Method and system equipment for intelligently analyzing carbon emission based on abnormal carbon factors in smart park in combination with energy storage, and storage medium

Abstract

The invention provides a method and system equipment for intelligently analyzing carbon emission based on abnormal carbon factors in a smart park in combination with energy storage, and a storage medium, and relates to the field of digital energy storage analysis. The method comprises the following steps: firstly, capturing an abnormal carbon factor in real time by combining a multi-time scale weighted threshold value and a fanotify technology through an intelligent electric meter comprising an intelligent sensing module and a carbon monitoring module; constructing four types of energy storage matrixes of hydropower, photovoltaic, wind power and non-green power, and predicting carbon factor response lag characteristics by using an NBP method; and finally, global simulation deduction is carried out through an MCTS algorithm, and an optimal energy-saving strategy is selected according to a comprehensive evaluation function. The system correspondingly comprises an intelligent electric meter deployment layer, an abnormal carbon factor sensing system and other modules. According to the invention, accurate real-time monitoring of abnormal carbon factors is realized, power grid cleanliness changes are dynamically adapted, a global optimal strategy is output, the park carbon management efficiency is improved, and the method is suitable for various smart park low-carbon management scenes.

Description

Technical Field

[0001] This invention relates to the field of digital energy storage analysis, and in particular to a method, system equipment, and storage medium for intelligent analysis of carbon emissions based on abnormal carbon factors in smart parks combined with energy storage. Background Technology

[0002] Two core issues exist in carbon emission management in smart industrial parks: 1) Traditional methods cannot detect and capture sudden abnormal carbon emission factors in the park in real time and accurately; 2) Existing carbon emission optimization strategies (such as traditional time-of-use pricing) are difficult to dynamically respond to changes in grid cleanliness (such as fluctuations in wind and solar power generation), resulting in decision-making lag and a tendency to fall into local optima. They cannot effectively guide energy storage scheduling and energy-saving strategies to achieve globally optimal carbon emission reduction. Summary of the Invention

[0003] Purpose of the invention: To address the difficulty of real-time sensing of abnormal carbon emission factors, this invention proposes a method for analyzing carbon emissions based on a combination of anomalous carbon factors in smart parks and energy storage. By deploying smart meters, combined with carbon monitoring instruments and a dynamic carbon factor engine based on Fanotify technology, the method monitors changes in carbon emission profiles in key areas in real time. It also accurately identifies abnormal carbon factors based on a weighted average threshold across multiple time scales (short-term / medium-term / long-term), thereby solving the problems of lag and underreporting in traditional monitoring methods and achieving proactive, real-time capture of abnormal carbon emission factors.

[0004] In a first aspect, this invention proposes a method for analyzing carbon emissions based on abnormal carbon factors in smart parks combined with energy storage, comprising the following steps: S1. Deploy smart meters in the public areas of the smart park and on each floor of the buildings. The smart meters integrate a smart sensing module and a carbon monitoring module. The smart sensing module calls the carbon monitor to collect real-time carbon emission factor data of the target monitoring area, and determines whether the real-time carbon emission factor data is an abnormal carbon factor based on a multi-timescale weighted average threshold. If it is determined to be an abnormal carbon factor, the smart sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine. The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors the changes in the configuration file storing carbon emission data in the key protection area in real time. If the change in the configuration file exceeds the preset carbon factor abnormal threshold, the execution component synchronizes the configuration file to the dynamic carbon factor engine to complete the perception and capture of abnormal carbon factors. S2. By analyzing the historical carbon emission logs and energy storage operation logs through the carbon monitoring module, the power generation source information of the energy storage system is extracted. The power generation sources include hydropower, photovoltaic, wind power and non-green electricity. Based on the energy characteristics and energy storage capacity of each power generation source, hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix and non-green electricity energy storage matrix are constructed respectively. S3. The subsequent behavior prediction (NBP) method is used to predict the carbon emission response lag characteristics corresponding to the abnormal carbon factors in the four energy storage matrices and output the carbon emission response lag probability. The NBP method is based on the historical change pattern of the abnormal carbon factors to construct a carbon factor behavior prediction model. S4. Based on the Monte Carlo tree search algorithm, a global optimization simulation is performed on the energy storage matrix where the abnormal carbon factors are located. During the simulation, three key indicators are recorded in real time: the accuracy of abnormal carbon factors, the carbon emission growth rate, and the number of abnormal carbon factors processed per unit time. Based on the preset comprehensive evaluation function, the scores of each candidate energy-saving strategy are calculated, and the energy-saving strategy with the highest score is selected as the optimal solution for carbon emission control in the smart park.

[0005] In a further embodiment of the first aspect, the process of constructing the multi-timescale weighted average threshold in step S1 includes: Extract the historical log data of carbon emission factors recorded by the carbon monitor, and calculate the weighted average for each group of historical data using the weighting rule of "recent data has a higher weight than long-term data", which will be used as the short-term threshold, medium-term threshold and long-term threshold respectively. The triggering condition for determining that the real-time carbon emission factor data is an abnormal carbon factor is: the current carbon emission factor data exceeds the short-term threshold and the medium-term threshold at the same time, or exceeds the short-term threshold, the medium-term threshold and the long-term threshold at the same time.

[0006] In a further embodiment of the first aspect, step S1 further includes an abnormal power consumption calculation step: Abnormal power consumption in the target area is calculated based on the formula "abnormal power consumption = abnormal carbon emission factor / power consumption coefficient corresponding to unit carbon emission". The power consumption coefficient corresponding to unit carbon emission is obtained by converting the carbon emission factor of the energy type in the park. When the carbon monitor collects carbon emission factor data, it first divides the target monitoring area into 10m×10m grid units and generates monitoring data files in the format of "grid number-collection time-carbon factor value". The fanotify component monitors the modification, deletion and addition operations of the corresponding data files according to the grid units to ensure that abnormal carbon factor data is captured without omission.

[0007] In a further embodiment of the first aspect, the photovoltaic energy storage matrix described in step S2 is constructed as follows: Number of photovoltaic arrays in series ;in The rated voltage of the photovoltaic grid connection node. This refers to the operating voltage of a single photovoltaic cell. Total voltage of photovoltaic array ; Real-time output power of photovoltaic array ;in This represents the number of parallel series in the photovoltaic array. This refers to the operating current of a single photovoltaic cell. Photovoltaic array number - real-time output power A photovoltaic energy storage matrix is ​​constructed using "-remaining energy storage capacity-carbon factor value" as a column vector.

[0008] In a further embodiment of the first aspect, the wind power energy storage matrix is ​​constructed in step S2 as follows: Constructing wind power generation to satisfy a piecewise function: When wind speed V < V c Or V > V F At that time, P=0; When V c ≤V≤V R When, P=P R ×(VV c ) / (V R -V c ); When V R <V≤V F When, P=P R ; Where V is the real-time wind speed, V c To cut off the wind speed, V F To cut off the wind speed, V R Where P is the rated wind speed, and P is the real-time power. R Rated power; A wind power energy storage matrix is ​​constructed using "wind turbine number - real-time wind speed V - real-time power P - remaining energy storage capacity - carbon factor value" as column vectors.

[0009] In a further embodiment of the first aspect, the hydropower storage matrix described in step S2 is constructed as follows: Hydropower generation capacity is estimated using incident wave energy. Where C is an empirical coefficient, calibrated according to the hydrological conditions of the park area; H is the significant wave height. The peak period; A hydropower energy storage matrix is ​​constructed using "hydropower station unit number - real-time wave height - real-time power - remaining energy storage capacity - carbon factor value" as column vectors; The non-green energy storage matrix is ​​constructed as follows: The response lag quantification of non-green energy storage is based on abnormal CPU power consumption caused by supply voltage fluctuations, and CPU power consumption. ,in Here, C is the power consumption correction factor, C is the CPU chip capacitor, V is the supply voltage, c is the voltage influence index, and f is the CPU operating frequency. A non-green energy storage matrix is ​​constructed using "non-green equipment number - real-time voltage - real-time power consumption - remaining energy storage capacity - carbon factor value" as column vectors.

[0010] In a further embodiment of the first aspect, the loss function of the NBP method in step S3 is cross-entropy, expressed as:

[0011] in, A training dataset for predicting the subsequent behavior of abnormal carbon factors collected by a carbon monitor over the past 6 months; This represents the actual probability of occurrence within the k-th time step in the historical data. is the probability of occurrence predicted by the model at the k-th time step; K is the number of subsequent time steps predicted; p is the number of negative samples.

[0012] In a further embodiment of the first aspect, step S4 specifically includes: S4-1. Set the number of simulations to 100. Candidate energy-saving strategies include intelligent lighting control, air conditioning system energy saving, and elevator group control optimization. S4-2. For each candidate strategy, perform 100 Monte Carlo simulations in the energy storage matrix where the anomalous carbon factor is located. Each simulation lasts for 24 hours. Record the anomalous carbon factor accuracy, carbon emission growth rate, and MTTR for each simulation. Wherein, the accuracy of anomalous carbon factors = the number of times anomalous carbon factors are correctly identified / the total number of anomalous carbon factors; the carbon emission growth rate = the increase in carbon emissions during the simulation period / the initial carbon emissions; and the MTTR = the total time for handling anomalous carbon factors during the simulation period / the total number of anomalous carbon factors. S4-3. Calculate the average score of each strategy based on the comprehensive evaluation function, which is:

[0013] The baseline value for MTTR is set to 300s. When MTTR ≤ 300s, The value is ≥0; when MTTR > 300s, Value < 0; S4-4. Compare the average scores of each candidate strategy and select the strategy with the highest score as the final carbon emission control scheme.

[0014] A second aspect of the present invention proposes a system for intelligent carbon emission analysis based on abnormal carbon factors in smart parks combined with energy storage, the system comprising: The smart meter deployment layer is used to deploy smart meters in the public areas of the smart park and on each floor of the buildings. The smart meters integrate a smart sensing module and a carbon monitoring module. The smart sensing module calls the carbon monitor to collect real-time carbon emission factor data of the target monitoring area, and determines whether the real-time carbon emission factor data is an abnormal carbon factor based on a multi-time-scale weighted average threshold. If it is determined to be an abnormal carbon factor, the smart sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine. The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors the configuration file changes of the carbon emission data stored in the key protection area in real time. If the change of the configuration file exceeds the preset carbon factor abnormality threshold, the execution component synchronizes the configuration file to the dynamic carbon factor engine to complete the perception and capture of abnormal carbon factors. The energy storage matrix management system includes a log analysis unit and a matrix construction unit. The log analysis unit parses carbon emission logs and energy storage operation logs to extract information on four types of power generation sources: hydropower, photovoltaic, wind power, and non-green electricity. The matrix construction unit constructs hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix, and non-green electricity energy storage matrix based on the energy parameters of each power generation source. The carbon factor prediction and optimization system includes an NBP prediction unit and an MCTS optimization unit. The NBP prediction unit uses a subsequent behavior prediction model to output the carbon emission response lag probability. The MCTS optimization unit performs Monte Carlo simulation on the energy storage matrix where the abnormal carbon factor is located, records the accuracy of the abnormal carbon factor, the carbon emission growth rate and MTTR, and calculates a score based on the evaluation function. The decision output system compares the scores of each candidate energy-saving strategy, outputs the optimal carbon emission control strategy with the highest score, and sends the strategy instruction to the park's energy control terminal.

[0015] A third aspect of the present invention provides an electronic device comprising: a processor and a memory storing computer program instructions; wherein the processor, when executing the computer program instructions, implements the method for intelligent analysis of carbon emissions based on abnormal carbon factors in smart parks combined with energy storage, as described in the first aspect.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium storing at least one executable instruction that, when executed on an electronic device, causes the electronic device to perform the method for intelligent analysis of carbon emissions based on abnormal carbon factors in smart parks combined with energy storage, as described in the first aspect.

[0017] Beneficial effects: (1) By combining the smart sensing module integrated in the smart meter with the dynamic carbon factor engine, and using the fanotify technology to monitor the carbon emission profile of key areas in real time, abnormal changes in carbon factors (such as a sudden surge in carbon emissions from high-energy-consuming equipment) can be actively captured, avoiding “missed reports or delayed reports” caused by traditional manual inspections or timed sampling.

[0018] (2) By constructing four independent energy storage matrices for hydropower, photovoltaic, wind power and non-green electricity, the carbon characteristics of different power generation sources are accurately correlated (such as the cleanliness fluctuation of photovoltaic / wind power and the high carbon emission attributes of non-green electricity). Combined with the prediction of the lag characteristics of carbon emission response by NBP subsequent behavior prediction, the grid cleanliness changes can be dynamically matched (such as prioritizing the dispatch of photovoltaic energy storage during the peak of photovoltaic output during the day and reducing the use of non-green electricity when the proportion of thermal power is high at night), avoiding the drawbacks of the traditional fixed time-of-use electricity price that "does not care about grid cleanliness and controls energy in a one-size-fits-all manner".

[0019] (3) The Monte Carlo Tree Search (MCTS) algorithm is used to perform 100 global simulations of the energy storage matrix. The strategy effect is quantified by the comprehensive evaluation function (Score) of “abnormal carbon factor accuracy + carbon emission growth rate + MTTR”, which can screen out the global optimal energy-saving scheme and improve the global carbon emission reduction efficiency. Attached Figure Description

[0020] Figure 1 This is an overall flowchart of the method for analyzing carbon emissions based on abnormal carbon factors in smart parks combined with energy storage in the embodiments.

[0021] Figure 2 This is a schematic diagram of the execution flow of the dynamic carbon factor engine in the embodiment. Detailed Implementation

[0022] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0023] To address the problem that carbon emission optimization strategies struggle to dynamically respond to clean energy fluctuations and are prone to getting trapped in local optima, this embodiment proposes constructing four independent matrices for the energy storage system based on power generation sources (hydropower, photovoltaic, wind power, and non-green electricity). It then performs subsequent behavior prediction (NBP) on captured anomalous carbon factors and employs a Monte Carlo Tree Search (MCTS) algorithm to globally optimize and simulate the matrices containing anomalous carbon factors. By designing a comprehensive evaluation function (Score), the performance of different energy-saving strategies in terms of carbon factor accuracy, carbon emission growth rate, and response time (MTTR) is quantitatively compared. This solves the pain points of traditional time-of-use pricing mechanisms being rigid and unable to adapt to real-time changes in grid cleanliness, and overcomes the shortcomings of traditional optimization methods that easily obtain local optima. This achieves intelligent energy-saving strategy decision-making based on global optima and dynamically adapting to clean energy fluctuations.

[0024] See Figure 1 As shown, the implementation details of the solution proposed in this embodiment are as follows: S1. Deploy smart meters in the public areas of the smart park and on each floor of the buildings. The smart meters integrate a smart sensing module and a carbon monitoring module. The smart sensing module calls the carbon monitor to collect real-time carbon emission factor data for the target monitoring area, and determines whether the real-time carbon emission factor data is abnormal based on a multi-timescale weighted average threshold. If it is determined to be an abnormal carbon factor, the smart sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine (the execution flow of the dynamic carbon factor engine is described in...). Figure 2 The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors changes in the configuration file storing carbon emission data in key protection areas in real time. If the change in the configuration file exceeds the preset carbon factor anomaly threshold, the execution component will synchronize the configuration file to the dynamic carbon factor engine to complete the detection and capture of abnormal carbon factors.

[0025] S2. By analyzing historical carbon emission logs and energy storage operation logs through the carbon monitoring module, the power generation source information of the energy storage system is extracted. The power generation sources include hydropower, photovoltaic, wind power and non-green electricity. Based on the energy characteristics and energy storage capacity of each power generation source, hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix and non-green electricity energy storage matrix are constructed respectively. S3. The subsequent behavior prediction (NBP) method is used to predict the carbon emission response lag characteristics corresponding to the abnormal carbon factors in the four energy storage matrices and output the carbon emission response lag probability. The NBP method uses the historical change pattern of the abnormal carbon factors as the training basis to construct a carbon factor behavior prediction model. S4. Based on the Monte Carlo tree search algorithm, a global optimization simulation is performed on the energy storage matrix where the abnormal carbon factors are located. During the simulation, three key indicators are recorded in real time: the accuracy of abnormal carbon factors, the carbon emission growth rate, and the number of abnormal carbon factors processed per unit time. Based on the preset comprehensive evaluation function, the scores of each candidate energy-saving strategy are calculated, and the energy-saving strategy with the highest score is selected as the optimal solution for carbon emission control in the smart park.

[0026] As a preferred approach, the process of constructing the multi-time-scale weighted average threshold in step S1 includes: Extract historical log data of carbon emission factors from the carbon monitoring instrument for the last 10, 100, and 1000 times. Calculate the weighted average for each set of historical data using the weighting rule that "recent data has a higher weight than long-term data", and use these as the short-term, medium-term, and long-term thresholds, respectively. The triggering condition for determining real-time carbon emission factor data as abnormal carbon factors is: the current carbon emission factor data exceeds the short-term threshold and the medium-term threshold at the same time, or exceeds the short-term threshold, the medium-term threshold and the long-term threshold at the same time.

[0027] As a preferred embodiment, step S1 also includes an abnormal power consumption calculation step: Abnormal power consumption in the target area is calculated based on the formula "abnormal power consumption = abnormal carbon emission factor / power consumption coefficient corresponding to unit carbon emission". The power consumption coefficient corresponding to unit carbon emission is obtained by converting the carbon emission factor of the energy type in the park. When the carbon monitor collects carbon emission factor data, it first divides the target monitoring area into 10m×10m grid units and generates monitoring data files in the format of "grid number-collection time-carbon factor value". The fanotify component monitors the modification, deletion and addition operations of the corresponding data files according to the grid units to ensure that no abnormal carbon factor data is missed.

[0028] As a preferred option, the photovoltaic energy storage matrix is ​​constructed in step S2 as follows: Number of photovoltaic arrays in series ;in The rated voltage of the photovoltaic grid connection node. This refers to the operating voltage of a single photovoltaic cell. Total voltage of photovoltaic array ; Real-time output power of photovoltaic array ;in This represents the number of parallel series in the photovoltaic array. This refers to the operating current of a single photovoltaic cell. Photovoltaic array number - real-time output power A photovoltaic energy storage matrix is ​​constructed using "-remaining energy storage capacity-carbon factor value" as a column vector.

[0029] As a preferred option, the wind power energy storage matrix is ​​constructed in step S2 as follows: Constructing wind power generation to satisfy a piecewise function: When wind speed V < V c Or V > V F At that time, P=0; When V c ≤V≤V R When, P=P R ×(VV c ) / (V R -V c ); When V R <V≤V F When, P=P R .

[0030] Where V is the real-time wind speed, V c To cut off the wind speed, V F To cut off the wind speed, V R Where P is the rated wind speed, and P is the real-time power. R Rated power; A wind power energy storage matrix is ​​constructed using "wind turbine number - real-time wind speed V - real-time power P - remaining energy storage capacity - carbon factor value" as column vectors.

[0031] As a preferred option, the hydropower storage matrix is ​​constructed in step S2 as follows: Hydropower generation capacity is estimated using incident wave energy. Where C is an empirical coefficient, calibrated according to the hydrological conditions of the park area; H is the significant wave height. The peak period; A hydropower energy storage matrix is ​​constructed using "hydropower station unit number - real-time wave height - real-time power - remaining energy storage capacity - carbon factor value" as column vectors.

[0032] As a preferred option, the non-green energy storage matrix is ​​constructed in step S2 as follows: The response lag quantification of non-green energy storage is based on abnormal CPU power consumption caused by supply voltage fluctuations, and CPU power consumption. ,in Here, C is the power consumption correction factor, C is the CPU chip capacitor, V is the supply voltage, c is the voltage influence index, and f is the CPU operating frequency. A non-green energy storage matrix is ​​constructed using "non-green equipment number - real-time voltage - real-time power consumption - remaining energy storage capacity - carbon factor value" as column vectors.

[0033] As a preferred approach, the loss function of the NBP method in step S3 is cross-entropy, expressed as:

[0034] in, A training dataset for predicting the subsequent behavior of abnormal carbon factors collected by a carbon monitor over the past 6 months; This represents the actual probability of occurrence within the k-th time step in the historical data. is the probability of occurrence predicted by the model at the k-th time step; K is the number of subsequent time steps predicted; p is the number of negative samples.

[0035] As a preferred embodiment, step S4 specifically includes: The simulation was set to run 100 times. The candidate energy-saving strategies included intelligent lighting control, air conditioning system energy saving, and elevator group control optimization. For each candidate strategy, 100 Monte Carlo simulations were performed in the energy storage matrix where the abnormal carbon factor was located. Each simulation lasted for 24 hours. The accuracy of the abnormal carbon factor, the carbon emission growth rate, and the MTTR of each simulation were recorded. Where: Abnormal carbon factor accuracy = number of times abnormal carbon factors were correctly identified / total number of abnormal carbon factors; carbon emission growth rate = increase in carbon emissions during the simulation period / initial carbon emissions; MTTR = total time spent handling abnormal carbon factors during the simulation period / total number of abnormal carbon factors; the average score of each strategy is calculated based on the comprehensive evaluation function, which is:

[0036] The baseline value for MTTR is set to 300s. When MTTR ≤ 300s, The value is ≥0; when MTTR > 300s, Value < 0; By comparing the average scores of each candidate strategy, the strategy with the highest score is selected as the final carbon emission control solution.

[0037] See Table 1 for examples of the results.

[0038] Table 1. Node strategies corresponding to different scores

[0039] Decision output: Select energy-saving strategy A (highest score).

[0040] Strategy A – Intelligent Lighting Control: This strategy replaces traditional, inefficient light sources with LED lighting fixtures and utilizes intelligent control technologies such as sensor control, light sensor control, and time control to achieve precise adjustment and intelligent management of lighting based on actual needs. Lights automatically turn off when no one is present and automatically turn on when someone is present, avoiding energy waste.

[0041] Strategy B – Energy Saving in Air Conditioning Systems: Optimizes air conditioning system control through intelligent algorithms, automatically adjusting operating status based on indoor and outdoor temperature and humidity. For example, using water-based cooling technology stores energy during off-peak hours and releases it during peak hours, saving 30%-60% of electricity.

[0042] Strategy C – Elevator Group Control Optimization: Achieve intelligent elevator scheduling by rationally allocating elevator operation based on usage frequency and passenger flow, reducing idle time and energy consumption.

[0043] The underlying logic of the methods disclosed in the above embodiments can be implemented as a system. This system can consist of a smart meter deployment layer, an energy storage matrix management system, a carbon factor prediction and optimization system, and a decision output system.

[0044] The smart meter deployment layer is used to deploy smart meters in the public areas of the smart park and on each floor of the buildings. The smart meters integrate a smart sensing module and a carbon monitoring module. The smart sensing module calls the carbon monitor to collect real-time carbon emission factor data of the target monitoring area, and determines whether the real-time carbon emission factor data is an abnormal carbon factor based on a multi-time-scale weighted average threshold. If it is determined to be an abnormal carbon factor, the smart sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine. The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors changes in the configuration file storing carbon emission data in the key protection area in real time. If the change in the configuration file exceeds the preset carbon factor abnormality threshold, the execution component synchronizes the configuration file to the dynamic carbon factor engine to complete the perception and capture of abnormal carbon factors.

[0045] The energy storage matrix management system includes a log analysis unit and a matrix construction unit. The log analysis unit parses carbon emission logs and energy storage operation logs to extract information on four types of power generation sources: hydropower, photovoltaic, wind power, and non-green electricity. The matrix construction unit constructs hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix, and non-green electricity energy storage matrix based on the energy parameters of each power generation source.

[0046] The carbon factor prediction and optimization system includes an NBP prediction unit and an MCTS optimization unit. The NBP prediction unit uses a subsequent behavior prediction model to output the carbon emission response lag probability. The MCTS optimization unit performs Monte Carlo simulation on the energy storage matrix where the abnormal carbon factor is located, records the accuracy of the abnormal carbon factor, the carbon emission growth rate and MTTR, and calculates a score based on the evaluation function.

[0047] The decision output system compares the scores of each candidate energy-saving strategy, outputs the optimal carbon emission control strategy with the highest score, and sends the strategy instruction to the park's energy control terminal.

[0048] Furthermore, the logical ideas behind the methods disclosed in the above embodiments can be implemented, in whole or in part, through software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product. A computer program product includes one or more computer instructions or computer programs.

[0049] When computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives (SSDs).

[0050] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0051] In summary, existing technologies in smart park carbon emission management have not effectively addressed the issues of real-time and accurate sensing and capture of abnormal carbon factors, nor have they established a dynamic carbon emission optimization decision-making mechanism capable of responding to changes in grid cleanliness and ensuring global optimality. Based on these key technological challenges, this invention proposes an innovative solution that integrates real-time anomaly sensing, multi-source energy storage matrix analysis, subsequent behavior prediction, and Monte Carlo tree search for global optimization.

[0052] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. Although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the invention itself. Various changes in form and detail may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for analyzing carbon emissions based on abnormal carbon factors in smart parks combined with energy storage intelligence, characterized in that, Includes the following steps: S1. Deploy smart meters in the public areas of the smart park and on each floor of the buildings. The smart meters integrate a smart sensing module and a carbon monitoring module. The smart sensing module calls the carbon monitor to collect real-time carbon emission factor data of the target monitoring area, and determines whether the real-time carbon emission factor data is an abnormal carbon factor based on a multi-timescale weighted average threshold. If it is determined to be an abnormal carbon factor, the smart sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine. The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors the changes in the configuration file storing carbon emission data in the key protection area in real time. If the change in the configuration file exceeds the preset carbon factor abnormal threshold, the execution component synchronizes the configuration file to the dynamic carbon factor engine to complete the perception and capture of abnormal carbon factors. S2. By analyzing the historical carbon emission logs and energy storage operation logs through the carbon monitoring module, the power generation source information of the energy storage system is extracted. The power generation sources include hydropower, photovoltaic, wind power and non-green electricity. Based on the energy characteristics and energy storage capacity of each power generation source, hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix and non-green electricity energy storage matrix are constructed respectively. S3. The subsequent behavior prediction (NBP) method is used to predict the carbon emission response lag characteristics corresponding to the abnormal carbon factors in the four energy storage matrices and output the carbon emission response lag probability. S4. Based on the Monte Carlo tree search algorithm, a global optimization simulation is performed on the energy storage matrix where the abnormal carbon factors are located. During the simulation, the accuracy of abnormal carbon factors, the carbon emission growth rate and the number of abnormal carbon factors processed per unit time are recorded in real time. The scores of each candidate energy-saving strategy are calculated based on a preset comprehensive evaluation function, and the energy-saving strategy with the highest score is selected as the optimal solution for carbon emission control in the smart park.

2. The method according to claim 1, characterized in that, The process of constructing the multi-time-scale weighted average threshold in step S1 includes: Extract historical log data of carbon emission factors recorded by the carbon monitor, and calculate the weighted average for each group of historical data using the weighting rule of "recent data has a higher weight than long-term data", which will be used as the short-term threshold, medium-term threshold and long-term threshold respectively. The triggering condition for determining that the real-time carbon emission factor data is an abnormal carbon factor is: the current carbon emission factor data exceeds the short-term threshold and the medium-term threshold at the same time, or exceeds the short-term threshold, the medium-term threshold and the long-term threshold at the same time.

3. The method according to claim 1, characterized in that, Step S1 also includes an abnormal power consumption calculation step: Abnormal power consumption in the target area is calculated based on the formula "abnormal power consumption = abnormal carbon emission factor / power consumption coefficient corresponding to unit carbon emission". The power consumption coefficient corresponding to unit carbon emission is obtained by converting the carbon emission factor of the energy type in the park. When the carbon monitor collects carbon emission factor data, it first divides the target monitoring area into 10m×10m grid units and generates monitoring data files in the format of "grid number-collection time-carbon factor value". The fanotify component monitors the modification, deletion and addition operations of the corresponding data files according to the grid units to ensure that abnormal carbon factor data is captured without omission.

4. The method according to claim 1, characterized in that, The photovoltaic energy storage matrix described in step S2 is constructed as follows: Number of photovoltaic arrays in series ;in The rated voltage of the photovoltaic grid connection node. This refers to the operating voltage of a single photovoltaic cell. Total voltage of photovoltaic array ; Real-time output power of photovoltaic array ;in This represents the number of parallel series in the photovoltaic array. This refers to the operating current of a single photovoltaic cell. Photovoltaic array number - real-time output power A photovoltaic energy storage matrix is ​​constructed using "-remaining energy storage capacity-carbon factor value" as a column vector. The wind power energy storage matrix is ​​constructed as follows: Constructing wind power generation to satisfy a piecewise function: When wind speed V < V c Or V > V F At that time, P=0; When V c ≤V≤V R When, P=P R ×(VV c ) / (V R -V c ); When V R <V≤V F When, P=P R ; Where V is the real-time wind speed, V c To cut off the wind speed, V F To cut off the wind speed, V R Where P is the rated wind speed, and P is the real-time power. R Rated power; A wind power energy storage matrix is ​​constructed using "wind turbine number - real-time wind speed V - real-time power P - remaining energy storage capacity - carbon factor value" as column vectors.

5. The method according to claim 1, characterized in that, The construction method of the hydropower storage matrix in step S2 is as follows: Hydropower generation capacity is estimated using incident wave energy. Where C is an empirical coefficient, calibrated according to the hydrological conditions of the park area; H is the significant wave height. The peak period; A hydropower energy storage matrix is ​​constructed using "hydropower station unit number - real-time wave height - real-time power - remaining energy storage capacity - carbon factor value" as column vectors; The non-green energy storage matrix is ​​constructed as follows: The response lag quantification of non-green energy storage is based on abnormal CPU power consumption caused by supply voltage fluctuations, and CPU power consumption. ,in Here, C is the power consumption correction factor, C is the CPU chip capacitor, V is the supply voltage, c is the voltage influence index, and f is the CPU operating frequency. A non-green energy storage matrix is ​​constructed using "non-green equipment number - real-time voltage - real-time power consumption - remaining energy storage capacity - carbon factor value" as column vectors.

6. The method according to claim 1, characterized in that, The loss function of the NBP method described in step S3 is cross-entropy, and its expression is: in, A training dataset for predicting the subsequent behavior of abnormal carbon factors collected by a carbon monitor over the past 6 months; This represents the actual probability of occurrence within the k-th time step in the historical data. is the probability of occurrence predicted by the model at the k-th time step; K is the number of subsequent time steps predicted; p is the number of negative samples.

7. The method according to claim 1, characterized in that, Step S4 specifically includes: S4-1. Set up simulation times. Candidate energy-saving strategies include intelligent lighting control, air conditioning system energy saving, and elevator group control optimization. S4-2. For each candidate strategy, perform 100 Monte Carlo simulations in the energy storage matrix where the anomalous carbon factor is located. Each simulation lasts for 24 hours. Record the anomalous carbon factor accuracy, carbon emission growth rate, and MTTR for each simulation. Wherein, the accuracy of anomalous carbon factors = the number of times anomalous carbon factors are correctly identified / the total number of anomalous carbon factors; the carbon emission growth rate = the increase in carbon emissions during the simulation period / the initial carbon emissions; and the MTTR = the total time for handling anomalous carbon factors during the simulation period / the total number of anomalous carbon factors. S4-3. Calculate the average score of each strategy based on the comprehensive evaluation function, which is: The baseline value for MTTR is set to 300s. When MTTR ≤ 300s, The value is ≥0; when MTTR > 300s, Value < 0; S4-4. Compare the average scores of each candidate strategy and select the strategy with the highest score as the final carbon emission control scheme.

8. A system for intelligent carbon emission analysis based on abnormal carbon factors in smart parks combined with energy storage, characterized in that, include: The smart meter deployment layer is used to deploy smart meters in the public areas of the smart park and on each floor of the building. The smart meters integrate a smart sensing module and a carbon monitoring module. The intelligent sensing module calls the carbon monitor to collect real-time carbon emission factor data of the target monitoring area, and determines whether the real-time carbon emission factor data is an abnormal carbon factor based on a multi-timescale weighted average threshold. If it is determined to be an abnormal carbon factor, the intelligent sensing module transmits the abnormal carbon factor to the dynamic carbon factor engine. The dynamic carbon factor engine has a built-in fanotify component and an execution component. The fanotify component monitors the changes in the configuration file storing carbon emission data in the key protection area in real time. If the change in the configuration file exceeds the preset carbon factor abnormal threshold, the execution component synchronizes the configuration file to the dynamic carbon factor engine to complete the perception and capture of abnormal carbon factors. The energy storage matrix management system includes a log analysis unit and a matrix construction unit. The log analysis unit parses carbon emission logs and energy storage operation logs to extract information on four types of power generation sources: hydropower, photovoltaic, wind power, and non-green electricity. The matrix construction unit constructs hydropower energy storage matrix, photovoltaic energy storage matrix, wind power energy storage matrix, and non-green electricity energy storage matrix based on the energy parameters of each power generation source. The carbon factor prediction and optimization system includes an NBP prediction unit and an MCTS optimization unit. The NBP prediction unit uses a subsequent behavior prediction model to output the carbon emission response lag probability. The MCTS optimization unit performs Monte Carlo simulation on the energy storage matrix where the abnormal carbon factor is located, records the accuracy of the abnormal carbon factor, the carbon emission growth rate and MTTR, and calculates a score based on the evaluation function. The decision output system compares the scores of each candidate energy-saving strategy, outputs the optimal carbon emission control strategy with the highest score, and sends the strategy instruction to the park's energy control terminal.

9. An electronic device, characterized in that, The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the method for intelligent analysis of carbon emissions based on abnormal carbon factors in smart parks combined with energy storage as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on an electronic device, causes the electronic device to perform the method for intelligent analysis of carbon emissions based on abnormal carbon factors in smart parks combined with energy storage, as described in any one of claims 1 to 7.

Images