Carbon emission metering system based on cloud computing

Abstract

The invention relates to the technical field of carbon emission, and discloses a cloud computing-based carbon emission metering system, which comprises a data acquisition and intelligent preprocessing module, a cloud computing module, an integration module and a real-time monitoring early warning module, and is characterized in that the data acquisition and intelligent preprocessing module is used for acquiring satellite remote sensing monitoring data; the urban air quality detection data and the carbon emission data of different enterprises are acquired, the multiple kinds of data are integrated to obtain a multi-dimensional data set, accurate fusion of multi-source data is achieved through the data acquisition and intelligent preprocessing module, multi-dimensional deep analysis is achieved through the integration module, and the real-time monitoring and early warning module constructs an intelligent monitoring and early warning system. Under a multi-source heterogeneous carbon emission data environment, accurate data fusion and multi-dimensional deep analysis are realized, and an intelligent carbon emission monitoring and early warning system suitable for a complex and changeable environment and enterprise development conditions is constructed, so that the requirements of accurate control of carbon emission of different enterprises and guidance of green development of the industry are met.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission, and more specifically to a carbon emission measurement system based on cloud computing. Background Art

[0002] With the acceleration of global industrialization, the impact of human activities on the environment has become increasingly significant, and carbon emissions have become a global focus. This urgently requires accurate and efficient carbon emission measurement technology to monitor and evaluate the carbon emissions of various countries, regions and enterprises, so as to formulate reasonable emission reduction strategies and policies and achieve effective control of carbon emissions.

[0003] Traditional carbon emission measurement often relies on data from a single source, such as energy consumption data reported by enterprises themselves or local environmental monitoring data. These data sources are limited. Therefore, how to achieve accurate data integration, multi-dimensional in-depth analysis, and build an intelligent carbon emission monitoring and early warning system that adapts to complex and changing environments and corporate development conditions in a multi-source heterogeneous carbon emission data environment to meet the needs of accurate carbon emission control of different enterprises and guide the green development of the industry has become an urgent problem to be solved. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a carbon emission measurement system based on cloud computing to solve the problems existing in the above-mentioned background technology.

[0005] The present invention provides the following technical solutions:

[0006] A carbon emission measurement system based on cloud computing, including a data acquisition and intelligent preprocessing module, a cloud computing module, an integration module and a real-time monitoring and early warning module;

[0007] Data acquisition and intelligent preprocessing module, used to obtain satellite remote sensing monitoring data, urban air quality detection data and carbon emission data of different enterprises, and integrate multiple data to obtain multi-dimensional data sets;

[0008] The cloud computing module is used to build carbon emission models for different industries, match the most appropriate carbon emission model according to the industry characteristics of different enterprises in the multi-dimensional data set, and use distributed computing technology to measure the carbon emissions of different enterprises;

[0009] The integration module is used to classify and summarize the carbon emissions of different companies through multi-dimensional data sets, and identify key companies by comparing and analyzing the carbon emissions of different companies and industries in the same time period;

[0010] The real-time monitoring and early warning module is used to monitor the carbon emissions of enterprises in real time. By automatically adjusting the threshold setting strategy, it will issue early warning information when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold.

[0011] Preferably, in the data acquisition and intelligent preprocessing module, multiple data are fused to obtain a multi-dimensional data set, and the fusion process is as follows:

[0012] Set the satellite remote sensing monitoring data set as D s , each piece of data t s Indicates the satellite remote sensing monitoring data timestamp, loc s Indicates the geographic location of satellite remote sensing monitoring data, Indicates the msth satellite remote sensing monitoring index value;

[0013] Set the urban air quality detection data set as D a , each piece of data t a Indicates the timestamp of urban air quality detection data, loc a Indicates the geographical location of the city's air quality detection data. Indicates the value of the ma-th air quality detection index;

[0014] Assume that the carbon emission data set of different enterprises is D c , each carbon emission data t c Indicates the carbon emission data timestamp of the enterprise, loc c Indicates the geographical location of the company’s carbon emissions data, Represents the value of the ma-th carbon emission index;

[0015] Match different data by time, the time matching function Mat(t i , t p )as follows:

[0016] Set the time tolerance threshold to Δt;

[0017] If |t i -t p |≤Δt, then Mat(t i , t p )=1, indicating that the time matching is successful;

[0018] If |t i -t p |>Δt, then Mat(t i , t p )=0, indicating that the time matching is unsuccessful;

[0019] In the formula, ti and t p Represents t s ,t a and t c , when matching, you can s ,t a and t c Cross-substitute into t i and t p Time matching of multiple data is performed in

[0020] Match different data through space, spatial matching function Mat(loc i ,loc p )as follows:

[0021] Set the geographical distance tolerance threshold to Δd;

[0022] if Then Mat(loc i ,loc p )=1, indicating successful spatial matching;

[0023] if Then Mat(loc i ,loc p )=0, indicating that the spatial matching is unsuccessful;

[0024] In the formula, loc i =(lon i ,lat i ),loc p =(lon p ,lat p ), indicating the geographical location of different data, loc i and loc p Represents loc s ,loc a and loc c , indicating the geographical location of different data, loc can be used for matching s ,loc a and loc c Cross-substitute into loc i and loc p Perform spatial matching of multiple data in

[0025] Set the multidimensional dataset to d fused =(D s , D a , D c );

[0026] Accurately identify and correct abnormal data points through intelligent algorithms.

[0027] Preferably, the abnormal data points can be accurately identified and corrected through intelligent algorithms. The correction process is as follows:

[0028] Set d in multi-dimensional data fused The data sequence in is X = {x1, x2, ..., x n}, x n represents the nth data sequence, and n represents the number of data in the data sequence X;

[0029] Calculate the mean of the data series using the mean calculation formula and standard deviation calculation formula And standard deviation σ:

[0030]

[0031] In the formula, x f represents the fth data value in the data sequence;

[0032] Set the volatility multiple threshold to k;

[0033] if Then determine x f is an abnormal data point, marked as x e ;

[0034] if Then determine x f is a normal data point;

[0035] For abnormal data points x e Make corrections. The correction process is as follows:

[0036] Select the abnormal data point x e The neighboring q data are {x e-l , x e-l+1 , …, x e-1 , x e+1 , x e+r}, where l and r are natural numbers greater than 0, l+r=q, and q is an odd number;

[0037]

[0038] In the formula, represents the corrected data point, and iz represents the index variable.

[0039] Preferably, in the cloud computing module, carbon emission models for different industries are constructed, and the construction process is as follows:

[0040] Set the ibth enterprise industry feature vector in the multidimensional data set to XH ib =(xh1, xh2, ..., xh b), where b represents the total number of enterprise industry characteristic vectors, ib = 1, 2, ..., b;

[0041] The enterprise industry feature vector is standardized, and the processing process is as follows:

[0042] The mean of the enterprise industry characteristics calculated by the mean calculation formula and standard deviation calculation formula and standard deviation Calculate the standardized enterprise industry characteristic vector Enterprise industry characteristic vector The calculation formula is as follows:

[0043]

[0044] Construct a model library containing a variety of carbon emission models for different industries, and set the carbon emission model MH = {MH1, MH2, ..., MH mf}, mf represents the total number of carbon emission models, and each carbon emission model MH kf All correspond to specific industry types, kf = {1, 2, ..., mf};

[0045] For the ibth enterprise, calculate the standardized enterprise industry characteristic vector With each carbon emission model MH kf The similarity between the ideal industry characteristics of the adaptation is calculated as follows:

[0046]

[0047] In the formula, xh ib，j represents the original value of the ib-th enterprise on the j-th feature dimension, MH kf，j Represents the carbon emission model MH on the jth feature dimension kf , j represents the index variable, Indicates that the ibth enterprise and MH kf The similarity between them ranges from [-1, 1];

[0048] According to the calculated similarity, the most suitable carbon emission model MH is matched for the ibth enterprise. kf , calculate the model index j that maximizes the similarity max , the calculation formula is as follows:

[0049]

[0050] In the formula, It means finding the parameters that make the function maximum;

[0051] Randomly select dc data points as the initial cluster centers, marked as μ ju, where ju = 1, 2, ..., dc;

[0052] For the standardized enterprise industry feature vector in the multidimensional data set Calculate the μ for each cluster center ju The distance is calculated as follows:

[0053]

[0054] In the formula, express With each cluster center μ ju distance;

[0055] For each cluster, recalculate its cluster center μ ju (ca+1), calculated as follows:

[0056]

[0057] In the formula, ca represents the number of iterations, μ ju (ca) represents the set of data points belonging to the juth cluster at the cath iteration;

[0058] Set the cluster center μ ju (ca+1) The maximum number of iterations is Ds max ;

[0059] Constantly ju (ca+1) is updated until the maximum number of iterations Ds is reached max , the iteration stops.

[0060] Preferably, regarding the integration module, the carbon emissions of different enterprises are classified and summarized through multi-dimensional data sets, and the classification and summary process is as follows:

[0061] Set the enterprise data set in the multidimensional data set to EQ = {eq1, eq2, ..., eq nq}, nq represents the total number of enterprises, each eq ib The industry category set IH, time period set IT, geographical region set IR, enterprise size set IS and carbon emission data set D of different enterprises represent the ib-th enterprise c ;

[0062] At this time, the classification function is set to CL = (IH, IT, IR, IS, D c );

[0063] Setting BK T It is a B-tree index built based on timestamps. The timestamp range is [t start , t end ];

[0064]

[0065] In the formula, QK T (t start , t end , B.K. T ) represents the B-tree query function, t start Indicates the starting timestamp, t end Indicates the end timestamp. Indicates the timestamp corresponding to the enterprise data;

[0066] Setting HM I For hash index based on industry category, hash query function QM I (ir,HM I ) can be expressed as:

[0067] QM I (ir,HM I )={eq ib |eq ib ∈EQ,ir};

[0068] In the formula, ir represents the industry category to which the enterprise belongs;

[0069] Calculate the carbon emission data of different companies through the mean calculation formula c The average of corporate carbon emissions is obtained

[0070] Calculate the carbon emission data d of the ibth enterprise c The variance is calculated as follows:

[0071]

[0072] In the formula, CDF 2 represents the variance of corporate carbon emissions;

[0073] Set the enterprise identity set to IFQ = {ifq1, ifq2, ..., ifq nq};

[0074] When drawing a bar chart, use the company logo ifq nq As the horizontal axis parameter, the carbon emission data d c as the vertical axis;

[0075] To identify key enterprises, the identification formula is as follows:

[0076]

[0077] In the formula, IZB(ib) represents the key enterprise judgment index;

[0078] The judgment threshold for key enterprises is set as Φ;

[0079] If IZB(ib)≥Φ, the enterprise is marked as a key enterprise IZB(ib) ZDY ;

[0080] If IZB(ib)<Φ, the enterprise is marked as a normal enterprise IZB(ib) PTY .

[0081] Preferably, in the real-time monitoring and early warning module, by automatically adjusting the threshold setting strategy, when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold, an early warning message is issued. The early warning process is as follows:

[0082] The carbon emissions of ordinary enterprises and key enterprises are set as TPL PTY and TPL ZDY , setting the carbon emission thresholds for ordinary enterprises and key enterprises to be FPt PTY and FPt ZDY ;

[0083] If TPL PTY ≥FPt PTY , then determine the general enterprise IZB(ib) PTY If carbon emissions exceed the limit, the first warning instruction will be issued;

[0084] If TPL PTY <FPt PTY , then determine the general enterprise IZB(ib) PTY Carbon emissions are normal and no instructions are issued;

[0085] If TPL ZDY ≥FPtZ DY , then determine the key enterprise IZB (ib) ZDY Carbon emissions exceed the standard and a second warning order is issued;

[0086] If TPL ZDY <FPtZ DY , then determine the key enterprise IZB (ib) ZDY Carbon emissions are normal and no instructions are issued;

[0087] After receiving the first warning instruction, the alarm flashes to give a warning;

[0088] After receiving the second warning command, the alarm will light up constantly to give a warning;

[0089] Adjust the threshold FPt based on the upgrade status of the industry to which the enterprise belongs PTY and FPt ZDY .

[0090] Preferably, the threshold FPt is adjusted based on the upgrade status of the industry to which the enterprise belongs PTY and FPt ZDY , the adjustment process is as follows:

[0091] The industrial upgrading status coefficient of the industries to which ordinary enterprises and key enterprises belong is set as SSJB PTY and SSJB ZDY , the value range is 0-1, 0 means that the enterprise is not in the stage of industrial upgrading, and 1 means that the enterprise is in the stage of deep industrial upgrading;

[0092] F P PTYt =FPt PTY ×(1+ksuq1×SSJB PTY );

[0093] Where, FPt PTYt represents the carbon emission threshold of ordinary enterprises after adjustment, and ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of ordinary enterprises;

[0094] F P ZDYt =FPt ZDY ×(1+ksuq2×SSJB ZDY );

[0095] Where, FPt ZDYt represents the carbon emission threshold of key enterprises after adjustment, ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of key enterprises;

[0096] Introducing external environmental factors to FPt PTYt and FPt ZDYt Continue to adjust.

[0097] Preferably, the introduction of external environmental factors will affect FPt PTYt and FPt ZDYt Continue to adjust, the adjustment process is as follows:

[0098] The external environment impact coefficient is set to JXTQ, with a value range of 0.8 to 1.2;

[0099] F P sj =FPt PTYt ×JXTQ;

[0100] Where, FPt sj It represents the carbon emission threshold of an average enterprise after adjusting for external environmental factors;

[0101] F P sjz =FPt ZDYt ×JXTQ;

[0102] Where, FPtsjz It indicates the carbon emission threshold of key enterprises after adjustment by introducing external environmental factors.

[0103] Technical effects and advantages of the present invention:

[0104] In the data collection and intelligent preprocessing stage, a multi-dimensional data set is constructed by integrating satellite remote sensing monitoring data, urban air quality detection data and carbon emission data of different enterprises. With the help of intelligent algorithms based on data fluctuation range and industry emission rules, abnormal data points in the data set are accurately identified and corrected. This high-precision data preprocessing not only effectively eliminates the interference of data noise and erroneous information, but also fully explores the inherent correlation between multi-source data, providing a solid and reliable data foundation for subsequent analysis and decision-making. In the cloud computing module, the clustering analysis algorithm of machine learning is used to deeply analyze the industry characteristic data of enterprises in the multi-dimensional data set, such as production process, main product type, energy use structure, etc., and on this basis, the most suitable carbon emission measurement model is matched for each enterprise, ensuring the pertinence and accuracy of the model.

[0105] Based on the multi-dimensional data set, the integration module classifies and summarizes the carbon emissions of different enterprises, breaking through the limitations of traditional classification based only on industry categories and time periods, and introducing multiple dimensions such as geographical regions and enterprise scale for cross-classification. In the geographical area classification, it can not only be macro-divided according to conventional urban areas and counties, but also further refined to smaller geographical units such as industrial parks and commercial areas, thereby realizing a detailed analysis of the micro and macro distribution of carbon emissions. In order to improve the speed of data query and retrieval, two efficient index structures are constructed for data classified in different dimensions. Among them, the B-tree index based on timestamp can quickly locate carbon emission data within a specific time period in chronological order in massive data, meeting the needs of time series analysis of carbon emission data. The hash index based on industry categories can instantly locate the carbon emission data of all enterprises in a certain industry, which provides great convenience for comparative research between industries. On this basis, the statistical analysis method is adopted The method calculates the statistical indicators of the mean and variance of carbon emissions of different enterprises and industries in the same time period in the multidimensional data set. Through these indicators, not only can the differences in carbon emission levels of various enterprises and industries be comprehensively and systematically evaluated, but also the advanced statistical methods such as variance analysis can be used to accurately judge the degree of dispersion of carbon emission data between different industries, and then accurately identify industries and enterprises with large fluctuations in carbon emissions, providing clear focus objects and research directions for environmental management departments and enterprises themselves, and effectively promoting the transformation of carbon emission control strategies from extensive to refined. In addition, the constructed data visualization comparison tool presents the carbon emission data of different enterprises and industries in the multidimensional data set in the form of intuitive bar charts. This visualization method converts abstract data into intuitive graphical information, which is convenient for decision makers to quickly capture the key enterprises and industry trends of carbon emissions, and provides intuitive and powerful support for the formulation of scientific and reasonable emission reduction policies and corporate development strategies, effectively improving the efficiency and scientificity of decision-making.

[0106] Through the real-time monitoring and early warning module, we can take into account the short-term carbon emission fluctuations that industries in the stage of industrial upgrading may face during the process of technological transformation, and appropriately relax their short-term carbon emission thresholds through scientific and reasonable threshold setting strategies. This flexible threshold adjustment mechanism provides enterprises with the necessary space for innovation and development, encourages enterprises to actively invest resources in technological transformation and upgrading, and helps promote the green transformation and sustainable development of the entire industry. On the contrary, for enterprises with high pollution, high emissions and poor rectification, the early warning module adopts a strategy of gradually tightening their carbon emission thresholds, and by increasing environmental cost pressure, it prompts enterprises to accelerate the pace of emission reduction, effectively curbing the adverse environmental behavior of such enterprises. At the same time, this module fully considers the impact of external environmental factors on corporate carbon emissions and introduces external environmental factors. The threshold is adjusted in real time based on environmental factors. For example, during the winter heating period, the energy consumption increases due to the sharp increase in heating demand. The carbon emission thresholds of some heating companies will be appropriately raised according to this actual situation to avoid false warnings caused by normal production needs. During large-scale international events, in order to ensure the air quality and environmental image during the events, the system will lower the carbon emission threshold in the entire city and strengthen emission control for all companies. This dynamic threshold adjustment mechanism based on changes in the external environment makes the carbon emission early warning system more in line with actual production and living scenarios, realizes the organic coordination of environmental control and social and economic activities, improves the scientificity, rationality and effectiveness of the entire carbon emission monitoring and early warning system, and provides strong technical support for building a green and low-carbon society. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] Figure 1 This is a block diagram of the carbon emission measurement system based on cloud computing of the present invention. DETAILED DESCRIPTION

[0108] The present invention is further described below in conjunction with specific embodiments. However, people familiar with the art should understand that the detailed description given here in conjunction with the drawings is for better explanation, and the structure of the present invention necessarily exceeds these limited embodiments. For some equivalent replacement schemes or common means, they are no longer described in detail herein, but still belong to the scope of protection of the present application.

[0109] Figure 1 The best embodiment of the present invention is shown below in conjunction with the attached Figure 1 The present invention is further described.

[0110] A carbon emission measurement system based on cloud computing, including a data acquisition and intelligent preprocessing module, a cloud computing module, an integration module and a real-time monitoring and early warning module;

[0111] Data acquisition and intelligent preprocessing module, used to obtain satellite remote sensing monitoring data, urban air quality detection data and carbon emission data of different enterprises, and integrate multiple data to obtain multi-dimensional data sets;

[0112] The cloud computing module is used to build carbon emission models for different industries, match the most appropriate carbon emission model according to the industry characteristics of different enterprises in the multi-dimensional data set, and use distributed computing technology to measure the carbon emissions of different enterprises;

[0113] The integration module is used to classify and summarize the carbon emissions of different companies through multi-dimensional data sets, and identify key companies by comparing and analyzing the carbon emissions of different companies and industries in the same time period;

[0114] The real-time monitoring and early warning module is used to monitor the carbon emissions of enterprises in real time. By automatically adjusting the threshold setting strategy, it will issue early warning information when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold.

[0115] As a further feature of the present invention, in the data acquisition and intelligent preprocessing module, multiple data are fused to obtain a multi-dimensional data set, and the fusion process is as follows:

[0116] Set the satellite remote sensing monitoring data set as D s , each piece of data t s Indicates the satellite remote sensing monitoring data timestamp, loc s Indicates the geographic location of satellite remote sensing monitoring data, Indicates the msth satellite remote sensing monitoring index value;

[0117] Set the urban air quality detection data set as D a , each piece of data t a Indicates the timestamp of urban air quality detection data, loc a Indicates the geographical location of the city's air quality detection data. Indicates the value of the ma-th air quality detection index;

[0118] Assume that the carbon emission data set of different enterprises is D c , each carbon emission data t c Indicates the carbon emission data timestamp of the enterprise, loc c Indicates the geographical location of the company’s carbon emissions data, Represents the value of the ma-th carbon emission index;

[0119] Match different data by time, the time matching function Mat(t i , t p)as follows:

[0120] Set the time tolerance threshold to Δt;

[0121] If |t i -t p |≤Δt, then Mat(t i , t p )=1, indicating that the time matching is successful;

[0122] If |t i -t p |>Δt, then Mat(t i , t p )=0, indicating that the time matching is unsuccessful;

[0123] In the formula, t i and t p Represents t s ,t a and t c , when matching, you can s ,t a and t c Cross-substitute into t i and t p Time matching of multiple data is performed in

[0124] Match different data through space, spatial matching function Mat(loc i ,loc p )as follows:

[0125] Set the geographical distance tolerance threshold to Δd;

[0126] if Then Mat(loc i ,loc p )=1, indicating successful spatial matching;

[0127] if Then Mat(loc i ,loc p )=0, indicating that the spatial matching is unsuccessful;

[0128] In the formula, loc i =(lon i ,lat i ),loc p =(lon p ,lat p ), indicating the geographical location of different data, loc i and loc p Represents loc s ,loc a and locc , indicating the geographical location of different data, loc can be used for matching s ,loc a and loc c Cross-substitute into loc i and loc p Perform spatial matching of multiple data in

[0129] Set the multidimensional dataset to d fused =(D s , D a , D c );

[0130] Accurately identify and correct abnormal data points through intelligent algorithms.

[0131] As a further aspect of the present invention, regarding the accurate identification and correction of abnormal data points through intelligent algorithms, the correction process is as follows:

[0132] Set d in multi-dimensional data fused The data sequence in is X = {x1, x2, ..., x n}, x n represents the nth data sequence, and n represents the number of data in the data sequence X;

[0133] Calculate the mean of the data series using the mean calculation formula and standard deviation calculation formula And standard deviation σ:

[0134]

[0135] In the formula, x f represents the fth data value in the data sequence;

[0136] Set the volatility multiple threshold to k;

[0137] if Then determine x f is an abnormal data point, marked as x e ;

[0138] if Then determine x f is a normal data point;

[0139] For abnormal data points x e Make corrections. The correction process is as follows:

[0140] Select the abnormal data point x e The neighboring q data are {x e-l , x e-l+1 , …, x e-1 , x e+1 , xe+r}, where l and r are natural numbers greater than 0, l+r=q, and q is an odd number;

[0141]

[0142] In the formula, represents the corrected data point, and iz represents the index variable.

[0143] As a further aspect of the present invention, in the cloud computing module, carbon emission models for different industries are constructed, and the construction process is as follows:

[0144] Set the ibth enterprise industry feature vector in the multidimensional data set to XH ib =(xh1, xh2, ..., xh b ), where b represents the total number of enterprise industry characteristic vectors, ib = 1, 2, ..., b;

[0145] The enterprise industry feature vector is standardized, and the processing process is as follows:

[0146] The mean of the enterprise industry characteristics calculated by the mean calculation formula and the standard deviation calculation formula and standard deviation Calculate the standardized enterprise industry characteristic vector Enterprise industry characteristic vector The calculation formula is as follows:

[0147]

[0148] Construct a model library containing a variety of carbon emission models for different industries, and set the carbon emission model MH = {MH1, MH2, ..., MH mf}, mf represents the total number of carbon emission models, and each carbon emission model MH kf All correspond to specific industry types, kf = {1, 2, ..., mf};

[0149] For the ibth enterprise, calculate the standardized enterprise industry characteristic vector With each carbon emission model MH kf The similarity between the ideal industry characteristics of the adaptation is calculated as follows:

[0150]

[0151] In the formula, xh ib，j represents the original value of the ib-th enterprise on the j-th feature dimension, MH kf，j Represents the carbon emission model MH on the jth feature dimension kf , j represents the index variable, Indicates that the ibth enterprise and MHkf The similarity between them ranges from [-1, 1];

[0152] According to the calculated similarity, the most suitable carbon emission model MH is matched for the ibth enterprise. kf , calculate the model index j that maximizes the similarity max , the calculation formula is as follows:

[0153]

[0154] In the formula, It means finding the parameters that make the function maximum;

[0155] Randomly select dc data points as the initial cluster centers, marked as μ ju , where ju = 1, 2, ..., dc;

[0156] For the standardized enterprise industry feature vector in the multidimensional data set Calculate the μ for each cluster center ju The distance is calculated as follows:

[0157]

[0158] In the formula, express With each cluster center μ ju distance;

[0159] For each cluster, recalculate its cluster center μ ju (ca+1), calculated as follows:

[0160]

[0161] In the formula, ca represents the number of iterations, μj u (ca) represents the set of data points belonging to the juth cluster at the cath iteration;

[0162] Set the maximum number of iterations of the cluster center μju(ca+1) to Ds max ;

[0163] Keep updating μju(ca+1) until the maximum number of iterations Ds is reached max , the iteration stops.

[0164] As a further aspect of the present invention, regarding the integration module, the carbon emissions of different enterprises are classified and summarized through a multi-dimensional data set, and the classification and summary process is as follows:

[0165] Set the enterprise data set in the multidimensional data set to EQ = {eq1, eq2, ..., eqnq}, nq represents the total number of enterprises, each eq ib The industry category set IH, time period set IT, geographical region set IR, enterprise size set IS and carbon emission data set D of different enterprises represent the ib-th enterprise c ;

[0166] At this time, the classification function is set to CL = (IH, IT, IR, IS, D c );

[0167] Setting BK T It is a B-tree index built based on timestamps. The timestamp range is [t start , t end ];

[0168]

[0169] In the formula, QK T (t start , t end , B.K. T ) represents the B-tree query function, t start Indicates the starting timestamp, t end Indicates the end timestamp. Indicates the timestamp corresponding to the enterprise data;

[0170] Setting HM I For hash index based on industry category, hash query function QM i (ir,HM I ) can be expressed as:

[0171] QM I (ir,HM I )={eq ib |eq ib ∈EQ,ir};

[0172] In the formula, ir represents the industry category to which the enterprise belongs;

[0173] Calculate the carbon emission data of different companies through the mean calculation formula c The average of corporate carbon emissions is obtained

[0174] Calculate the carbon emission data d of the ibth enterprise c The variance is calculated as follows:

[0175]

[0176] In the formula, CDF 2 represents the variance of corporate carbon emissions;

[0177] Set the enterprise identity set to IFQ = {ifq1, ifq2, ..., ifq nq};

[0178] When drawing a bar chart, use the company logo ifq nq As the horizontal axis parameter, the carbon emission data d c as the vertical axis;

[0179] To identify key enterprises, the identification formula is as follows:

[0180]

[0181] In the formula, IZB(ib) represents the key enterprise judgment index;

[0182] The judgment threshold for key enterprises is set as Φ;

[0183] If IZB(ib)≥Φ, the enterprise is marked as a key enterprise IZB(ib) ZDY ;

[0184] If IZB(ib)<Φ, the enterprise is marked as a normal enterprise IZB(ib) PTY .

[0185] As a further feature of the present invention, in the real-time monitoring and early warning module, by automatically adjusting the threshold setting strategy, when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold, an early warning message is issued. The early warning process is as follows:

[0186] The carbon emissions of ordinary enterprises and key enterprises are set as TPL PTY and TPL ZDY , setting the carbon emission thresholds for ordinary enterprises and key enterprises to be FPt PTY and FPt ZDY ;

[0187] If TPL PTY ≥FPt PTY , then determine the general enterprise IZB(ib) PTY If carbon emissions exceed the limit, the first warning instruction will be issued;

[0188] If TPL PTY <FPt PTY , then determine the general enterprise IZB(ib) PTY Carbon emissions are normal and no instructions are issued;

[0189] If TPL ZDY ≥FPt ZDY , then determine the key enterprise IZB (ib) ZDY Carbon emissions exceed the standard and a second warning order is issued;

[0190] If TPL ZDY <FPt ZDY , then determine the key enterprise IZB (ib) ZDY Carbon emissions are normal and no instructions are issued;

[0191] After receiving the first warning instruction, the alarm flashes to give a warning;

[0192] After receiving the second warning command, the alarm will light up constantly to give a warning;

[0193] Adjust the threshold FPt based on the upgrade status of the industry to which the enterprise belongs PTY and FPt ZDY .

[0194] As a further aspect of the present invention, the threshold FPt is adjusted based on the upgrade status of the industry to which the enterprise belongs. PTY and FPt ZDY , the adjustment process is as follows:

[0195] The industrial upgrading status coefficient of the industries to which ordinary enterprises and key enterprises belong is set as SSJB PTY and SSJB ZDY , the value range is 0-1, 0 means that the enterprise is not in the stage of industrial upgrading, and 1 means that the enterprise is in the stage of deep industrial upgrading;

[0196] F P PTYt =FPt PTY ×(1+ksuq1×SSJB PTY );

[0197] Where, FPt PTYt represents the carbon emission threshold of ordinary enterprises after adjustment, and ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of ordinary enterprises;

[0198] F P ZDYt =FPt ZDY ×(1+ksuq2×SSJB ZDY );

[0199] Where, FPt ZDYt represents the carbon emission threshold of key enterprises after adjustment, ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of key enterprises;

[0200] Introducing external environmental factors to FPt PTYt and FPt ZDYt Continue to adjust.

[0201] As a further aspect of the present invention, regarding the introduction of external environmental factors, FPt PTYt and FPt ZDYtContinue to adjust, the adjustment process is as follows:

[0202] The external environment impact coefficient is set to JXTQ, with a value range of 0.8 to 1.2;

[0203] F P sj =FPt PTYt ×JXTQ;

[0204] Where, FPt sj It represents the carbon emission threshold of an average enterprise after adjusting for external environmental factors;

[0205] F P sjz =FPt ZDYt ×JXTQ;

[0206] Where, FPt sjz It indicates the carbon emission threshold of key enterprises after adjustment by introducing external environmental factors.

[0207] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A carbon emission measurement system based on cloud computing, characterized in that: It includes data collection and intelligent pre-processing module, cloud computing module, integration module and real-time monitoring and early warning module; Data acquisition and intelligent preprocessing module, used to obtain satellite remote sensing monitoring data, urban air quality detection data and carbon emission data of different enterprises, and integrate multiple data to obtain multi-dimensional data sets; The cloud computing module is used to build carbon emission models for different industries, match the most appropriate carbon emission model according to the industry characteristics of different enterprises in the multi-dimensional data set, and use distributed computing technology to measure the carbon emissions of different enterprises; The integration module is used to classify and summarize the carbon emissions of different companies through multi-dimensional data sets, and identify key companies by comparing and analyzing the carbon emissions of different companies and industries in the same time period; The real-time monitoring and early warning module is used to monitor carbon emissions of enterprises in real time. By automatically adjusting the threshold setting strategy, it will issue early warning information when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold; In the data acquisition and intelligent preprocessing module, multiple data are fused to obtain a multi-dimensional data set. The fusion process is as follows: Set the satellite remote sensing monitoring data set as D s , each piece of data t s Indicates the satellite remote sensing monitoring data timestamp, loc s Indicates the geographic location of satellite remote sensing monitoring data, Indicates the msth satellite remote sensing monitoring index value; Set the urban air quality detection data set as D a , each piece of data t a Indicates the timestamp of urban air quality detection data, loc a Indicates the geographical location of the city's air quality detection data. Indicates the value of the ma-th air quality detection index; Assume that the carbon emission data set of different enterprises is D c , each carbon emission data t c Indicates the carbon emission data timestamp of the enterprise, loc c Indicates the geographical location of the company’s carbon emissions data, Represents the value of the ma-th carbon emission index; Match different data by time, the time matching function Mat(t i , t p )as follows: Set the time tolerance threshold to Δt; If |t i -t p |≤Δt, then Mat(t i , t p )=1, indicating that the time matching is successful; If |t i -t p |>Δt, then Mat(t i , t p )=0, indicating that the time matching is unsuccessful; Where, t i and t p Represents t s ,t a and t c , when matching, t s ,t a and t c Cross-substitute into t i and t p Time matching of multiple data is performed in Match different data through space, spatial matching function Mat(loc i ,loc p )as follows: Set the geographical distance tolerance threshold to Δd; if Then Mat(loc i ,loc p )=1, indicating successful spatial matching; if Then Mat(loc i ,loc p )=0, indicating that the spatial matching is unsuccessful; In the formula, loc i =(lon i ,lat i ),loc p =(lon p ,lat p ), indicating the geographical location of different data, loc i and loc p Represents loc s ,loc a and loc c , indicating the geographical location of different data, loc s ,loc a and loc c Cross-substitute into loc i and loc p Perform spatial matching of multiple data in Set the multidimensional dataset to d fused =(D s , D a , D c ); Accurately identify and correct abnormal data points through intelligent algorithms; Regarding the use of intelligent algorithms to accurately identify and correct abnormal data points, the correction process is as follows: Set d in multi-dimensional data fused The data sequence in is X = {x1, x2, ..., x n }, x n represents the nth data sequence, and n represents the number of data in the data sequence X; Calculate the mean of the data series using the mean calculation formula and standard deviation calculation formula And standard deviation σ: In the formula, x f represents the fth data value in the data sequence; Set the volatility multiple threshold to k; if Then determine x f is an abnormal data point, marked as x e ; if Then determine x f is a normal data point; For abnormal data points x e Make corrections. The correction process is as follows: Select the abnormal data point x e The neighboring q data are {x e-l , x e-l+1 , …, x e-1 , x e+1 , x e+r }, where l and r are natural numbers greater than 0, l+r=q, and q is an odd number; In the formula, represents the corrected data point, iz represents the index variable; Regarding the construction of carbon emission models for different industries in the cloud computing module, the construction process is as follows: Set the ibth enterprise industry feature vector in the multidimensional data set to XH ib =(xh1, xh2, ..., xh b ), where b represents the total number of enterprise industry characteristic vectors, ib = 1, 2, ..., b; The enterprise industry feature vector is standardized, and the processing process is as follows: The mean of the enterprise industry characteristics calculated by the mean calculation formula and the standard deviation calculation formula and standard deviation Calculate the standardized enterprise industry characteristic vector Enterprise industry characteristic vector The calculation formula is as follows: Construct a model library containing a variety of carbon emission models for different industries, and set the carbon emission model MH = {MH1, MH2, ..., MH mf }, mf represents the total number of carbon emission models, and each carbon emission model MH kf All correspond to specific industry types, kf = {1, 2, ..., mf}; For the ibth enterprise, calculate the standardized enterprise industry characteristic vector With each carbon emission model MH kf The similarity between the ideal industry characteristics of the adaptation is calculated as follows: In the formula, xh ib，j represents the original value of the ib-th enterprise on the j-th feature dimension, MH kf，j Represents the carbon emission model MH on the jth feature dimension kf , j represents the index variable, Indicates that the ibth enterprise and MH kf The similarity between them ranges from [-1, 1]; According to the calculated similarity, the most suitable carbon emission model MH is matched for the ibth enterprise. kf , calculate the model index j that maximizes the similarity max , the calculation formula is as follows: In the formula, It means finding the parameters that make the function maximum; Randomly select dc data points as the initial cluster centers, marked as μ ju , where ju = 1, 2, ..., dc; For the standardized enterprise industry feature vector in the multidimensional data set Calculate the μ for each cluster center ju The distance is calculated as follows: In the formula, express With each cluster center μ ju distance; For each cluster, recalculate its cluster center μ ju (ca+1), calculated as follows: In the formula, ca represents the number of iterations, μ ju (ca) represents the set of data points belonging to the juth cluster at the cath iteration; Set the cluster center μ ju (ca+1) The maximum number of iterations is Ds max ; Constantly ju (ca+1) is updated until the maximum number of iterations Ds is reached max , the iteration stops.

2. A carbon emission measurement system based on cloud computing according to claim 1, characterized in that: Regarding the integration module, the carbon emissions of different companies are classified and summarized through multi-dimensional data sets. The classification and summary process is as follows: Set the enterprise data set in the multidimensional data set to EQ = {eq1, eq2, ..., eq nq }, nq represents the total number of enterprises, each eq ib The industry category set IH, time period set IT, geographical area set IR, enterprise size set IS and carbon emission data set D of different enterprises represent the ibth enterprise c ; At this time, the classification function is set to CL = (IH, IT, IR, IS, D c ); Setting BK T It is a B-tree index built based on timestamps. The timestamp range is [t start , t end ]; In the formula, QK T (t start , t end , B.K. T ) represents the B-tree query function, t start Indicates the starting timestamp, t end Indicates the end timestamp. Indicates the timestamp corresponding to the enterprise data; Setting HM I For hash index based on industry category, hash query function QM I (ir,HM I ) is expressed as: QM I (ir,HM I )={eq ib |eq ib ∈EQ,ir}; In the formula, ir represents the industry category to which the enterprise belongs; Calculate the carbon emission data of different companies through the mean calculation formula c The average of corporate carbon emissions is obtained Calculate the carbon emission data d of the ibth enterprise c The variance is calculated as follows: In the formula, CDF 2 represents the variance of corporate carbon emissions; Set the enterprise identity set to IFQ = {ifq1, ifq2, ..., ifq nq }; When drawing a bar chart, use the company logo ifq nq As the horizontal axis parameter, the carbon emission data d c as the vertical axis; To identify key enterprises, the identification formula is as follows: In the formula, IZB(ib) represents the key enterprise judgment index; The judgment threshold for key enterprises is set as Φ; If IZB(ib)≥Φ, the enterprise is marked as a key enterprise IZB(ib) ZDY ; If IZB(ib)<Φ, the enterprise is marked as a normal enterprise IZB(ib) PTY .

3. The carbon emission measurement system based on cloud computing according to claim 1, characterized in that: In the real-time monitoring and early warning module, by automatically adjusting the threshold setting strategy, when the carbon emissions of ordinary enterprises and key enterprises reach or exceed the early warning threshold, an early warning message is issued. The early warning process is as follows: The carbon emissions of ordinary enterprises and key enterprises are set as TPL PTY and TPL ZDY , setting the carbon emission thresholds for ordinary enterprises and key enterprises to be FPt PTY and FPt ZDY ; If TPL PTY ≥FPt PTY , then determine the general enterprise IZB(ib) PTY If carbon emissions exceed the limit, the first warning instruction will be issued; If TPL PTY <FPt PTY , then determine the general enterprise IZB(ib) PTY Carbon emissions are normal and no instructions are issued; If TPL ZDY ≥FPt ZDY , then determine the key enterprise IZB (ib) ZDY Carbon emissions exceed the limit, and a second warning order is issued; If TPL ZDY <FPt ZDY , then determine the key enterprise IZB (ib) ZDY Carbon emissions are normal and no instructions are issued; After receiving the first warning instruction, the alarm flashes to give a warning; After receiving the second warning command, the alarm will light up constantly to give a warning; Adjust the threshold FPt based on the upgrade status of the industry to which the enterprise belongs PTY and FPt ZDY .

4. The carbon emission measurement system based on cloud computing according to claim 3 is characterized in that: Regarding adjusting the threshold FPt based on the upgrade of the industry to which the enterprise belongs PTY and FPt ZDY , the adjustment process is as follows: The industrial upgrading status coefficient of the industries to which ordinary enterprises and key enterprises belong is set as SSJB PTY and SSJB ZDY , the value range is 0-1, 0 means that the enterprise is not in the stage of industrial upgrading, and 1 means that the enterprise is in the stage of deep industrial upgrading; FPT PTYt =FPt PTY ×(1+ksuq1×SSJB PTY ; Where, FPt PTYt represents the carbon emission threshold of ordinary enterprises after adjustment, and ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of ordinary enterprises; FPT ZDYt =FPt ZDY ×(1+ksuq2×SSJB ZDY ); Where, FPt ZDYt represents the carbon emission threshold of key enterprises after adjustment, ksuq1 represents the threshold adjustment coefficient corresponding to the industrial upgrading stage of key enterprises; Introducing external environmental factors to FPt PTYt and FPt ZDYt Continue to adjust.

5. The carbon emission measurement system based on cloud computing according to claim 4 is characterized in that: Regarding the introduction of external environmental factors, the impact of FPt PTYt and FPt ZDYt Continue to adjust, the adjustment process is as follows: The external environment impact coefficient is set to JXTQ, with a value range of 0.8 to 1.2; FPT sj =FPt PTYt ×JXTQ; Where, FPt sj It represents the carbon emission threshold of an average enterprise after adjusting for external environmental factors; FPT sjz =FPt ZDYt ×JXTQ; Where, FPt sjz It indicates the carbon emission threshold of key enterprises after adjustment by introducing external environmental factors.

Images