support for intelligent decision-making in the process of carbon emission and carbon sink management through an intuitive and real-time graphical interface.

[0014] Preferably, the carbon emission control strategy of the control module is as follows:

[0015]

[0016] in, This indicates the intensity of carbon emission control; α is the carbon emission control coefficient, and ΔE(t) is the carbon emission deviation value.

[0017] The carbon sequestration control strategy of the control module is as follows:

[0018]

[0019] in, This indicates the intensity of carbon sink regulation; β is the carbon sink regulation coefficient, and ΔC(t) is the carbon sink deviation value;

[0020] Based on equations (3) and (4), the control module adjusts carbon emission and carbon sink targets according to real-time observation data and generates corresponding control schemes.

[0021] Furthermore, the control module includes intelligent optimization of the control scheme, and the intelligent optimization objective of the control scheme is:

[0022]

[0023] The above formula represents the sum of squares of the deviation between carbon emissions and carbon sinks.

[0024] Furthermore, the intelligent optimization of the control scheme adopts an adaptive optimization mechanism based on deep learning based on the feedback results, and optimizes the control decision by training with reinforcement learning historical and real-time data.

[0025] Preferably, the feedback result is as follows:

[0026] F adjust (t)=|ΔE(t)|+|ΔC(t)| (6)

[0027] When F adjust (t) When the preset threshold is exceeded, the control module triggers intelligent optimization;

[0028] The intelligent optimization adopts a deep Q-reinforcement learning model, and the optimization formula of the deep Q-reinforcement learning model is as follows:

[0029] Q(s,a)=R(s,a)+γ·max a Q(s′,a′) (7)

[0030] Where Q(s,a) represents the expected reward for taking action a in state s; R(s,a) is the immediate reward; γ is the discount factor, which determines the degree of influence of future rewards; s′ is the next state, and a′ is the optimal next action.

[0031] Preferably, the carbon emission deviation is calculated using the following formula:

[0032] ΔE(t)=E pollution (t)-E target (t) (1)

[0033] Where ΔE(t) is the carbon emission deviation at time t; E pollution (t) represents the actual carbon emission concentration; E target (t) represents the target carbon emission concentration;

[0034] The carbon sink deviation is calculated as follows:

[0035] ΔC(t)=C sink (t)-C target (t) (2)

[0036] Where ΔC(t) is the carbon sink deviation at time t; C sink (t) represents the actual carbon sequestration capacity; C target (t) represents the target carbon sink capacity.

[0037] Preferably, the interactive visualization and decision support module includes:

[0038] Data input and real-time display: The front-end interface presents the received multi-source data in real time, and based on the fused data, displays the following graphical information: carbon emission heat map, carbon sink distribution map, and dynamic trend map of carbon emission and carbon sink.

[0039] Interactive adjustment and real-time feedback: An interactive interface is provided for adjusting control parameters and selecting different management objectives, and the response after adjustment is displayed. The control parameters include: carbon emission targets and carbon sink compensation plans.

[0040] Decision support and optimization suggestions: Based on real-time data and optimization algorithms, the platform not only provides a visual display of monitoring results, but also generates optimization suggestions through intelligent algorithms. These suggestions include: automatically providing adjustment suggestions based on the current carbon emissions and carbon sink status of the mining area, and displaying the expected effects of these adjustments in real time.

[0041] Preferably, the decision support and optimization suggestions include model prediction and decision support, which uses historical and real-time data to predict future changes in carbon emissions and carbon sinks through machine learning models or time series analysis models. Based on the prediction results, the carbon emission and carbon sink targets are automatically adjusted to avoid exceeding carbon emission limits or insufficient carbon sink capacity.

[0042] Preferably, the model prediction uses an ARIMA or LSTM model for time series prediction, and the ARIMA or LSTM model outputs the following predicted values:

[0043]

[0044] Among them, f pollution and f sink It is a predictive model, where t represents the current time and n is the time window length of historical data.

[0045] Furthermore, the decision support and optimization suggestions generate the optimal strategy based on model optimization, and the optimization objective of the optimal strategy is as follows:

[0046]

[0047] The above formula aims to minimize the difference between carbon emission and carbon sink targets simultaneously, and finds the optimal solution through an optimization algorithm, such as gradient descent or linear programming.

[0048] The beneficial effects of this invention are as follows:

[0049] 1. This invention can acquire multi-source data in real time and automatically adjust carbon emission and carbon sink management measures according to real-time changes, significantly improving the accuracy, efficiency, and flexibility of carbon flux observation and control.

[0050] 2. Through an adaptive control mechanism and optimization algorithm, the platform can intelligently generate control schemes, reduce manual intervention, and improve decision-making efficiency.

[0051] 3. Through multi-source data fusion and spatiotemporal synchronization, this invention enables the platform to provide more accurate carbon emission and carbon sink observation results, helping managers to carry out efficient regulation.

[0052] 4. The interactive interface of this invention enhances the user experience, allowing managers to intuitively view data changes and quickly adjust control strategies.

[0053] 5. This invention combines a feedback mechanism with an optimization algorithm to ensure the continuous optimization of carbon emission and carbon sink management measures. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below.

[0055] The embodiment discloses a real-time dynamic interaction and control platform for carbon flux based on "sky tower ground well" observations (hereinafter referred to as "the platform"), as follows:

[0056] 1. Platform architecture design:

[0057] This platform is based on the integrated carbon emission and carbon sink monitoring technology of "Sky Tower Ground Well" and uses the following multiple data sources:

[0058] The "sky" level comprises two main modules. The first is the satellite remote sensing module: Utilizing multispectral, high-resolution remote sensing satellites, it periodically scans the mining area and its surroundings to acquire environmental parameters such as vegetation indices, soil moisture, and temperature, thereby predicting changes in carbon flux within the mining area. Particularly useful for large-scale, long-term dynamic observations, satellite remote sensing provides carbon flux data for the mining area and its surrounding regions, filling gaps in large-scale mining area observation techniques and suitable for continuously acquiring macroscopic data on carbon flux changes. The second is the UAV remote sensing module: In areas with insufficient satellite data updates or complex local environments, UAVs equipped with high-precision sensors supplement the acquisition of local environmental images and radiation data. This is particularly suitable for short-term, dynamic observations of localized areas within the mining area, offering unique advantages, especially in the detailed observation of local hotspots.

[0059] At the "tower" level: Eddy covariance flux towers are deployed in key areas of the mining area. Using gas flow sensors, temperature and humidity sensors, and weather stations, real-time data on wind speed, temperature, carbon dioxide, and other greenhouse gas concentrations are collected, and atmospheric carbon exchange flux is calculated. These towers enable real-time monitoring of carbon emissions and carbon sink changes within the mining area. Their observation range is typically 1.5-2 times the tower's height, depending on surface roughness and turbulence characteristics. By measuring high-frequency time-series data of wind speed and gas concentration and calculating their covariance, the vertical gas exchange rate in turbulence can be obtained, thereby estimating the gas flux in the vertical direction.

[0060] At the "ground well" level, there are two main modules. The surface observation module deploys a network of fixed or mobile sensors, combined with regular manual sampling, to accurately capture the release and absorption of carbon within the mining area, including the impact of mining activities on the carbon sequestration capacity of soil and vegetation, and to monitor key indicators of carbon emissions and absorption in the surface environment in real time. The underground observation module is similar to the surface module, deploying high-temperature resistant, waterproof, moisture-proof, and dust-resistant explosion-proof carbon flux monitoring sensors in key underground areas. Combined with manual sampling, it acquires underground carbon flux data and other relevant environmental parameters.

[0061] The aforementioned multi-source data, through platform integration and fusion, enables comprehensive monitoring of carbon emissions and carbon sinks in coal mining areas.

[0062] 2. Real-time dynamic interactive control mechanism

[0063] The real-time dynamic control mechanism is one of the platform's core innovations. It aims to automatically generate and adjust carbon emission and carbon sink management measures in real time based on collected multi-source carbon emission and carbon sink data. Through intelligent algorithms, the platform can automatically generate and optimize management plans based on real-time changes in carbon emissions and carbon sinks in coal mining areas, ensuring real-time control and optimization of carbon emissions and carbon sinks, thereby improving the efficiency and responsiveness of environmental management.

[0064] The core objective of the real-time dynamic control mechanism is to dynamically adjust carbon emission and carbon sink management measures in coal mining areas. Specific objectives include: carbon emission control: when the platform observes that carbon emissions in a certain area exceed a preset threshold, it automatically initiates control measures, such as increasing the absorption of fugitive gas, to reduce carbon emissions; carbon sink regulation: when carbon sink capacity is insufficient, the platform will automatically adjust the carbon sink compensation area or introduce new carbon sink measures to enhance carbon sink capacity, such as carbon storage in goaf areas to assist in increasing carbon sinks.

[0065] However, practical applications face the following technical challenges: First, data timeliness and real-time performance are crucial, as carbon emissions and carbon sinks change very rapidly, requiring real-time data streams to support rapid decision-making. Second, the precision of regulatory response is critical; how to accurately adjust carbon emission and carbon sink management measures based on real-time data to maximize carbon emission reduction and enhance carbon sink capacity.

[0066] The working principle of the real-time dynamic control mechanism is based on the following main steps:

[0067] Data input and real-time observation: The platform collects carbon emissions, carbon sinks, and meteorological data in real time from multiple data sources (including satellite remote sensing, UAV imagery, ground and downhole sensors, etc.). The carbon emission concentration is set to E. pollution (t), representing the carbon emission concentration at time t; carbon sink capacity C sink (t), representing the carbon sink capacity at time t.

[0068] Target setting and deviation calculation. The platform sets target values ​​E for carbon emissions and carbon sinks. target (t) and C target (t), and calculate the deviation between actual carbon emissions and carbon sink capacity and the target. The carbon emission deviation is expressed as:

[0069] ΔE(t)=E pollution (t)-E target (t) (1)

[0070] Where ΔE(t) represents the carbon emission deviation at time t; E pollution (t) represents the actual carbon emission concentration; E target (t) represents the target carbon emission concentration.

[0071] Carbon sink deviation is expressed as:

[0072] ΔC(t)=C sink (t)-C target (t) (2)

[0073] Where ΔC(t) represents the carbon sink deviation at time t; C sink (t) represents the actual carbon sequestration capacity; C target (t) represents the target carbon sink capacity.

[0074] The platform automatically generates control strategies based on the deviation between carbon emissions and carbon sinks. Specifically, it generates adjustment strategies according to the magnitude of the deviation to quickly respond to excessive carbon emissions or insufficient carbon sinks. The calculation formula for the control strategy is as follows:

[0075]

[0076] in, The value represents the intensity of carbon emission control; α is the control coefficient, representing the adjustment range, and its value is based on historical data and the optimization model; ΔE(t) is the deviation value of carbon emissions.

[0077] The carbon sink adjustment strategies are as follows:

[0078]

[0079] in, β represents the intensity of carbon sink regulation; β is the carbon sink regulation coefficient, representing the intensity of improving carbon sink capacity; ΔC(t) is the deviation value of carbon sink.

[0080] Using the above formula, the platform intelligently adjusts carbon emission and carbon sink targets based on real-time observation data, generating corresponding control schemes.

[0081] This embodiment also introduces an intelligent optimization algorithm to optimize the control strategy in real time. After the control strategy is executed, the platform will further optimize the control measures based on feedback data to ensure the accuracy of the control results. The optimization objectives are as follows:

[0082]

[0083] The optimization objective aims to minimize the sum of squares of the deviations between carbon emissions and carbon sinks. In this way, the platform can dynamically adjust management plans to achieve the optimization of carbon emission and carbon sink management.

[0084] The intelligent feedback and adjustment mechanism aims to continuously evaluate the effectiveness of regulation and adjust the regulatory strategy accordingly. The feedback result is calculated using the following formula:

[0085] F adjust (t)=|ΔE(t)|+|ΔC(t)| (6)

[0086] If F adjust If (t) exceeds the preset threshold, it indicates that the current carbon emission and carbon sink adjustment effect is not ideal, and the platform will trigger a re-optimization process to readjust the strategy.

[0087] After receiving real-time feedback, the platform employs a deep learning-based adaptive optimization mechanism, using techniques such as reinforcement learning to train on historical and real-time data to optimize control decisions. The algorithm is based on a deep Q-reinforcement learning model, continuously improving decision accuracy based on environmental changes and data feedback. The optimization formula for the deep Q-learning model (DQN) is as follows:

[0088] Q(s,a)=R(s,a)+γ·max a Q(s′,a′) (7)

[0089] Where Q(s,a) represents the expected reward for taking action a in state s; R(s,a) is the immediate reward; γ is a discount factor that determines the degree of influence of future rewards; s′ is the next state; and a′ is the optimal next action. This model dynamically adjusts carbon emission and carbon sink management measures through real-time data and feedback.

[0090] The platform employs a dynamic control mechanism, utilizing artificial intelligence and optimization algorithms to intelligently predict future trends in carbon emissions and carbon sinks based on real-time observation data. When the platform detects excessive carbon emissions or insufficient carbon sinks, it automatically generates control measures and pushes them to mine managers. These measures include: adjusting mining intensity to reduce carbon emissions; increasing carbon sink compensation areas to enhance carbon sink capacity; and changing production processes or introducing green technologies to reduce carbon emissions. Through this real-time dynamic control mechanism, the platform can effectively address sudden carbon emission issues and provide managers with optimal control solutions.

[0091] With automatic feedback and optimization, the platform tracks the effects of adjustments in real time and adjusts management strategies based on new data feedback. Through this feedback mechanism, the platform can continuously optimize carbon emission and carbon sink management measures to ensure optimal environmental benefits.

[0092] 3. Interactive visualization and decision support

[0093] The interactive visualization and decision support system uses an intuitive, real-time graphical interface to help coal mine managers make intelligent decisions in carbon emission and carbon sink management. Through the interactive interface, users can view real-time trends in carbon emissions and carbon sinks, perform data analysis, adjust management strategies, and see the effects of these adjustments. The platform supports real-time feedback and historical data review, enabling managers to adjust plans and optimize the decision-making process based on different carbon emission targets, carbon sink targets, and specific mine conditions. The working principle is as follows:

[0094] First, there's the data input and real-time display. The platform receives multi-source data (such as carbon emissions, carbon sequestration, and meteorological data), which is then presented in real-time through the front-end interface. Based on the integrated data, the platform displays the following graphical information:

[0095] Carbon emission heat map: Displays the intensity of carbon emissions in different areas of the mining area, helping to identify carbon emission hotspots;

[0096] Carbon sink distribution map: Shows the spatial distribution of carbon sinks within the mining area, helping managers understand the overall carbon sink capacity and which areas have insufficient carbon sinks;

[0097] Carbon Emissions and Carbon Sequestration Dynamic Trend Chart: Based on real-time observation data, this chart displays the changing trends of carbon emissions and carbon sequestration, helping policymakers understand the dynamic balance between carbon emissions and carbon sequestration.

[0098] Interactive adjustment and real-time feedback: Users can adjust control parameters, select different management objectives, and observe the system's response after adjustments through the platform's interactive interface. Specific interactive functions include:

[0099] Carbon emission target adjustment: Users can adjust the carbon emission target (E) of the mining area based on real-time carbon emission data. target (t));

[0100] Carbon offset program adjustment: Users can choose to adjust their carbon offset program and dynamically change the offset area according to the actual situation of carbon offset.

[0101] Decision support and optimization suggestions, based on real-time data and optimization algorithms, not only provide a visual display of monitoring results but also generate optimization suggestions through intelligent algorithms. These suggestions automatically provide adjustment recommendations based on the current carbon emissions and carbon sink status of the mining area and display the expected effects of these adjustments in real time. For example, the system can generate control plans under different scenarios based on historical and current data, and calculate the trends of carbon emissions and carbon sink changes after adjustments, providing decision support for managers.

[0102] In addition to real-time decision support, this embodiment also includes model-based prediction capabilities. The platform can predict future changes in carbon emissions and carbon sinks using machine learning models or time series analysis models based on historical and real-time data. Based on the prediction results, the system automatically adjusts carbon emission and carbon sink targets to avoid exceeding emission limits or insufficient carbon sink capacity. ARIMA (Auto-Regressive Integrated Moving Average) or LSTM (Long Short-Term Memory) models are used for time series prediction. The model output predicted values ​​are... and

[0103]

[0104] Among them, f pollution and f sink It is a predictive model, where t represents the current time and n is the time window length of historical data.

[0105] In the platform's intelligent prediction and optimization module, the system combines historical and real-time data to predict carbon emissions and carbon sinks for a future period. This prediction provides data support for managers, enabling decision-makers to make adjustments in advance.

[0106] Prediction and Adjustment Strategies. Historical Data-Based Prediction: By analyzing historical data (such as carbon emissions and weather conditions), the system uses machine learning models to predict future carbon emissions and carbon sinks. Strategy Generation and Optimization: The platform generates the optimal control plan based on the prediction results and optimizes the plan to ensure that the set carbon emission and carbon sink targets are achieved within the future timeframe.

[0107] The platform optimizes the mathematical model for proposed solutions, generating the best strategy based on the model to minimize the bias between carbon emissions and carbon sinks. The optimization objective is:

[0108]

[0109] The objective function aims to minimize the difference between carbon emissions and carbon sink targets simultaneously, and finds the optimal solution through optimization algorithms (such as gradient descent, linear programming, etc.).

[0110] The platform can monitor carbon emissions and carbon sinks in coal mining areas in real time and automatically adjust management measures through a dynamic control mechanism to ensure rapid response to different environmental changes. In addition, the platform provides an interactive visual interface, allowing managers to operate intuitively, adjust management plans in real time, and generate optimization suggestions. The platform combines integrated observation technology from sky, tower, ground, and well sites with advanced data fusion algorithms to ensure high accuracy and efficiency in carbon emission and carbon sink management.

[0111] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A real-time dynamic interaction and regulation platform for carbon flux based on "sky tower well" observation, characterized in that, The "sky tower and ground well" comprehensive observation and data acquisition module, a regulation and control module, and an interactive visualization and decision support module are comprised; The "sky tower and ground well" comprehensive observation and data acquisition module comprises a satellite remote sensing module, a UAV remote sensing module, a vorticity covariance flux tower, a ground observation module, and a well observation module; The regulation and control module adjusts and controls carbon emissions and / or carbon sinks in real time according to the data obtained by the data acquisition module, reduces carbon emissions, and improves carbon sink capacity; The interactive visualization and decision support module provides support for intelligent decision-making in the process of carbon emission and carbon sink management through intuitive and real-time graphical interfaces; The carbon emission control strategy of the regulation and control module is as follows: ， wherein, represents the intensity of carbon emission regulation; a is the carbon emission regulation coefficient, and ΔE(t) is the carbon emission deviation value; The carbon sink control strategy of the regulation and control module is as follows: ， wherein, represents the intensity of carbon sink regulation; β is the carbon sink regulation coefficient, and ΔC(t) is the carbon sink deviation value; Based on formula (3) and formula (4), the regulation and control module adjusts the carbon emission and carbon sink targets according to real-time observation data and generates a corresponding control scheme; The regulation and control module includes intelligent optimization of the control scheme, and the intelligent optimization target of the control scheme is: ， The above formula represents the minimization of the square sum of the deviation of carbon emissions and carbon sinks; The intelligent optimization of the control scheme adopts a self-adaptive optimization mechanism based on deep learning according to the feedback results, trains through historical reinforcement learning and real-time data, and optimizes the control decision; The feedback results are as follows: ， When F adjust (t) the control module triggers intelligent optimization when exceeding a preset threshold; The intelligent optimization adopts a deep Q reinforcement learning model, and the optimization formula of the deep Q reinforcement learning model is as follows: ， Wherein, Q(s,a) represents the expected reward of taking action a in state s; R(s,a) is the immediate reward; γ is the discount factor, which determines the degree of influence of future rewards; s' is the next state, and a' is the optimal next action.

2. The real-time dynamic interaction and regulation platform for carbon flux based on "sky tower well" observation according to claim 1, characterized in that, The carbon emission deviation is calculated as follows: ， wherein, ΔE(t) is the carbon emission deviation at time t; E pollution (t) is the actual carbon emission concentration; E target (t) is the set target carbon emission concentration; The carbon sink deviation is calculated as follows: ， Wherein, ΔC(t) is the carbon sink deviation at time t; C sink (t) is the actual carbon sink capacity; C target (t) is the set target carbon sink capacity. 3.The real-time dynamic interaction and regulation platform of carbon flux based on "sky tower and ground well" observation according to claim 1, wherein, The interactive visualization and decision support module comprises: Data input and real-time display, real-time presentation of received multi-source data through the front-end interface, and display of the following graphical information based on the fused data: carbon emission heat map, carbon sink distribution map, and carbon emission and carbon sink dynamic change trend map; Interactive adjustment and real-time feedback, providing an interactive interface for adjusting control parameters and selecting different management targets, and displaying the response after adjustment, wherein the control parameters include carbon emission targets and carbon sink compensation plans; Decision support and optimization suggestions, based on real-time data and optimization algorithms, the platform not only provides visual display of monitoring results, but also generates optimization suggestions through intelligent algorithms, including automatically providing adjustment suggestions based on the current carbon emission and carbon sink conditions of the mining area, and real-time display of the expected effects of these adjustments.

4. The real-time dynamic interaction and regulation platform for carbon flux based on "sky tower well" observation according to claim 3, characterized in that, The decision support and optimization suggestions include model prediction and decision support, which predict future changes in carbon emissions and carbon sinks based on historical data and real-time data through machine learning models or time series analysis models, automatically adjust the targets of carbon emissions and carbon sinks according to the prediction results, and avoid excessive carbon emissions or insufficient carbon sink capacity.

5. The real-time dynamic interactive and regulation platform of carbon flux based on "sky tower well" observation according to claim 4, characterized in that, The model prediction adopts ARIMA or LSTM models for time series prediction, and the ARIMA or LSTM model outputs the predicted values as follows: ， where f pollution and f sink are prediction models, t represents the current time, and n is the length of the time window of historical data.

6. The real-time dynamic interactive and regulation platform of carbon flux based on "sky tower well" observation according to claim 5, characterized in that, The decision support and optimization recommendations generate optimal strategies based on model optimization, with the optimization goal of the optimal strategies as follows: （9） The above formula aims to minimize the difference between carbon emission and carbon sink targets at the same time, and the optimal solution is found through an optimization algorithm, which is gradient descent or linear programming.

Images