Carbon emission evaluation method and system based on data analysis

Abstract

The present application relates to the technical field of data processing, and especially relates to a carbon emission evaluation method and system based on data analysis, which comprises the following steps: collecting wind speed and carbon dioxide emission according to a sampling frequency, and constructing a wind speed time sequence and an emission time sequence; performing coarse-grained processing on the emission time sequence to obtain a plurality of coarse-grained sequences, wherein the value of a coarse-grained point in the coarse-grained sequence is obtained by weighted average calculation of the emission; the calculation of the weight comprises: calculating an abnormal score of an emission data point, the correlation between the emission data point and a corresponding wind speed data point, and a wind speed influence factor; the weight is equal to the ratio of the wind speed influence factor corresponding to the coarse-grained point to the sum of all wind speed influence factors corresponding to the coarse-grained scale; calculating a multi-scale entropy value of the coarse-grained sequence; and evaluating whether the carbon dioxide emission meets a standard or not. The present application adjusts the weight by analyzing the wind speed, reduces the influence of the wind speed on the data during coarse-grained processing, and thus reduces the detection deviation.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a carbon emission assessment method and system based on data analysis. Background Art

[0002] Currently, by monitoring anomalies in factory carbon dioxide emission data, potential illegal emissions, equipment failures or improper operations can be discovered in a timely manner, so that effective intervention and optimization measures can be taken.

[0003] Multiscale entropy algorithms have been applied in various fields. This paper applies multiscale entropy algorithms to anomaly detection in factory CO2 emissions data, leveraging their adaptability and potential high detection accuracy to promptly identify anomalies in emissions data. The multiscale entropy algorithm uses coarse-graining to construct a series of new time series at a predetermined scale.

[0004] However, in practical applications, the collection of factory CO2 emissions data can be affected by environmental factors, resulting in the collected data failing to truly reflect the factory's actual emissions. This distortion can affect the new time series generated by the coarse-grained multiscale entropy algorithm to varying degrees, ultimately leading to inaccurate calculated multiscale entropy values ​​and, in turn, biased anomaly detection results for CO2 emissions data. Summary of the Invention

[0005] In order to solve the above-mentioned technical problem of deviation in abnormal detection results of carbon dioxide emission data due to environmental factors, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a carbon emission assessment method based on data analysis, which includes: collecting wind speed and carbon dioxide emissions according to a sampling frequency, constructing a wind speed time series and an emissions time series; coarse-graining the emissions time series to obtain coarse-grained sequences under multiple preset scales, wherein the values ​​of the coarse-grained points in the coarse-grained sequence are obtained by weighted average calculation of the emissions; calculating the weight includes: calculating the anomaly score of the emission data point in the emissions time series, calculating the correlation between the emission data point and the corresponding wind speed data point in its wind speed time series, the wind speed impact factor being equal to the product of the negative exponential function value of the anomaly score and its correlation; the weight is the ratio of the wind speed impact factor corresponding to the emission data point to the sum of all wind speed impact factors; calculating the sample entropy value of the coarse-grained sequence under each of the preset scales; and calculating the multi-scale entropy value based on the sample entropy value.

[0007] The present invention optimizes the coarse-graining process by combining the wind speed influencing factor of each data point, which can significantly improve the accuracy of multi-scale entropy calculations and thus enhance the reliability of carbon dioxide emission anomaly detection. Specifically, this method dynamically adjusts the weight of each data point by analyzing wind speed data, reducing the contribution of distorted data that is significantly affected by wind speed during the coarse-graining process, making the coarse-graining results at each scale closer to the actual emission level of the factory, thereby improving the reliability of multi-scale entropy. This improvement helps to effectively distinguish between true emission anomalies and pseudo-anomalies caused by wind speed interference, and reduces false positives and missed positives of carbon dioxide emission anomalies due to the influence of ambient wind.

[0008] As a further improvement to the method of the present invention, the anomaly score of the emission data point in the emission time series is ,in, It is The anomaly score of each emission, It is Emissions per emission data point, is the minimum value in the emission time series, is the maximum value in the emission time series, The emission time series The second-order difference of data points, is a numerical stability term, is a natural exponential function.

[0009] The anomaly score, a quantitative indicator, accurately measures the degree of anomaly in CO2 emissions data points. By combining emission value normalization with second-order difference calculations, it closely correlates data trends with the influence of wind speed in the environment, effectively identifying data points that violate the gradual change pattern of CO2 emissions. The higher the score, the greater the likelihood that the data point is anomalous due to wind speed interference; the lower the score, the more reliable the data. This indicator provides an intuitive basis for assessing the reliability of carbon emissions data, helping to accurately screen out anomalous data in environmental monitoring and analysis, ensuring that the data truly reflects emissions, and enhancing the scientific nature of carbon emissions research and management.

[0010] As a further improvement of the method of the present invention, the calculation of the correlation between the emission data point and the wind speed data point corresponding to the wind speed time series includes: constructing a data set of variation of the emission and a data set of variation of the wind speed; the correlation ;in, It is The ranking of the emissions data point in the emissions change dataset, It is The ranking of the wind speed data point corresponding to the emission data point in the wind speed variation data set, is the absolute value, is a numerical stability term.

[0011] By constructing a dataset of emissions changes and wind speed changes and conducting correlation analysis, we can accurately quantify the negative correlation between CO2 emissions and wind speed. Using data point ranking as a link, we can explore potential connections between emissions trends and wind speed fluctuations. The smaller the correlation value, the stronger the negative correlation, which directly reveals the degree of interference from ambient wind on carbon emissions data. This allows us to accurately identify data points that are significantly affected by wind speed and have low credibility. This provides a scientific basis for subsequent calculations of wind speed influencing factors and data reliability assessments, effectively improving the accuracy of carbon emissions data processing and the reliability of environmental analysis.

[0012] As a further improvement of the method of the present invention, the emission variation data set is a data set constructed by calculating the difference between each data point in the emission time series and its previous data point, and sorting the differences from small to large; the wind speed variation data set is a data set constructed by calculating the difference between each data point in the wind speed time series and its previous data point, and sorting the differences from large to small.

[0013] By constructing separate datasets for emissions and wind speed variation and processing the data differences in a specific sorting method, we can clearly visualize the changing characteristics of both. This provides an ordered data foundation for analyzing the correlation between emissions and wind speed, facilitates accurate capture of the impact of wind speed changes on emissions data, helps more accurately identify abnormal emissions data affected by ambient wind interference, and improves the reliability of carbon emissions data analysis.

[0014] As a further improvement to the method of the present invention, the first Second-order differences of data points ;in, The emission time series Emissions per data point, The emission time series Emissions per data point.

[0015] Second-order differences measure the difference between a data point and the data points before and after it. Because CO2 emissions typically change gradually, a large difference between a point and the values ​​immediately preceding and following it indicates an emission anomaly, potentially influenced by wind speed. The magnitude of the second-order difference can be used to assess the reliability of emission data. Small values ​​indicate high reliability and no wind speed influence, while large values ​​indicate low reliability, high anomaly, and a significant wind speed influence.

[0016] As another improvement to the method of the present invention, the correlation between the emission data point and the corresponding wind speed data point in its wind speed time series is calculated using a Pearson correlation coefficient formula.

[0017] Using the Pearson correlation coefficient to calculate the correlation between emissions and wind speed data points can objectively measure the degree of linear correlation between the two, and use standardized numerical values ​​to reflect the strength and direction of the relationship, providing a quantitative basis for analyzing the impact of wind speed on emissions, and helping to accurately determine the extent to which emission data is affected by wind speed interference.

[0018] As another improvement to the method of the present invention, the abnormality score of the emission data point in the emission time series is calculated by a Z-score method.

[0019] The Z-score method is used to calculate anomaly scores, comparing data points with the dataset mean and standard deviation to standardize the degree of anomaly. This method can quickly locate data points that deviate from the normal range and effectively identify outliers in CO2 emissions, laying the foundation for subsequent anomaly detection and analysis.

[0020] As another improvement of the method of the present invention, the calculating of the multi-scale entropy value based on the sample entropy value includes: calculating the average value of the sample entropy values ​​of the coarse-grained sequence under the multiple preset scales; and normalizing the average value to obtain the multi-scale entropy value.

[0021] By calculating the mean entropy of the coarse-grained sequence samples at each preset scale and normalizing it to multi-scale entropy, we can integrate information from different scales and comprehensively characterize the complexity and uncertainty of the emission time series, providing richer information for evaluating the pattern of carbon dioxide emissions from a multi-scale perspective and improving the accuracy of analysis.

[0022] As another improvement of the method of the present invention, the evaluating whether the carbon dioxide emissions meet the standards based on the multi-scale entropy value includes: setting an emission threshold, when the multi-scale entropy value is greater than the emission threshold, marking the carbon dioxide emissions corresponding to the emission time series as abnormal; when the difference between any data point in the emission time series and its previous data point is greater than 0, marking the carbon dioxide emissions corresponding to the emission time series as abnormally increased.

[0023] In the second aspect, the present invention provides a carbon emission assessment system based on data analysis, which includes a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the carbon emission assessment method based on data analysis of the first aspect of the present invention is implemented.

[0024] By adopting the above technical solution, the carbon emission assessment method based on data analysis of the first aspect of the present invention is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0025] The present invention utilizes wind speed influencing factors to optimize the coarse-graining process, improving the accuracy of multiscale entropy calculations and enhancing the reliability of carbon dioxide emission anomaly detection. By analyzing wind speed and dynamically adjusting data point weights, the impact of distorted data due to wind speed interference is reduced, making the coarse-graining results more consistent with actual emissions. This method effectively distinguishes true anomalies from false wind speed anomalies, reducing false positives and missed detections due to ambient wind conditions, and providing more accurate technical support for carbon emission monitoring and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flow chart of a carbon emission assessment method based on data analysis provided in an embodiment of the present invention;

[0027] Figure 2 A flowchart of the coarse-grained processing process provided by an embodiment of the present invention;

[0028] Figure 3 This is a structural block diagram of a carbon emission assessment system based on data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The first aspect of the embodiment of the present invention provides a carbon emission assessment method based on data analysis, such as Figure 1 As shown, the method includes steps S100 to S500:

[0030] Step S100: Collect wind speed and carbon dioxide emissions according to a sampling frequency, and construct a wind speed time series and an emission time series.

[0031] To elaborate, the equipment for collecting carbon dioxide emissions can use an infrared gas analyzer or a non-dispersive infrared sensor, and the equipment for collecting wind speed can use an anemometer.

[0032] CO2 emissions and wind speed data should be collected at the same location, meaning the collection equipment needs to be set up in the same place. This ensures that the CO2 emissions and wind speed data collected at the same time are for the same spatial area, reducing interference caused by spatial differences and providing a basis for analyzing their correlation.

[0033] It should be noted that the sampling frequency of carbon dioxide emissions and wind speed also needs to be consistent.

[0034] This ensures that the two data sets have the same collection duration and sampling time. Time-synchronized data can provide a foundation for subsequent analysis.

[0035] For example, set the collection duration for both to 24 hours and the collection frequency to once every 5 seconds.

[0036] Regarding the construction of time series, a time series refers to a series of data points arranged in chronological order. As mentioned earlier, wind speed data and carbon dioxide emission data at multiple time points are collected according to the sampling frequency. The carbon dioxide emission time series and wind speed time series of the present invention are both collected in chronological order according to the sampling time.

[0037] According to the above method, two time series of equal length can be obtained. Assume that the emission time series is: ,in, The emission time series The emissions of data points. Assume that the wind speed time series is: ,in, The wind speed time series The wind speed of each data point.

[0038] Step S200: performing coarse-graining processing on the emission time series to obtain coarse-grained sequences at multiple preset scales.

[0039] To elaborate, the multi-scale entropy algorithm is a method for analyzing the complexity of time series. It evaluates the complexity of the signal by calculating the sample entropy at multiple time scales. Its calculation steps include coarse-graining processing and sample entropy calculation. The present invention improves the coarse-graining processing, such as Figure 2 The process shown includes steps S210 to S240:

[0040] Step S210: Calculate the anomaly score of the emission data point in the emission time series.

[0041] To expand on this, the anomaly score is a numerical value that quantifies the extent to which each data point in a time series is an outlier. The higher the anomaly score, the more likely the data point is an outlier.

[0042] In the present invention, the influence of wind speed on carbon dioxide emissions is mainly considered among environmental factors. If there is an influence of wind speed, the corresponding carbon dioxide emissions will be more abnormal. Therefore, the reliability of the collected carbon dioxide emissions can be judged by calculating the anomaly score.

[0043] The anomaly score calculation formula for an emissions data point is:

[0044] ;

[0045] in, It is Anomaly score for each emission data point, It is Emissions per emission data point, is the minimum value in the emission time series, is the maximum value in the emission time series, The emission time series The second-order difference of data points, is a numerical stability term, is a natural exponential function.

[0046] in the formula Indicates the The relative size of the value of each emission data point. The smaller the value, the The greater the abnormal low value of the data point, the The more prominent the value of an emission data point, the more likely it is that its sampling is not authentic and the greater its degree of specificity.

[0047] In the formula, the first Second-order differences of data points:

[0048] ;

[0049] in, The emission time series Emissions per data point, The emission time series Emissions per data point, The emission time series Emissions per data point.

[0050] The second-order difference calculates the The second-order difference is the difference between the values ​​of a data point and the two data points before and after it. Under normal circumstances, carbon dioxide emissions data does not decrease instantly, but changes gradually over a period of time. Therefore, if the difference in the value change of a data point and the data points on its left and right is large, it means that the carbon dioxide emissions of this data point are abnormal and are likely affected by environmental factors. Therefore, the size of the second-order difference can be used to evaluate whether the carbon dioxide emissions of this data point are reliable. The smaller the value, the more reliable the carbon dioxide emissions of this data point are, and it is not affected by wind speed. The larger the value, the less reliable the carbon dioxide emissions of this data point are, the higher the degree of abnormality, and the greater the degree of influence of wind speed.

[0051] In addition, it should be noted that This is a numerical stability term. This term is used to avoid the situation where the second-order difference is 0. When the second-order difference is 0, the calculated anomaly score will be 0. Therefore, this value can be between 0 and 0.2. However, it should be noted that this numerical stability term must be the same for each emission data point. In this invention, it is set to 0.001.

[0052] In summary, the anomaly score can assess the degree of anomaly of data affected by wind speed in the emission time series.

[0053] Optionally, the anomaly score may also be calculated using a Z-score method or an interquartile range method.

[0054] Step S220: Calculate the wind speed impact factor corresponding to the emission data point.

[0055] To elaborate, through the analysis of the above steps, an anomaly score was obtained for each emission data point. However, data points with low CO2 emission values ​​that actually exist may also have large anomaly scores, so the impact of data points with low CO2 emission values ​​that actually exist needs to be eliminated in this step. After scenario research, the change in the CO2 emission data points with low values ​​due to ambient wind compared to the previous moment showed a strong negative correlation with the change in wind speed data at the same moment compared to the previous moment. Therefore, the impact of wind speed on CO2 emissions can be calculated and evaluated by calculating the correlation.

[0056] Specifically, the calculation formula of the wind speed influence factor is:

[0057] ;

[0058] in, It is The wind speed impact factor corresponding to each emission data point is: is the normalization function, It is Anomaly score for each emission data point, It is The correlation between each emission data point and its corresponding wind speed data point in the wind speed time series, is a natural exponential function.

[0059] in the formula The larger the When analyzing the CO2 emission data point based on the CO2 emission data itself, the greater the possibility of its sampling being untrue, the greater the corresponding wind speed impact factor will be. The correlation between the carbon dioxide emission data and the wind speed data at the sampling moment is smaller, the stronger the negative correlation is, indicating that the The greater the degree to which the carbon dioxide emission data point at the sampling moment is affected by the ambient wind, the The greater the possibility that the sampling of carbon dioxide emission data corresponding to a moment is untrue, the greater the wind speed impact factor will be.

[0060] This formula can be used to determine whether the CO2 emissions at that data point are abnormally low due to wind speed or are truly abnormally low. By considering wind speed as an external factor, we can reduce the chances of misidentifying normal data affected by the environment as abnormal, thereby improving the accuracy of anomaly detection.

[0061] There are two ways to calculate the correlation in this formula, which are introduced one by one below:

[0062] The first is an implementation method: the correlation is calculated by analyzing the data of two time series.

[0063] Specifically, the correlation calculation involves first constructing two data sets: one for emissions based on the emissions time series, and one for wind speed based on the wind speed time series. The emissions data set is constructed by calculating the difference between each data point in the emissions time series and its previous data point, and sorting the differences from smallest to largest. The wind speed data set is constructed by calculating the difference between each data point in the wind speed time series and its previous data point, and sorting the differences from largest to smallest.

[0064] Then calculate the correlation:

[0065] ;

[0066] in, It is The ranking of the emission data point in the emission change dataset, It is The ranking of the wind speed data point corresponding to each emission data point in the wind speed variation data set, is the absolute value, is a numerical stability term.

[0067] In this formula, the correlation Analyzed the There is a negative correlation between the emission data point and the corresponding wind speed data. The smaller the value, the stronger the negative correlation, indicating that the The greater the degree to which the carbon dioxide emissions of an emission data point are affected by the ambient wind, the greater the possibility that the carbon dioxide emissions corresponding to the data point are untrue, and the greater the wind speed impact factor calculated from it.

[0068] In addition, it should be noted that is a numerical stability term, which is used to avoid When is 0, When the value is 0, the calculated wind speed impact factor will be 0, so the value is set as the numerical stabilizer. The value can be between 0 and 0.2, but it is noted that the numerical stabilizer should be the same for each emission data point, and the present application sets it to 0.1.

[0069] The second is an implementation: calculated by the Pearson correlation coefficient formula.

[0070] Specifically, the correlation is:

[0071] ;

[0072] wherein, is the total number of data points in the time series, is the emission of the th data point in the emission time series, is the average value of the emissions of all data points in the emission time series, is the wind speed of the th data point in the wind speed time series, is the average value of all wind speed data points in the wind speed time series.

[0073] indicates a negative correlation, the closer to -1, the stronger the negative correlation. It is explained that the th emission data point is more affected by the environmental wind, the greater the possibility that the carbon dioxide emission corresponding to the data point is not real, and the greater the wind speed impact factor calculated therefrom.

[0074] Step S230, calculate the weight corresponding to the emission data point.

[0075] In detail, the present application sets a weight when calculating the coarse-grained point in the coarse-grained sequence, which is the ratio of the wind speed impact factor corresponding to the emission data point to the sum of all wind speed impact factors, that is:

[0076] ;

[0077] wherein, is the weight corresponding to the th emission data point, is the wind speed impact factor corresponding to the th emission data point, is the total number of data points in the time series, indicates the sum of all wind speed impact factors in the wind speed time series, is a natural exponential function.

[0078] As mentioned earlier, the greater the wind speed influence factor of an emission data point, the greater the degree to which the sampling value of this data point is affected by the ambient wind, and the lower the credibility of the carbon dioxide emissions obtained from the collection of this emission data point.

[0079] In this formula, weights are determined by the wind speed influence factor and used to weight the coarse-grained points in the coarse-grained sequence, effectively preventing environmental wind interference from misleading carbon emission data. When an emission data point is significantly affected by wind speed, its calculated weight is smaller, making the coarse-grained results closer to the actual emission trend and preventing abnormal fluctuations from interfering with the multi-scale entropy analysis. Conversely, data points less affected by wind speed are assigned greater weights, ensuring that valid data is fully utilized. This mechanism enhances the anti-interference ability of data processing and improves the accuracy and reliability of multi-scale entropy analysis, providing a more scientific decision-making basis for carbon emission monitoring and environmental assessment.

[0080] Step S240: Set the scale factor and construct a coarse-grained sequence.

[0081] To expand on this, in the calculation of the multi-scale entropy algorithm, it is necessary to set the scale factor and then construct the coarse-grained sequence of the original time series at each scale. In the present invention, the scale factor can be set to , is the maximum scale factor.

[0082] The coarse-graining process requires constructing a coarse-grained sequence at each preset scale for the emission time series. The calculation formula for the coarse-grained points in the coarse-grained sequence is:

[0083] ;

[0084] in, Representation scale In the coarse-grained sequence The data corresponding to the coarse-grained points, It is The weight corresponding to each emission data point, The emission time series Emissions per data point, is the index value of the data point in the emissions time series, is the scale factor, is the index value of the current data point in the emissions time series Based on , the index value after the scale factor is offset.

[0085] This formula calculates weights based on the wind speed influence factor and applies them to the coarse-grained sequence construction, effectively eliminating the interference of ambient wind on carbon emission data. Data points significantly affected by wind speed are assigned smaller weights, reducing the impact of abnormal fluctuations and making the coarse-grained results more consistent with actual emission trends. Data points less affected by wind speed are assigned larger weights, fully leveraging the available data. Ultimately, this improves the accuracy and reliability of multiscale entropy analysis, providing a more scientific and precise basis for decision-making in carbon emission monitoring and environmental assessment.

[0086] In addition, it should be noted that if the remaining data points at a certain scale cannot form a complete set, the last few data can be discarded.

[0087] Step S300: Calculate the sample entropy value of the coarse-grained sequence at each preset scale.

[0088] To elaborate, the existing technology for calculating sample entropy values ​​will not be introduced here. It is only explained that the embedding dimension when calculating sample entropy at each scale can be preset to an empirical value of 2, and the tolerance can be preset to an empirical value of 0.08. Of course, in actual use, it can be set according to needs and empirical values.

[0089] Step S400: Calculate multi-scale entropy values ​​based on sample entropy values.

[0090] To expand on this, we average the sample entropy of the coarse-grained sequence at all scales, and then calculate the entropy by After the normalization function normalizes the result, the normalized multi-scale entropy value can be obtained.

[0091] Step S500: Evaluate whether the carbon dioxide emissions meet the standards based on the multi-scale entropy value.

[0092] To elaborate, in order to detect whether there are any abnormalities in carbon dioxide emissions, especially when there are too many abnormalities, it is necessary to set the emission threshold experience value to 0.9. Of course, it can also be set according to needs and experience.

[0093] Specifically, an emission threshold is set. When the multi-scale entropy value is greater than the emission threshold, the abnormal fluctuation of carbon dioxide emissions corresponding to the emission time series is marked.

[0094] On this basis, it is also necessary to determine whether the abnormal fluctuation is an abnormally decreasing fluctuation or an abnormally increasing fluctuation. This can be determined by judging whether the difference between any data point in the emission time series and its previous data point is greater than 0. When the difference is greater than 0, it indicates that the carbon dioxide emissions have increased abnormally.

[0095] Normally, abnormal enlargement requires attention and the staff should be notified to make timely adjustments.

[0096] The second aspect of this embodiment provides a carbon emission assessment system based on data analysis, such as Figure 3 As shown, the carbon emission assessment system based on data analysis includes a memory and a processor. The memory stores computer program instructions. When the computer program instructions are executed by the processor, the carbon emission assessment method based on data analysis of the first aspect of the present invention is implemented.

[0097] The carbon emission assessment system based on data analysis also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0098] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory, dynamic random access memory, static random access memory, enhanced dynamic random access memory, high bandwidth memory, hybrid memory cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium may be part of, accessible to, or connectable to the device.

[0099] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A carbon emission assessment method based on data analysis, characterized in that: include: Collect wind speed and carbon dioxide emissions according to the sampling frequency, and construct wind speed time series and emission time series; The emission time series is coarse-grained to obtain coarse-grained sequences at multiple preset scales. The values ​​of the coarse-grained points in the coarse-grained sequences are calculated by weighted average of the emissions. The weight calculation includes: calculating the anomaly score of the emission data point in the emission time series, calculating the correlation between the emission data point and the corresponding wind speed data point in its wind speed time series, and the wind speed impact factor is equal to the product of the negative exponential function value of the anomaly score and its correlation. The weight is the ratio of the wind speed impact factor corresponding to the emission data point to the sum of all wind speed impact factors. Anomaly score for an emissions data point in the emissions time series , It is The anomaly score of each emission, It is Emissions per emission data point, 、 are the minimum and maximum values ​​in the emission time series, The emission time series The second-order difference of data points, is a numerical stability term, is a natural exponential function; Calculate the correlation between an emission data point and the corresponding wind speed data point in its wind speed time series, including: Construct a dataset of emission variation and a dataset of wind speed variation; Correlation ; It is The ranking of the emission data point in the emission change dataset, It is The ranking of the wind speed data point corresponding to the emission data point in the wind speed variation data set, is the absolute value, is a numerical stability term Calculate the sample entropy value of the coarse-grained sequence at each preset scale; Calculate multi-scale entropy values ​​based on sample entropy values; Evaluate whether carbon dioxide emissions meet the standards based on multi-scale entropy values.

2. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: The emission change dataset is constructed by calculating the difference between each data point in the emission time series and its previous data point, and sorting the differences from small to large; The wind speed variation data set is constructed by calculating the difference between each data point in the wind speed time series and its previous data point, and sorting the differences from large to small.

3. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: The emission time series Second-order differences of data points ; in, The emission time series Emissions per data point, The emission time series Emissions per data point.

4. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: The correlation between the emission data point and the corresponding wind speed data point in its wind speed time series is calculated using the Pearson correlation coefficient formula.

5. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: The anomaly score of the emission data point in the emission time series is calculated using the Z-score method.

6. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: Calculating a multi-scale entropy value based on the sample entropy value includes: Calculating an average of the sample entropy values ​​of the coarse-grained sequences at the plurality of preset scales; The average value is normalized to obtain a multi-scale entropy value.

7. The carbon emission assessment method based on data analysis according to claim 1, characterized in that: Evaluating whether the carbon dioxide emissions meet the standards according to the multi-scale entropy value includes: Setting an emission threshold, and when the multi-scale entropy value is greater than the emission threshold, marking the carbon dioxide emissions corresponding to the emission time series as abnormal; When the difference between any data point in the emission time series and its previous data point is greater than 0, it is marked that the carbon dioxide emissions corresponding to the emission time series have increased abnormally.

8. A carbon emission assessment system based on data analysis, characterized in that: The carbon emission assessment system based on data analysis includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the carbon emission assessment method based on data analysis according to any one of claims 1 to 7 is implemented.

Images