A method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis

Abstract

The present invention proposes a method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis, which relates to the field of carbon emission supervision technology, including: collecting enterprise carbon emission related data; preprocessing the collected data to form a standardized data set; performing multi-dimensional analysis on the standardized data set to calculate the carbon emission assessment coefficient; based on the carbon emission assessment coefficient, using a mathematical model to calculate the enterprise carbon emission index, and quantitatively assessing the enterprise carbon emission status; based on the calculated carbon emission index, combined with industry standards and historical data, assessing the current operation of the enterprise, identifying abnormal emissions and optimization space; based on the assessment results, combined with the expert knowledge base, forming energy-saving and emission reduction decision-making suggestions and optimization plans. The present invention can achieve accurate quantitative assessment of the enterprise's carbon emission status, provide scientific energy-saving and emission reduction decision-making suggestions and optimization plans for enterprises, thereby effectively improving the enterprise's carbon emission management level.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission regulation, and in particular to a method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis. Background Art

[0002] The corporate carbon emission supervision system involves multiple dimensions such as energy consumption measurement, process control, and environmental impact assessment, including specific contents such as energy utilization efficiency evaluation, production process optimization, and pollutant emission control. This puts forward strict requirements on the production and operation management capabilities of the enterprise, forcing the enterprise to establish a more standardized and complete carbon emission management mechanism.

[0003] At present, corporate carbon emission monitoring mainly relies on traditional monitoring and statistical methods, including sampling and testing at fixed points, manual collection of energy consumption data, and regular monitoring of environmental parameters. These traditional methods mainly conduct regular sampling and analysis by setting up fixed monitoring points, using manual recording to count energy consumption, and periodically testing relevant parameters through conventional environmental monitoring equipment. At the data analysis level, basic statistical methods are mainly used to process monitoring data, including simple statistical operations such as data aggregation, mean calculation, and trend analysis, and provide data support for corporate management decisions by generating statistical reports and trend analysis charts. Although this analysis method can meet basic management needs, it has obvious deficiencies in in-depth analysis and prediction and early warning.

[0004] Existing carbon emission monitoring and management methods have significant limitations. First, the data collection dimension is insufficient, and the setting of monitoring points and sampling frequency cannot meet actual needs, which makes it difficult to accurately reflect the overall picture of corporate carbon emissions, affecting the representativeness and integrity of the data; second, the data analysis method is too simplistic, only staying at the basic statistical analysis level, unable to achieve in-depth mining of data value, resulting in insufficient ability to identify potential problems and difficulty in discovering the laws and trends hidden in the data; third, the early warning mechanism is not yet sound, lacking systematic abnormality discrimination standards and disposal processes, making it difficult to achieve timely discovery and disposal of abnormal conditions, affecting the emergency response capabilities of enterprises; fourth, the existing methods do not adequately consider the time and space dimensions, ignoring the influence of time and space factors in the emission process, affecting the scientificity and reliability of the evaluation results. These technical bottlenecks not only restrict the improvement of corporate carbon emission management level, but may also lead to the failure of enterprises to obtain the expected results in their investment in energy conservation, emission reduction and environmental protection, affecting the sustainable development capabilities of enterprises. Summary of the invention

[0005] In view of this, the present invention proposes a method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis. Through multi-dimensional data collection and standardized processing, combined with multiple evaluation dimensions such as energy utilization efficiency, process technology level and emission time series attenuation, a complete mathematical model is established to achieve accurate quantitative evaluation of the enterprise's carbon emissions status, and can timely identify abnormal emission events, provide enterprises with scientific energy-saving and emission reduction decision-making recommendations and optimization plans, thereby effectively improving the company's carbon emission management level.

[0006] The technical solution of the present invention is achieved in this way:

[0007] The present invention provides a method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis, comprising:

[0008] S1. Collect enterprise carbon emission related data;

[0009] S2. Preprocess the collected data, including data cleaning, outlier processing, data standardization and data structuring, to form a standardized data set;

[0010] S3. Perform multi-dimensional analysis on the standardized data set and calculate the carbon emission assessment coefficient;

[0011] S4. Based on the carbon emission assessment coefficient, a mathematical model is used to calculate the carbon emission index of the enterprise and quantitatively assess the carbon emission status of the enterprise;

[0012] S5. Based on the calculated carbon emission indicators, combined with industry standards and historical data, evaluate the current operation of the enterprise, identify abnormal emissions and optimization space;

[0013] S6. Based on the evaluation results and combined with the expert knowledge base, formulate decision-making recommendations and optimization plans for energy conservation and emission reduction.

[0014] On the basis of the above technical scheme, preferably, the enterprise carbon emission related data include real-time collected data, spatial distribution data, historical accumulation data and external reference data, the real-time collected data include energy consumption data, production process parameters, environmental monitoring data and equipment operation status data, the spatial distribution data include emission source location information, three-dimensional coordinates of monitoring points and plant terrain data, the historical accumulation data include energy consumption history records, equipment operation history data, process parameter history records and environmental historical monitoring data, the external reference data include industry standards and national standards; among them, environmental monitoring data and environmental historical monitoring data both include pollutant concentration, temperature, humidity, wind direction, wind speed and air pressure.

[0015] On the basis of the above technical solution, preferably, in step S3, the carbon emission assessment coefficient calculated based on the standardized data set includes an energy utilization efficiency coefficient, a process technology level coefficient and an emission time series attenuation coefficient.

[0016] Based on the above technical solution, preferably, the calculation process of energy utilization efficiency is as follows:

[0017] Determine the baseline efficiency η based on the energy consumption history in the standardized dataset 0 and reference energy consumption E 0 , where the base efficiency η 0 The average efficiency selected from the optimal operating period of the production equipment;

[0018] Determine the adjustment coefficient k by analyzing the fluctuation range of energy consumption data in the standardized data set 1 ;

[0019] The equipment efficiency attenuation coefficient μ is obtained by fitting the equipment operation history data in the standardized data set;

[0020] Construct a calculation model for energy efficiency coefficient:

[0021]

[0022] In the formula, is the energy efficiency coefficient of the i-th energy source, is the consumption of the i-th energy, and t is the monitoring time.

[0023] Based on the above technical solution, preferably, the calculation process of the process technology level coefficient is as follows:

[0024] Determine the baseline technology level by combining industry standards and data from the best performance periods in the history of process parameters 0 and standard output P 0 ;

[0025] The exponential coefficient α is determined by analyzing the relationship between output and technology level in the standardized data set;

[0026] Calculate the standard deviation of the production process parameters to get the volatility σ p ;

[0027] Determine the correction factor β based on the impact of volatility on technical level;

[0028] Establish the evaluation equation of process technology level coefficient:

[0029]

[0030] In the formula, is the technological level coefficient of the jth process, is the product output of the j-th process, and t is the monitoring time.

[0031] Based on the above technical solution, preferably, the calculation process of the emission time series attenuation coefficient is as follows:

[0032] Determine the decay rate γ and the natural decay coefficient λ of the emissions by analyzing the decay trends of performance-related indicators in the equipment operation history data;

[0033] Use Fourier analysis to identify the periodic characteristics of the production process and determine the period T and fluctuation intensity ω;

[0034] Construct the expression for the emission time series decay coefficient:

[0035]

[0036] In the formula, is the emission time series attenuation coefficient, and t is the monitoring time.

[0037] Based on the above technical solution, preferably, in step S4, the mathematical model for calculating the carbon emission index of the enterprise is as follows:

[0038]

[0039] In the formula, Indicates the carbon emission index of the enterprise; Indicates the calculation cycle; represents the monitoring time, i.e., the time variable; Represents the spatial position vector; represents a historical time variable; is the carbon emission factor of the i-th energy source; represents the consumption of the i-th energy; represents the energy utilization efficiency coefficient of the i-th energy source; represents the carbon emission factor of the jth process; represents the product output of the jth process; Represents the technological level coefficient of the jth technological process; Represents the system control volume, i.e., the spatial scope for carbon emission calculation; represents the spatial distribution function of emissions; represents the emission flux vector; represents the spatial gradient operator; represents the historical emission intensity; represents the emission time series attenuation coefficient;

[0040] Among them, the carbon emission factor , From external reference data.

[0041] Based on the above technical solution, preferably, the emission space distribution function The calculation formula is as follows:

[0042]

[0043] In the formula, is the spatial distance between the monitoring point and the emission source; The characteristic diffusion radius is determined by analyzing the spatial distribution of pollutant concentrations and combining it with the plant area topographic data; is the wind direction influence coefficient; is the angle with the dominant wind direction, which is calculated based on the wind direction of real-time data collection and the dominant wind direction of environmental historical monitoring data; h is the vertical height of the monitoring point; H is the maximum impact height, which is determined by the vertical distribution law of pollutant concentration in environmental historical monitoring data;

[0044] Emission flux vector The calculation formula is as follows:

[0045]

[0046] In the formula, The baseline emission flux is determined by the standard operating condition data in the process parameter history record; It is the ratio of the emission rate at the current moment to the benchmark emission rate, which is calculated by comparing the production process parameters in the real-time collected data with the standard operating conditions in the historical accumulated data; is the spatial attenuation coefficient, which is obtained by fitting the pollutant concentration in the real-time collected data and the environmental historical monitoring data in the historical accumulation data; is the unit vector of the emission direction;

[0047] Historical emissions intensity The calculation formula is as follows:

[0048]

[0049] In the formula, The baseline emission intensity is determined by combining historical environmental monitoring data and historical records of process parameters; is the seasonal fluctuation coefficient, which is obtained by analyzing the seasonal variation law of environmental historical monitoring data; is the calculation cycle; is the historical time variable; It is the annual decay rate, calculated through long-term trend analysis of historical accumulated data.

[0050] Based on the above technical solution, preferably, step S5 includes:

[0051] S51. Establish an assessment benchmark for corporate carbon emissions indicators, determine the upper and lower limits of the normal operating range by analyzing historical accumulated data, and set early warning thresholds and alarm thresholds in accordance with industry standards;

[0052] S52. Calculate the deviation rate between the company's current carbon emission indicators and the assessment benchmark, and use time series analysis methods to identify trend changes and assess the company's operating status;

[0053] S53. Use statistical test methods to identify abnormal emission events, including instantaneous anomalies and cumulative anomalies, and quantify the degree of anomalies;

[0054] S54. By benchmarking against industry best practices and combining the company's historical optimal operating data, analyze gaps in energy efficiency, process level and emission control to identify potential optimization space;

[0055] S55. Generate an evaluation report, including operation status evaluation, abnormal event analysis and optimization space quantification content.

[0056] Based on the above technical solution, preferably, the method for identifying abnormal emission events in step S53 is as follows:

[0057] Calculate the Normalized Anomaly Index:

[0058]

[0059] In the formula, The carbon emission index of the enterprise at the current moment; is the average value of the carbon emission index of the enterprise in the base period; is the standard deviation of the carbon emission index of the enterprise in the base period;

[0060] Define the exception type judgment criteria:

[0061] when When , it is judged as a transient abnormality;

[0062] when When , it is judged as cumulative abnormality;

[0063] in, is the instantaneous abnormality discrimination coefficient, is the cumulative abnormal discriminant coefficient, is the accumulation period.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] (1) The present invention realizes accurate quantitative evaluation and intelligent management of corporate carbon emissions by establishing a complete data collection, processing, analysis and evaluation system. The system covers the entire process from data collection to final decision-making, including key links such as data cleaning, outlier processing, data standardization and data structuring, which improves the accuracy and reliability of carbon emission monitoring. Through systematic data processing and analysis methods, it provides a scientific basis for corporate energy conservation and emission reduction decisions, and effectively supports companies to achieve refined management and continuous improvement;

[0066] (2) The present invention constructs a three-dimensional evaluation coefficient system, including energy efficiency coefficient, process technology level coefficient and emission time series attenuation coefficient. The energy efficiency coefficient is quantified through the mathematical model of benchmark efficiency, adjustment coefficient and efficiency attenuation coefficient; the process technology level coefficient is calculated based on the correction of benchmark technology level, production index and parameter fluctuation; the emission time series attenuation coefficient is characterized by combining equipment performance attenuation rate, natural attenuation coefficient and periodic characteristic function. This evaluation system realizes the systematic quantification of factors affecting corporate carbon emissions;

[0067] (3) The present invention provides a spatiotemporal coupled carbon emission index calculation model, which realizes the quantitative characterization of the carbon emission process by accurately solving the emission spatial distribution function and the emission flux vector. The model comprehensively considers factors such as spatial distance attenuation, wind direction influence coefficient, and vertical distribution characteristics, and describes the time evolution characteristics of the emission process through the historical emission intensity function, thus establishing an accurate mathematical description of the carbon emission process;

[0068] (4) The present invention proposes an abnormal event identification method based on a standardized abnormal index, and establishes a dual abnormality discrimination mechanism through the instantaneous abnormality discrimination coefficient and the cumulative abnormality discrimination coefficient. This method can effectively distinguish between instantaneous abnormalities and cumulative abnormal events, achieve quantitative rating of the degree of abnormality, and provide a scientific basis for the abnormal management and control of carbon emissions of enterprises;

[0069] (5) The present invention establishes a systematic evaluation and optimization mechanism, and identifies the optimization space of enterprises in terms of energy efficiency, process level and emission control through benchmarking analysis methods. The system sets the normal operating range limit values ​​based on historical data and the early warning thresholds based on industry standards, and realizes real-time monitoring of the operating status. Through time series analysis and statistical test methods, the system can identify trend changes and abnormal events in carbon emissions, and form a professional evaluation report including operation evaluation, abnormal analysis and optimization suggestions, providing decision support for the continuous improvement of carbon emission management of enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0071] Figure 1 is a flow chart of the method of the present invention;

[0072] Figure 2 It is a technical implementation diagram of the present invention. DETAILED DESCRIPTION

[0073] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] like Figure 1 and Figure 2 As shown, the present invention provides a method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis, including:

[0075] S1. Collect enterprise carbon emission related data;

[0076] S2. Preprocess the collected data, including data cleaning, outlier processing, data standardization and data structuring, to form a standardized data set;

[0077] S3. Perform multi-dimensional analysis on the standardized data set and calculate the carbon emission assessment coefficient;

[0078] S4. Based on the carbon emission assessment coefficient, a mathematical model is used to calculate the carbon emission index of the enterprise and quantitatively assess the carbon emission status of the enterprise;

[0079] S5. Based on the calculated carbon emission indicators, combined with industry standards and historical data, evaluate the current operation of the enterprise, identify abnormal emissions and optimization space;

[0080] S6. Based on the evaluation results and combined with the expert knowledge base, formulate decision-making recommendations and optimization plans for energy conservation and emission reduction.

[0081] Specifically, in one embodiment of the present invention, enterprise carbon emission-related data include real-time collected data, spatial distribution data, historical accumulation data and external reference data. The real-time collected data include energy consumption data, production process parameters, environmental monitoring data and equipment operation status data. The spatial distribution data include emission source location information, three-dimensional coordinates of monitoring points and plant terrain data. The historical accumulation data include energy consumption history records, equipment operation history data, process parameter history records and environmental historical monitoring data. The external reference data include industry standards and national standards. Among them, environmental monitoring data and environmental historical monitoring data both include pollutant concentration, temperature, humidity, wind direction, wind speed and air pressure.

[0082] In this embodiment, the data collection process adopts a systematic multi-dimensional collection method, and through the establishment of a complete data collection system, the comprehensive acquisition of enterprise carbon emission related data is achieved. Data collection is divided into four categories: real-time collection data, spatial distribution data, historical accumulation data and external reference data. Each type of data is obtained through a specific collection method and channel, and is stored and managed according to a unified data specification.

[0083] In terms of real-time data collection, energy consumption data is collected through automation equipment such as smart meters and flow meters, production process parameters are collected using DCS systems and PLC controllers, environmental monitoring data is collected through online monitoring systems and weather stations, and equipment operating status data is collected through equipment status monitoring systems. The frequency of data collection ranges from 1 minute to 15 minutes.

[0084] Spatial distribution data is collected mainly through GIS systems, GPS positioning and field measurements, including the acquisition of emission source location information, three-dimensional coordinates of monitoring points and plant terrain data. This type of data needs to be updated and verified regularly, usually once a quarter or a year.

[0085] Historical accumulated data is exported through various management systems, including historical data from energy management systems, equipment management systems, production management systems, and environmental monitoring systems. These data are kept for the past 2-3 years and are used to establish benchmarks and analyze trends. External reference data are mainly industry standards and national standards, which need to be updated in a timely manner as standards are updated.

[0086] Specifically, in one embodiment of the present invention, step S2 includes:

[0087] Data preprocessing begins with data cleaning, using corresponding cleaning methods for different types of data. For real-time data collection, linear interpolation is used to supplement short-term missing data for energy consumption data, forward filling is used to maintain the most recent valid value for process parameters, and moving average filling based on time windows is used for environmental monitoring data. The status values ​​in the equipment operation status data remain unchanged, and linear interpolation is used for numerical parameters. For spatially distributed data, the rationality of coordinate data is mainly verified, including checking the coordinate range and verifying the consistency of the coordinate system. At the same time, the terrain data is checked for continuity and abnormal mutation points are corrected. The focus of cleaning historical accumulated data is the integrity check of the time series, including supplementing missing time points, unifying the timestamp format, and performing data consistency checks to ensure unified units and standardized formats. External reference data mainly unifies text formats and extracts key parameters.

[0088] A hierarchical detection method is used in the outlier processing stage. For real-time collected data, the 3σ principle is first used for preliminary screening, and then the moving Z-score method based on the sliding window is used for fine detection. At the same time, reasonable parameter range limits are set in combination with the characteristics of different data types. For spatially distributed data, spatial statistical methods are used in combination with GIS systems for anomaly detection, and verified in combination with actual terrain features. For historical accumulated data, seasonal decomposition methods are mainly used to identify anomalies, and comparative analysis is performed in combination with historical data of the same period. For detected outliers, obviously erroneous values ​​are directly deleted, and suspicious outliers are marked and retained, while abnormal characteristics are recorded for subsequent analysis.

[0089] The data standardization process uses corresponding standardization methods for different types of data. For numerical data, energy consumption data is normalized using the minimum-maximum standardization, process parameters are normalized using the Z-score standardization, and environmental monitoring data is normalized according to standard limits. The standardization of spatial data includes unified coordinate system conversion and elevation data normalization. The standardization of time series data mainly involves the unification of timestamp formats and the standardization of sampling frequencies.

[0090] Finally, the data is structured and a relational database model including real-time data tables, spatial data tables, historical data tables and standard parameter tables is established. The time association, spatial association, equipment association and process association among data are also established. In terms of storage structure, real-time data is stored in a time series database, spatial data is stored in a GIS database, historical data is stored in a relational database, and reference data is stored in a document database. During the entire preprocessing process, quality control measures such as integrity inspection, accuracy inspection, consistency inspection and timeliness inspection are used to ensure the reliability of the preprocessing results. At the same time, a data processing traceability mechanism is established to record the processing process, abnormal conditions and processing results.

[0091] Specifically, in one embodiment of the present invention, the goal of step S3 is to calculate the carbon emission assessment coefficient through multi-dimensional analysis of the standardized data set. Based on the characteristics of corporate carbon emissions, the assessment coefficient system includes three key dimensions: energy efficiency coefficient, process technology level coefficient and emission time series attenuation coefficient. The selection of these three dimensions is based on the three core characteristics of corporate carbon emissions: energy consumption characteristics, production process characteristics and spatiotemporal evolution characteristics.

[0092] The calculation process of energy efficiency is as follows:

[0093] Determine the baseline efficiency η based on the energy consumption history in the standardized dataset 0 and reference energy consumption E 0 , where the base efficiency η 0 The average efficiency selected from the optimal operating period of the production equipment;

[0094] Determine the adjustment coefficient k by analyzing the fluctuation range of energy consumption data in the standardized data set 1 ;

[0095] Combined with the historical data of equipment operation, the regression analysis method is used to fit the equipment efficiency attenuation coefficient μ, which reflects the attenuation characteristics of equipment performance over time;

[0096] Construct a calculation model for energy efficiency coefficient:

[0097]

[0098] In the formula, is the energy efficiency coefficient of the i-th energy source, is the consumption of the i-th energy, and t is the monitoring time.

[0099] In this embodiment, the energy efficiency coefficient The design principle is based on the logarithmic response characteristics of energy efficiency. In the formula, It reflects the nonlinear relationship between energy efficiency and consumption. Characterizes the natural degradation of equipment performance over time, based on the law that equipment aging leads to reduced energy efficiency.

[0100] The calculation process of the process technology level coefficient is as follows:

[0101] Determine the baseline technology level by combining industry standards and data from the best performance periods in the history of process parameters 0 and standard output P 0 ;

[0102] The exponential coefficient α is determined by analyzing the relationship between output and technology level in the standardized data set;

[0103] Calculate the standard deviation of the production process parameters to get the volatility σ p ;

[0104] Determine the correction factor β based on the impact of volatility on technical level;

[0105] Establish the evaluation equation of process technology level coefficient:

[0106]

[0107] In the formula, is the technological level coefficient of the jth process, is the product output of the j-th process, and t is the monitoring time.

[0108] In this embodiment, the process technology level coefficient The design of combines scale effect and stability evaluation. In the formula, The item reflects the scale effect. As an exponential coefficient, it reflects the impact of output changes on technological level. The term reflects the influence of process stability, which is expressed by the fluctuation rate To quantify the negative impact of fluctuations in process parameters on the technology level.

[0109] The calculation process of the emission time series attenuation coefficient is as follows:

[0110] Determine the decay rate γ and the natural decay coefficient λ of the emissions by analyzing the decay trends of performance-related indicators in the equipment operation history data;

[0111] Use Fourier analysis to identify the periodic characteristics of the production process and determine the period T and fluctuation intensity ω;

[0112] Construct the expression for the emission time series decay coefficient:

[0113]

[0114] In the formula, is the emission time series attenuation coefficient, and t is the monitoring time.

[0115] In this embodiment, the emission time series attenuation coefficient The design takes into account three key factors: equipment performance degradation , natural attenuation of pollutants and production periodicity This composite function can fully describe the time evolution characteristics of the emission process.

[0116] Specifically, in one embodiment of the present invention, the core of step S4 is to construct a comprehensive carbon emission index calculation model, which comprehensively considers the influence of four dimensions: energy consumption, process, spatial distribution and historical accumulation. The mathematical model for calculating the carbon emission index of an enterprise is as follows:

[0117]

[0118] In the formula, Indicates the carbon emission index of the enterprise; Indicates the calculation cycle; represents the monitoring time, i.e., the time variable; Represents the spatial position vector; represents a historical time variable; is the carbon emission factor of the i-th energy source; represents the consumption of the i-th energy; represents the energy utilization efficiency coefficient of the i-th energy source; represents the carbon emission factor of the jth process; represents the product output of the jth process; Represents the technological level coefficient of the jth technological process; Represents the system control volume, i.e., the spatial scope for carbon emission calculation; represents the spatial distribution function of emissions; represents the emission flux vector; represents the spatial gradient operator; represents the historical emission intensity; represents the emission time series attenuation coefficient;

[0119] Among them, the carbon emission factor , From external reference data.

[0120] Emission spatial distribution function The calculation formula is as follows:

[0121]

[0122] In the formula, is the spatial distance between the monitoring point and the emission source; The characteristic diffusion radius is determined by analyzing the spatial distribution of pollutant concentrations and combining it with the plant area topographic data; is the wind direction influence coefficient; is the angle with the dominant wind direction, which is calculated based on the wind direction of real-time data collection and the dominant wind direction of environmental historical monitoring data; h is the vertical height of the monitoring point; H is the maximum impact height, which is determined by the vertical distribution law of pollutant concentration in environmental historical monitoring data;

[0123] Emission flux vector The calculation formula is as follows:

[0124]

[0125] In the formula, The baseline emission flux is determined by the standard operating condition data in the process parameter history record; It is the ratio of the emission rate at the current moment to the benchmark emission rate, which is calculated by comparing the production process parameters in the real-time collected data with the standard operating conditions in the historical accumulated data; is the spatial attenuation coefficient, which is obtained by fitting the pollutant concentration in the real-time collected data and the environmental historical monitoring data in the historical accumulation data; is the unit vector of the emission direction;

[0126] Historical emissions intensity The calculation formula is as follows:

[0127]

[0128] In the formula, The baseline emission intensity is determined by combining historical environmental monitoring data and historical records of process parameters; is the seasonal fluctuation coefficient, which is obtained by analyzing the seasonal variation law of environmental historical monitoring data; is the calculation cycle; is the historical time variable; It is the annual decay rate, calculated through long-term trend analysis of historical accumulated data.

[0129] In this embodiment, CE is calculated by converting discrete monitoring data into continuous evaluation results in the form of multiple integrations, thereby achieving accurate quantification of the carbon emission status of the enterprise.

[0130] The first term is the time impact term. In this time dimension, the model uses a definite integral form to calculate the cumulative effect within the period T. Energy consumption term The comprehensive impact of carbon emission factors, actual consumption and energy efficiency of different energy types is taken into account; process items It reflects the combined effect of carbon emission factors, product output and process technology level of different processes. These two items constitute the main source of carbon emissions, and the overall emissions within the cycle can be obtained through time integration.

[0131] The second term is the spatial impact term. In this spatial dimension, the model describes the diffusion process of emissions in space in the form of volume integrals. Spatial distribution function It describes the distribution of emissions in space. Its calculation needs to consider three factors: distance attenuation, wind direction and height. The flow characteristics of the emissions are characterized, including the baseline emission flux, the real-time emission rate and the spatial attenuation effect. The product of these two functions is calculated through the spatial gradient operator ∇ and then integrated in the control volume V to obtain the spatial distribution effect.

[0132] The third item is the historical cumulative impact item, which reflects the cumulative impact of historical emissions. Among them, the historical emission intensity M(τ) describes the changing pattern of historical emission levels through three factors: baseline emission intensity, seasonal fluctuations, and annual decay; the emission time series decay coefficient η considers the combined impact of equipment performance decay, natural decay of emissions, and production periodicity. By integrating historical time, the cumulative impact of historical emissions on the current situation is calculated.

[0133] Specifically, in the implementation of this model, it is necessary to first determine the values ​​of various parameters, including carbon emission factors, baseline values ​​and various coefficients. Then, based on real-time monitoring data and historical data, calculate the emission intensity at each time point. Finally, through the numerical integration method, solve the multiple integrals to obtain the final carbon emission index. This method takes into account both the immediate effect of emissions and the cumulative impact, and can comprehensively and accurately evaluate the carbon emission status of enterprises.

[0134] Specifically, in one embodiment of the present invention, step S5 includes:

[0135] S51. Establish an assessment benchmark for corporate carbon emission indicators, determine the upper and lower limits of the normal operating range by analyzing historical accumulated data, and set early warning thresholds and alarm thresholds in accordance with industry standards.

[0136] Specifically, by statistically analyzing historical accumulated data, the distribution characteristics of the company's carbon emission indicators under normal operating conditions are calculated, and the upper and lower limits of the normal operating range are determined. The specific method is: first, filter the normal operating period data in the historical data, calculate the mean and standard deviation of the carbon emission indicators; then determine the normal operating range based on the 3σ principle; finally, combine industry standards to set the early warning threshold (such as 2σ) and alarm threshold (such as 3σ).

[0137] S52. Calculate the deviation rate between the company's current carbon emission indicators and the assessment benchmark, and use time series analysis methods to identify trend changes and evaluate the company's operating status.

[0138] Specifically, the deviation rate between the company's current carbon emission index and the assessment benchmark is calculated, and the time series analysis method is used to identify the changing trend of the index. In specific implementation, the relative deviation between the current index and the benchmark mean is first calculated; then the moving average method and trend analysis method are used to identify the changing trend of the index; finally, according to the size of the deviation rate and the direction of the trend, the company's operating status is divided into different categories such as stable operation, fluctuating operation, and trend deterioration.

[0139] S53. Use statistical test methods to identify abnormal emission events, including instantaneous anomalies and cumulative anomalies, and quantify the degree of anomaly rating.

[0140] Specifically, the normalized anomaly index is calculated:

[0141]

[0142] In the formula, The carbon emission index of the enterprise at the current moment; is the average value of the carbon emission index of the enterprise in the base period; is the standard deviation of the carbon emission index of the enterprise in the base period;

[0143] Define the exception type judgment criteria:

[0144] when When , it is judged as a transient abnormality;

[0145] when When , it is judged as cumulative abnormality;

[0146] in, is the instantaneous abnormality discrimination coefficient, is the cumulative abnormal discriminant coefficient, is the accumulation period.

[0147] According to the size of the abnormality index, the degree of abnormality is quantitatively rated, specifically: slight abnormality (1-2 times the standard deviation), moderate abnormality (2-3 times the standard deviation) and severe abnormality (more than 3 times the standard deviation).

[0148] S54. By benchmarking against industry best practices and combining the company's historical optimal operating data, analyze the gaps in energy efficiency, process level and emission control to identify potential optimization space.

[0149] Specifically, we first establish a multi-dimensional benchmark database, extract three evaluation coefficients by screening the optimal operating period in the historical operating data of enterprises, collect the operating indicators of industry benchmark enterprises, and use the Z-score standardization method to standardize the data. Construct the industry benchmark matrix B = [b ij ], where b ij Indicates the benchmark value of the jth indicator in the i-th dimension.

[0150] Quantitative analysis of three evaluation coefficients:

[0151] Energy Efficiency Gap Index:

[0152]

[0153] in is the benchmark energy efficiency coefficient, is the current energy efficiency coefficient.

[0154] Technology level gap index:

[0155]

[0156] in is the benchmark process technology level coefficient, is the coefficient of current process technology level.

[0157] Emission Control Gap Index:

[0158]

[0159] in is the baseline emission time series attenuation coefficient, is the current emission time series attenuation coefficient.

[0160] According to the size of each gap index, each dimension is divided into: slight gap (<10%), medium gap (10%-30%), and significant gap (>30%). Identify the dimension with the largest gap index as the key optimization direction. Specific suggestions are made for different dimensions:

[0161] In terms of energy efficiency: improving equipment efficiency and optimizing energy management; in terms of process level: optimizing process parameters and upgrading technology; in terms of emission control: improving emission treatment facilities and optimizing operation management.

[0162] S55. Generate an evaluation report, including operation status evaluation, abnormal event analysis and optimization space quantification content.

[0163] Specifically, the above analysis results are combined to generate a system evaluation report. The report content includes: the operation status evaluation part, which summarizes the overall evaluation of the current operation status of the enterprise and the analysis of key indicators; the abnormal event analysis part, which describes in detail the abnormal events found, the degree of abnormality and the possible cause analysis; the optimization space quantification part, which lists the optimization space of each dimension and specific improvement suggestions. The report adopts a combination of pictures and texts, with both quantitative data analysis and qualitative evaluation conclusions.

[0164] Specifically, in one embodiment of the present invention, step S6 includes:

[0165] Based on the evaluation report generated by S5, the carbon emission problems of enterprises are divided into three categories: abnormal operation status, equipment efficiency deviation and process parameter deviation. For each type of problem, key characteristic parameters are extracted, including the degree of abnormality, duration and scope of impact, and the problem feature vector is established.

[0166] The problem feature vector is matched with the cases in the expert knowledge base for similarity. The knowledge base contains historical solutions, industry best practices, and expert processing experience. Each case includes a problem description, solution, and implementation effect. By calculating the similarity of the feature vector, the most similar solution is selected.

[0167] According to the matching results and the actual situation of the enterprise, a hierarchical optimization plan is generated: short-term optimization suggestions: for operating parameter adjustments and management measures improvements; mid-term improvement plans: for equipment efficiency improvement and process optimization; long-term development plans: for technological transformation and equipment upgrades.

[0168] Conduct feasibility assessment on the generated optimization scheme, mainly considering three aspects: implementation cost, technical difficulty and expected effect. Through simple cost-benefit analysis, determine the priority of the scheme and form decision-making suggestions for step-by-step implementation.

[0169] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis, characterized in that: include: S1. Collect enterprise carbon emission related data; S2. Preprocess the collected data, including data cleaning, outlier processing, data standardization and data structuring, to form a standardized data set; S3. Perform multi-dimensional analysis on the standardized data set and calculate the carbon emission assessment coefficient; The carbon emission assessment coefficients calculated based on the standardized data set include energy efficiency coefficient, process technology level coefficient and emission time series attenuation coefficient; The calculation process of energy efficiency is as follows: Determine the benchmark efficiency η0 and reference energy consumption E0 based on the energy consumption history records in the standardized data set, where the benchmark efficiency η0 is selected from the average efficiency of the production equipment during the optimal operation period; The adjustment coefficient k1 is determined by analyzing the fluctuation range of energy consumption data in the standardized data set; The equipment efficiency attenuation coefficient μ is obtained by fitting the equipment operation history data in the standardized data set; Construct a calculation model for energy efficiency coefficient: ， In the formula, is the energy efficiency coefficient of the i-th energy source, is the consumption of the i-th energy source, and t is the monitoring time; The calculation process of the process technology level coefficient is as follows: Determine the benchmark technology level φ0 and standard output P0 by combining industry standards and data from the best performance period in the history of process parameters; The exponential coefficient α is determined by analyzing the relationship between output and technology level in the standardized data set; Calculate the standard deviation of the production process parameters to get the volatility σ p ; Determine the correction factor β based on the impact of volatility on technical level; Establish the evaluation equation of process technology level coefficient: ， In the formula, is the technological level coefficient of the jth process, is the product output of the jth process, and t is the monitoring time; The calculation process of the emission time series attenuation coefficient is as follows: Determine the decay rate γ and the natural decay coefficient λ of the emissions by analyzing the decay trends of performance-related indicators in the equipment operation history data; Use Fourier analysis to identify the periodic characteristics of the production process and determine the period T and fluctuation intensity ω; Construct the expression for the emission time series decay coefficient: ， In the formula, is the emission time series attenuation coefficient, t is the monitoring time; S4. Based on the carbon emission assessment coefficient, a mathematical model is used to calculate the carbon emission index of the enterprise and quantitatively assess the carbon emission status of the enterprise; S5. Based on the calculated carbon emission indicators, combined with industry standards and historical data, evaluate the current operation of the enterprise, identify abnormal emissions and optimization space; S6. Based on the evaluation results and combined with the expert knowledge base, formulate decision-making recommendations and optimization plans for energy conservation and emission reduction.

2. The method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis as claimed in claim 1, characterized in that: Enterprise carbon emission-related data include real-time collected data, spatial distribution data, historical accumulation data and external reference data. Real-time collected data include energy consumption data, production process parameters, environmental monitoring data and equipment operation status data. Spatial distribution data include emission source location information, three-dimensional coordinates of monitoring points and plant terrain data. Historical accumulation data include energy consumption history records, equipment operation history data, process parameter history records and environmental historical monitoring data. External reference data include industry standards and national standards. Among them, environmental monitoring data and environmental historical monitoring data both include pollutant concentration, temperature, humidity, wind direction, wind speed and air pressure.

3. The method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis as claimed in claim 1, characterized in that: In step S4, the mathematical model for calculating the enterprise carbon emission index is as follows: ， In the formula, Indicates the carbon emission index of the enterprise; Indicates the calculation cycle; represents the monitoring time, i.e., the time variable; Represents the spatial position vector; represents a historical time variable; is the carbon emission factor of the i-th energy source; represents the consumption of the i-th energy; represents the energy utilization efficiency coefficient of the i-th energy source; represents the carbon emission factor of the jth process; represents the product output of the jth process; Represents the technological level coefficient of the jth technological process; Represents the system control volume, i.e., the spatial scope for carbon emission calculation; represents the spatial distribution function of emissions; represents the emission flux vector; represents the spatial gradient operator; represents the historical emission intensity; represents the emission time series attenuation coefficient; Among them, the carbon emission factor , From external reference data.

4. The method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis as claimed in claim 3, characterized in that: Emission spatial distribution function The calculation formula is as follows: ， In the formula, is the spatial distance between the monitoring point and the emission source; The characteristic diffusion radius is determined by analyzing the spatial distribution of pollutant concentrations and combining it with the plant area topographic data; is the wind direction influence coefficient; is the angle with the dominant wind direction, which is calculated based on the wind direction of real-time data collection and the dominant wind direction of environmental historical monitoring data; h is the vertical height of the monitoring point; H is the maximum impact height, which is determined by the vertical distribution law of pollutant concentration in environmental historical monitoring data; Emission flux vector The calculation formula is as follows: ， In the formula, The baseline emission flux is determined by the standard operating condition data in the process parameter history record; It is the ratio of the emission rate at the current moment to the benchmark emission rate, which is calculated by comparing the production process parameters in the real-time collected data with the standard operating conditions in the historical accumulated data; is the spatial attenuation coefficient, which is obtained by fitting the pollutant concentration in the real-time collected data and the environmental historical monitoring data in the historical accumulation data; is the unit vector of the emission direction; Historical emissions intensity The calculation formula is as follows: ， In the formula, The baseline emission intensity is determined by combining historical environmental monitoring data and historical records of process parameters; is the seasonal fluctuation coefficient, which is obtained by analyzing the seasonal variation law of environmental historical monitoring data; is the calculation cycle; is the historical time variable; It is the annual decay rate, calculated through long-term trend analysis of historical accumulated data.

5. The method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis as claimed in claim 1, characterized in that: Step S5 includes: S51. Establish an assessment benchmark for corporate carbon emissions indicators, determine the upper and lower limits of the normal operating range by analyzing historical accumulated data, and set early warning thresholds and alarm thresholds in accordance with industry standards; S52. Calculate the deviation rate between the company's current carbon emission indicators and the assessment benchmark, and use time series analysis methods to identify trend changes and assess the company's operating status; S53. Use statistical test methods to identify abnormal emission events, including instantaneous anomalies and cumulative anomalies, and quantify the degree of anomalies; S54. By benchmarking against industry best practices and combining the company's historical optimal operating data, analyze gaps in energy efficiency, process level and emission control to identify potential optimization space; S55. Generate an evaluation report, including operation status evaluation, abnormal event analysis and optimization space quantification content.

6. The method for intelligent monitoring and management of enterprise carbon emissions based on big data analysis as claimed in claim 5, characterized in that: The method for identifying abnormal emission events in step S53 is as follows: Calculate the Normalized Anomaly Index: ， In the formula, The carbon emission index of the enterprise at the current moment; is the average value of the carbon emission index of the enterprise in the base period; is the standard deviation of the carbon emission index of the enterprise in the base period; Define the exception type judgment criteria: when When , it is judged as a transient abnormality; when When , it is judged as cumulative abnormality; in, is the instantaneous abnormality discrimination coefficient, is the cumulative abnormal discriminant coefficient, is the accumulation period.

Images