A method and system for calculating enterprise energy consumption and carbon emissions based on differentiated calibration

Abstract

This invention proposes a method and system for calculating enterprise energy consumption and carbon emissions based on differentiated calibration. This method first collects historical production and operation information from enterprises in the same industry within a region. Then, based on this production and operation information, all enterprises are clustered and analyzed. The enterprise at the center of each cluster is selected as a benchmark enterprise. The benchmark enterprise's installed energy consumption monitoring system then obtains its current measured energy consumption (the current measured carbon emissions are calculated based on the current measured energy consumption). Based on this historical data, the current energy consumption and carbon emissions forecasts for all enterprises in each cluster are then calculated. The deviation rate between the measured and predicted values ​​is then calculated. Based on the benchmark enterprise's deviation rate and taking into account cluster spatial distance, the current energy consumption and carbon emissions forecasts for each non-benchmark enterprise are differentially calibrated. This invention improves the accuracy of enterprise energy consumption and carbon emissions accounting and reduces the economic cost of accounting.

Description

Technical Field

[0001] The present invention belongs to the field of electric power technology, and specifically relates to a method and system for calculating enterprise energy consumption and carbon emissions based on differentiated calibration. Background Art

[0002] Key energy-consuming enterprises in sectors such as industry and construction are an important component of society's carbon emissions structure. Their production processes consume large amounts of energy and produce large amounts of greenhouse gases. Timely and accurate accounting of corporate energy consumption and carbon emissions will help steadily advance society's green and low-carbon transformation. Currently, energy consumption and carbon emissions accounting at the enterprise level is subject to significant time lags. Enterprises often need to use complex energy consumption statistics processes to calculate actual energy consumption and carbon emissions, making it impossible for relevant departments to obtain timely and accurate monthly energy consumption data from enterprises, impacting policy adjustments, indicator assessments, and dynamic supervision. Installing an energy consumption monitoring system in every enterprise would increase related fiscal expenditures and business operating costs, making implementation difficult.

[0003] Chinese Patent: Invention application number 202411317701.8 discloses an online energy consumption detection device and detection system. The proposed online energy consumption detection system includes a master station management layer, a network communication layer, and an on-site measurement and control layer. It can realize functions such as data detection, enterprise energy consumption analysis, and data management, and can effectively realize dynamic monitoring of enterprise energy consumption data. However, if all enterprises install detection systems and related equipment, it will increase the economic costs and operational pressure of enterprises. Moreover, most data needs to be manually entered, which increases labor costs. Therefore, the large-scale promotion and application of energy consumption detection systems faces practical difficulties, especially for small and medium-sized enterprises. Summary of the Invention

[0004] The purpose of the present invention is to address the above-mentioned problems existing in the prior art and to provide a method and system for calculating enterprise energy consumption and carbon emissions based on differentiated calibration, which can improve the accuracy and timeliness of enterprise energy consumption and carbon emissions accounting.

[0005] To achieve the above objectives, the technical solutions of the present invention are as follows:

[0006] In a first aspect, the present invention proposes a method for calculating enterprise energy consumption based on differentiated calibration, comprising:

[0007] A1. Collect historical production and operation information of enterprises in the same industry within the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output;

[0008] A2. Conduct cluster analysis on all enterprises in the same industry within the region based on production and operation information to obtain multiple enterprise clusters. Select the enterprise at the center of each cluster as the benchmark enterprise.

[0009] A3. For benchmark enterprises, their current energy consumption values ​​are obtained through their installed energy consumption monitoring systems. For all enterprises in each enterprise cluster, their current energy consumption forecast values ​​are obtained based on historical data.

[0010] A4. Calculate the deviation rate between the actual measured energy consumption of each benchmark enterprise and the current energy consumption forecast value. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current energy consumption forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding energy consumption accounting result.

[0011] In the A4, the deviation rate The calculation formula is:

[0012] ;

[0013] In the above formula, 、 are the measured value and predicted value of current energy consumption respectively;

[0014] The current energy consumption forecast values ​​of each non-benchmark enterprise are calibrated differently according to the following formula:

[0015] ;

[0016] ;

[0017] In the above formula, is the calibrated current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the energy consumption calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

[0018] A2 uses the weighted K-Means++ clustering algorithm for cluster analysis. The weighted K-Means++ clustering algorithm uses electricity consumption, energy consumption, carbon emissions, production output value, and product output as clustering features and uses the following distance calculation formula:

[0019] ;

[0020] In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

[0021] In A3, the current energy consumption forecast values ​​of each enterprise obtained based on historical data include:

[0022] A31. Based on the historical electricity consumption and energy consumption data of each enterprise, the historical electricity factor of each enterprise is calculated using the following formula:

[0023] ;

[0024] In the above formula, is the electricity factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively;

[0025] A32. Based on the historical electricity factors of each enterprise, the ARIMA model is used to predict the current electricity factors of each enterprise;

[0026] A33. Calculate the current energy consumption forecast for each enterprise using the current electricity factor:

[0027] ;

[0028] In the above formula, is the current energy consumption forecast value, is the current electric energy factor, The current electricity consumption.

[0029] In a second aspect, the present invention proposes a method for calculating corporate carbon emissions based on differentiated calibration, comprising:

[0030] B1. Collect historical production and operation information of enterprises in the same industry within the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output;

[0031] B2. Conduct cluster analysis on all enterprises in the same industry within the region based on production and operation information to obtain multiple enterprise clusters. Select the enterprise at the center of each cluster as the benchmark enterprise.

[0032] B3. For benchmark enterprises, obtain their current measured energy consumption values ​​through their installed energy consumption monitoring systems, and calculate their current measured carbon emissions values ​​based on these values. For all enterprises in each enterprise cluster, obtain their current predicted carbon emissions values ​​based on historical data.

[0033] B4. Calculate the deviation rate between the actual carbon emissions of each benchmark enterprise and the current carbon emissions forecast value. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster , considering the clustering space distance, the current carbon emission forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current carbon emission forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding carbon emission accounting result.

[0034] In B4, the deviation rate The calculation formula is:

[0035] ;

[0036] In the above formula, 、 are the measured and predicted values ​​of carbon emissions for the current period, respectively;

[0037] The current carbon emissions forecast values ​​of each non-benchmark enterprise are calibrated differently according to the following formula:

[0038] ;

[0039] ;

[0040] In the above formula, is the calibrated predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of the current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the carbon emission calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

[0041] B2 uses the weighted K-Means++ clustering algorithm to perform cluster analysis. The weighted K-Means++ clustering algorithm uses electricity consumption, energy consumption, carbon emissions, production output value, and product output as clustering features and uses the following distance calculation formula:

[0042] ;

[0043] In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

[0044] In B3, the current carbon emissions forecast values ​​of each enterprise obtained based on historical data include:

[0045] B31. Based on the historical electricity consumption and carbon emissions data of each enterprise, the following formula is used to calculate the historical electricity carbon factor of each enterprise:

[0046] ;

[0047] In the above formula, is the electricity carbon factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively;

[0048] B32. Based on the historical electricity carbon factor of each enterprise, the ARIMA model is used to predict the current electricity carbon factor of each enterprise;

[0049] B33. Calculate the current carbon emissions forecast for each enterprise using the current electricity carbon factor:

[0050] ;

[0051] In the above formula, is the predicted value of carbon emissions for the current period, is the current electric carbon factor, The current electricity consumption.

[0052] In a third aspect, the present invention proposes an enterprise energy consumption accounting system based on differentiated calibration, comprising an information collection module, a cluster analysis module, an energy consumption monitoring module, an energy consumption prediction module, a deviation rate calculation module, and a differentiated calibration module;

[0053] The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output;

[0054] The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise located at the cluster center in each enterprise cluster as the benchmark enterprise;

[0055] The energy consumption monitoring module is used to obtain the actual value of the current energy consumption of the benchmark enterprise through the energy consumption monitoring system installed;

[0056] The energy consumption prediction module is used to obtain the current energy consumption prediction value of all enterprises in each enterprise cluster based on historical data;

[0057] The deviation rate calculation module is used to calculate the deviation rate between the current energy consumption measured value and the current energy consumption forecast value of each benchmark enterprise. ;

[0058] The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current energy consumption forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding energy consumption accounting result.

[0059] The energy consumption prediction module includes a historical electric energy factor calculation unit, a current electric energy factor prediction unit, and a current energy consumption prediction unit;

[0060] The historical power factor calculation unit is used to calculate the historical power factor of each enterprise based on the historical power consumption and energy consumption data of each enterprise using the following formula:

[0061] ;

[0062] In the above formula, is the electricity factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively;

[0063] The current power factor prediction unit is used to predict the current power factor of each enterprise based on the historical power factor of each enterprise using the autoregressive difference moving average (ARIMA) model;

[0064] The current energy consumption prediction unit is used to calculate the current energy consumption prediction value of each enterprise through the current electric energy factor:

[0065] ;

[0066] In the above formula, is the current energy consumption forecast value, is the current electric energy factor, The current electricity consumption.

[0067] The deviation rate calculation module is used to calculate according to the following formula :

[0068] ;

[0069] In the above formula, 、 are the measured value and predicted value of current energy consumption respectively.

[0070] The differential calibration module is used to perform differential calibration according to the following formula:

[0071] ;

[0072] ;

[0073] In the above formula, is the calibrated current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the energy consumption calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

[0074] In a fourth aspect, the present invention proposes an enterprise carbon emission accounting system based on differential calibration, comprising an information collection module, a cluster analysis module, a carbon emission monitoring module, a carbon emission prediction module, a deviation rate calculation module, and a differential calibration module;

[0075] The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output;

[0076] The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise located at the cluster center in each enterprise cluster as the benchmark enterprise;

[0077] The carbon emission monitoring module is used to obtain the actual value of the current energy consumption of the benchmark enterprise through the energy consumption monitoring system installed by the benchmark enterprise, and calculate the actual value of the current carbon emission based on the actual value of the current energy consumption;

[0078] The carbon emission prediction module is used to obtain the current carbon emission prediction value of each enterprise in each enterprise cluster based on historical data;

[0079] The deviation rate calculation module is used to calculate the deviation rate between the actual value of the current carbon emissions of each benchmark enterprise and the current carbon emissions forecast value. ;

[0080] The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the clustering space distance, the current carbon emission forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current carbon emission forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding carbon emission accounting result.

[0081] The carbon emission prediction module includes a historical electricity-carbon factor calculation unit, a current electricity-carbon factor prediction unit, and a current carbon emission prediction unit;

[0082] The historical electricity carbon factor calculation unit calculates the historical electricity carbon factor of each enterprise based on the historical electricity consumption and carbon emission data of each enterprise using the following formula:

[0083] ;

[0084] In the above formula, is the electricity carbon factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively;

[0085] The current electricity carbon factor prediction unit is used to predict the current electricity carbon factor of each enterprise based on the historical electricity carbon factor of each enterprise using the autoregressive difference moving average ARIMA model;

[0086] The current carbon emission prediction unit is used to calculate the current carbon emission prediction value of each enterprise through the current electricity carbon factor:

[0087] ;

[0088] In the above formula, is the predicted value of carbon emissions for the current period, is the current electric carbon factor, The current electricity consumption.

[0089] The deviation rate calculation module is used to calculate according to the following formula :

[0090] ;

[0091] In the above formula, 、 are the measured value and predicted value of carbon emissions for the current period, respectively.

[0092] The differential calibration module is used to perform differential calibration according to the following formula:

[0093] ;

[0094] ;

[0095] In the above formula, is the calibrated predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of the current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the carbon emission calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

[0096] Compared with the prior art, the present invention has the following beneficial effects:

[0097] 1. A method for calculating enterprise energy consumption and carbon emissions based on differentiated calibration in the present invention first collects historical production and operation information of enterprises in the same industry in a region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output. Then, cluster analysis is performed on all enterprises in the same industry in the region based on the production and operation information to obtain multiple enterprise clusters, and the enterprise located at the cluster center in each enterprise cluster is selected as the benchmark enterprise. Then, the current energy consumption measured value (the current carbon emission measured value is calculated based on the current energy consumption measured value) is obtained through the energy consumption monitoring system installed by the benchmark enterprise. Based on the historical data, the current energy consumption / carbon emission predicted value of all enterprises in each enterprise cluster is obtained. Subsequently, the deviation rate between the measured value and the predicted value of each benchmark enterprise is calculated. According to the deviation rate of the benchmark enterprise and taking into account the cluster space distance, the current energy consumption / carbon emission predicted value of each non-benchmark enterprise in the corresponding enterprise cluster is differentially calibrated. The current energy consumption / carbon emission predicted value of each non-benchmark enterprise after differential calibration is used as the corresponding energy consumption / carbon emission accounting result. On the one hand, this method conducts cluster analysis on all enterprises in the same industry in the region, takes the enterprise at the cluster center as the benchmark enterprise, and only installs energy consumption monitoring systems in the benchmark enterprises, thereby reducing the economic cost of energy consumption and carbon emission accounting; on the other hand, when carrying out enterprise energy consumption and carbon emission accounting, the measured energy consumption and carbon emission data of the benchmark enterprises are used as a benchmark, and the clustering spatial distance between non-benchmark enterprises and benchmark enterprises is considered. The estimated results of non-benchmark enterprises that have not installed energy consumption monitoring systems are differentially calibrated to avoid the result deviation caused by unified calibration, thereby maximizing the accuracy of enterprise energy consumption and carbon emission accounting.

[0098] 2. The present invention provides an enterprise energy consumption and carbon emission accounting method based on differentiated calibration. It utilizes the high accuracy and strong real-time performance of electricity data, and realizes the characterization of energy consumption and carbon emissions by electricity consumption by constructing electric energy factors and electric carbon factors. In combination with historical trends, the dynamic calculation of enterprise energy consumption and carbon emissions is completed through prediction algorithms, which solves the problem of lagging energy consumption and carbon emission accounting caused by slow statistics of enterprise energy consumption data and complex processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 This is a flowchart of the method described in Example 1.

[0100] Figure 2 This is a flowchart of the method described in Example 2.

[0101] Figure 3 This is a structural diagram of the system described in Example 3.

[0102] Figure 4 This is a structural diagram of the system described in Example 4. DETAILED DESCRIPTION

[0103] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0104] Example 1:

[0105] This embodiment takes all enterprises in a certain industry in a certain region as the research object, and implements the enterprise energy consumption accounting method based on differentiated calibration described in the present invention, such as Figure 1 The specific steps are as follows:

[0106] 1. Collect all historical production and operation information of all enterprises, including electricity consumption, energy consumption, carbon emissions, production output value, and product output. Among them, energy consumption is obtained by converting the consumption of various energy sources into standard coal and summing them up:

[0107] ;

[0108] In the above formula, is the energy consumption of the enterprise in the tth period of history, and the time scale can be years or months; is the consumption of the kth energy source, out of a total of n energy sources, including electricity; is the standard coal conversion coefficient of the kth energy source, including electricity.

[0109] Carbon emissions are calculated from the consumption of various energy sources using the emission factor method:

[0110] ;

[0111] In the above formula, is the carbon emissions of the enterprise in the tth period of history; is the consumption of the kth energy source, totaling n-1 energy sources, excluding electricity; is the average lower calorific value of the kth energy source, excluding electricity; is the CO2 emission factor of the kth energy source, excluding electricity; is the electricity consumption of the enterprise in period t, and the time scale can be years or months; For the average carbon dioxide emission factor of electricity, the latest data released by the Ministry of Ecology and Environment is selected.

[0112] 2. Using the five types of production and operation information collected as clustering features, the weighted K-Means++ clustering algorithm is used to perform cluster analysis on all enterprises, so that enterprises with similar scale, production level, energy consumption level and carbon emission level are clustered together, and multiple enterprise clusters are obtained.

[0113] The distance calculation formula of the weighted K-Means++ clustering algorithm is:

[0114] ;

[0115] In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, , 、 、 、 、 are the weights of electricity consumption, energy consumption, carbon emissions, production output value, and product output, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

[0116] This embodiment considers the influence of different enterprise production and operation indicators, takes electricity consumption, energy consumption, and carbon emissions as the main classification conditions, and takes production output value and product output as secondary classification conditions, reflecting the influence of different characteristics on clustering results, and realizing cluster analysis that is more in line with the energy consumption and carbon emission accounting needs of enterprises. The weight of each cluster feature is determined by principal component analysis, and 、 、 ≥ , 、 、 ≥ .

[0117] For the cluster center, in an enterprise cluster, the sum of the spatial distances from a certain enterprise to all other enterprises is the smallest compared to other enterprises. This enterprise is considered to be the cluster center of the enterprise cluster. The calculation formula is:

[0118] ;

[0119] In the above formula, It is the sum of the spatial distances from the i-th enterprise to all other enterprises. Its value is the smallest and can be considered as the cluster center and the benchmark enterprise.

[0120] 3. For each enterprise cluster, the enterprise located at the cluster center is selected as the benchmark enterprise. By installing an energy consumption monitoring system, the energy consumption data of the benchmark enterprise is collected in real time. Other enterprises are non-benchmark enterprises and do not install a monitoring system.

[0121] The enterprise energy consumption monitoring system can realize the collection of energy consumption data for the entire production process, covering all production processes such as raw material acquisition and product production. The collected energy consumption data includes major energy sources such as coal, oil, and gas. Coal can be subdivided into types such as bituminous coal and coking coal, oil can be subdivided into types such as crude oil, fuel oil, and gasoline, and gas can be subdivided into types such as natural gas and liquefied petroleum gas.

[0122] 4. Based on the historical electricity consumption and energy consumption data of each enterprise, the historical electricity factor of each enterprise is calculated using the following formula:

[0123] ;

[0124] In the above formula, is the electricity factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t respectively.

[0125] This embodiment considers the strong correlation between electricity consumption and energy consumption, and constructs an electric energy factor to characterize the total energy consumption through electricity consumption, that is, the unit energy consumption corresponding to the unit electricity consumption.

[0126] 5. Considering the progress of science and technology and the improvement of enterprise production level, the corresponding relationship between electricity consumption and energy consumption is changing dynamically. Based on the historical electricity factors of each enterprise, the autoregressive differential moving average (ARIMA) model is used to predict the current electricity factors of each enterprise, including:

[0127] The ARIMA model consists of three parts: autoregression (AR), differencing (I), and moving average (MA). It is used to analyze and predict time series data. The AR part captures the impact of historical observations on current values. The I part stabilizes the series through differencing, eliminating trends and seasonality. The MA part considers the impact of historical forecast errors on current values, which meets the requirements of electric energy factors as time series. The calculation formula is as follows:

[0128] ;

[0129] In the above formula, is the time series data of the electric energy factor, arrive is the AR model parameter, which describes the relationship between the current power factor and the power factor at the past p time points; arrive is the MA model parameter, which describes the relationship between the current power factor and the error of the power factor at the past q time points; is the error term at time t; is a constant term.

[0130] 6. Calculate the current energy consumption forecast for each enterprise using the current electricity factor:

[0131] ;

[0132] In the above formula, is the current energy consumption forecast value, is the current electric energy factor, The current electricity consumption.

[0133] 7. Calculate the deviation rate between the actual measured value of current energy consumption of each benchmark enterprise and the predicted value of current energy consumption collected by the energy consumption monitoring system :

[0134] ;

[0135] In the above formula, 、 are the measured value and predicted value of current energy consumption respectively.

[0136] 8. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the cluster , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the cluster is calibrated differently. Among them, the energy consumption calibration rate calculation formula of non-benchmark enterprises is:

[0137] ;

[0138] In the above formula, is the energy consumption calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, is the kth cluster eigenvalue of the i-th non-benchmark enterprise, 、 are the kth cluster eigenvalues ​​of the benchmark enterprise and the non-benchmark enterprise farthest from the benchmark enterprise, As the basic ratio, it is assumed that the most remote enterprise also needs to be calibrated according to the deviation rate of the benchmark enterprise, which is set to 0.5.

[0139] When performing differentiated calibration, this embodiment takes into account the cluster space distance between the calibrated enterprise and the benchmark enterprise, and the cluster space distance between the non-benchmark enterprise at the farthest end of the enterprise cluster and the benchmark enterprise, and uses the ratio of the two as the judgment basis for differentiated calibration. That is, the closer the enterprise is to the benchmark enterprise (cluster center), the greater the influence of the actual measured data of the benchmark enterprise, and the greater the proportion of calibration according to the benchmark enterprise deviation rate. Conversely, the farther the enterprise is from the benchmark enterprise (cluster center), the less influence of the actual measured data of the benchmark enterprise, and the smaller the proportion of calibration according to the benchmark enterprise deviation rate.

[0140] Non-benchmark enterprises shall be calibrated according to the above calibration rate, and the calculation formula is:

[0141] ;

[0142] In the above formula, is the calibrated current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of current energy consumption of the i-th non-benchmark enterprise in the enterprise cluster.

[0143] The calibrated current energy consumption forecast value of each non-benchmark enterprise is the corresponding energy consumption accounting result.

[0144] Example 2:

[0145] This example uses the same research object as Example 1, such as Figure 1 As shown in the figure, the enterprise carbon emission accounting method based on differentiated calibration is implemented. The specific steps are as follows:

[0146] 1. Collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output. The details are the same as step 1 in Example 1.

[0147] 2. Perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise at the cluster center in each enterprise cluster as the benchmark enterprise. The specific content is the same as step 2 in Example 1.

[0148] 3. For the benchmark enterprise, obtain the actual value of its current energy consumption through the energy consumption monitoring system installed by the benchmark enterprise. The specific content is the same as step 3 in embodiment 1.

[0149] 4. Based on the actual measured value of current energy consumption, the carbon emission factor method is used to calculate the actual measured value of current carbon emissions.

[0150] 5. Based on the historical electricity consumption and carbon emissions data of each enterprise, the historical electricity-carbon factor of each enterprise is calculated using the following formula:

[0151] ;

[0152] In the above formula, is the electricity carbon factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t respectively.

[0153] This embodiment considers the strong correlation between an enterprise's electricity consumption and carbon emissions, and constructs an electricity-carbon factor to characterize the enterprise's unit carbon emissions through unit electricity consumption.

[0154] 6. Based on the historical electricity carbon factor of each enterprise, the autoregressive differential moving average (ARIMA) model is used to predict the current electricity carbon factor of each enterprise. The specific content is the same as step 5 in Example 1.

[0155] Calculate the current carbon emissions forecast for each enterprise using the current electricity carbon factor:

[0156] ;

[0157] In the above formula, is the predicted value of carbon emissions for the current period, is the current electric carbon factor, The current electricity consumption.

[0158] 7. Calculate the deviation rate between the actual measured carbon emissions of each benchmark enterprise and the current carbon emissions forecast value :

[0159] ;

[0160] In the above formula, 、 are the measured value and predicted value of carbon emissions for the current period, respectively.

[0161] 8. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster , considering the cluster space distance, the current carbon emission forecast value of each non-benchmark enterprise in the enterprise cluster is calibrated differently.

[0162] The calculation formula for the carbon emission calibration rate of non-benchmark enterprises is:

[0163] ;

[0164] In the above formula, is the carbon emission calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the base ratio, take 0.5.

[0165] 9. Non-benchmark enterprises shall calibrate according to the above calibration rate. The calculation formula is:

[0166] ;

[0167] In the above formula, is the calibrated predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster.

[0168] The current carbon emission forecast value after differentiated calibration of each non-benchmark enterprise is the corresponding carbon emission accounting result.

[0169] Example 3:

[0170] An enterprise energy consumption accounting system based on differentiated calibration, such as Figure 3 As shown, it includes information collection module, cluster analysis module, energy consumption monitoring module, energy consumption prediction module, deviation rate calculation module, and differential calibration module.

[0171] The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output.

[0172] The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise at the cluster center in each enterprise cluster as the benchmark enterprise. The cluster analysis includes:

[0173] The five types of production and operation information collected were used as clustering features, and the weighted K-Means++ clustering algorithm was used to perform cluster analysis on all enterprises, and multiple enterprise clusters were obtained.

[0174] The distance calculation formula of the weighted K-Means++ clustering algorithm is:

[0175] ;

[0176] In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, , 、 、 、 、 are the weights of electricity consumption, energy consumption, carbon emissions, production output value, and product output, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

[0177] Considering the influence of production and operation indicators of different enterprises, electricity consumption, energy consumption and carbon emissions are used as the main classification conditions, and production output value and product output are used as secondary classification conditions. The weight of each cluster feature is determined by principal component analysis. 、 、 ≥ , 、 、 ≥ .

[0178] The energy consumption monitoring module is used to obtain the actual measured value of the current energy consumption of the benchmark enterprise through the energy consumption monitoring system installed by the benchmark enterprise.

[0179] The energy consumption prediction module is used to obtain the current energy consumption prediction value of all enterprises in each enterprise cluster based on historical data, including a historical power factor calculation unit, a current power factor prediction unit, and a current energy consumption prediction unit.

[0180] The historical power factor calculation unit is used to calculate the historical power factor of each enterprise based on the historical power consumption and energy consumption data of each enterprise using the following formula:

[0181] ;

[0182] In the above formula, is the electricity factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t respectively.

[0183] The current electric energy factor prediction unit is used to predict the current electric energy factor of each enterprise based on the historical electric energy factor of each enterprise using the autoregressive difference moving average (ARIMA) model.

[0184] The current energy consumption prediction unit is used to calculate the current energy consumption prediction value of each enterprise through the current electric energy factor:

[0185] ;

[0186] In the above formula, is the current energy consumption forecast value, is the current electric energy factor, The current electricity consumption.

[0187] The deviation rate calculation module is used to calculate the deviation rate between the current energy consumption measured value and the current energy consumption forecast value of each benchmark enterprise. :

[0188] ;

[0189] In the above formula, 、 are the measured value and predicted value of current energy consumption respectively.

[0190] The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the enterprise cluster is calibrated as follows:

[0191] ;

[0192] ;

[0193] In the above formula, is the calibrated current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the energy consumption calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the base ratio, take 0.5.

[0194] The current energy consumption forecast value after differentiated calibration of each non-benchmark enterprise is the corresponding energy consumption accounting result.

[0195] Example 4:

[0196] A corporate carbon emissions accounting system based on differentiated calibration, such as Figure 4 As shown, it includes information collection module, cluster analysis module, carbon emission monitoring module, carbon emission prediction module, deviation rate calculation module, and differential calibration module.

[0197] The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output.

[0198] The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise at the cluster center in each enterprise cluster as the benchmark enterprise. The cluster analysis includes:

[0199] The five types of production and operation information collected were used as clustering features, and the weighted K-Means++ clustering algorithm was used to perform cluster analysis on all enterprises, and multiple enterprise clusters were obtained.

[0200] The distance calculation formula of the weighted K-Means++ clustering algorithm is:

[0201] ;

[0202] In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, , 、 、 、 、 are the weights of electricity consumption, energy consumption, carbon emissions, production output value, and product output, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

[0203] Considering the influence of production and operation indicators of different enterprises, electricity consumption, energy consumption and carbon emissions are used as the main classification conditions, and production output value and product output are used as secondary classification conditions. The weight of each cluster feature is determined by principal component analysis. 、 、 ≥ , 、 、 ≥ .

[0204] The carbon emission monitoring module is used to obtain the current energy consumption actual value through the energy consumption monitoring system installed by the benchmark enterprise, and calculate the current carbon emission actual value based on the current energy consumption actual value.

[0205] The carbon emission prediction module is used to obtain the current carbon emission prediction value of each enterprise in each enterprise cluster based on historical data, including a historical electricity-carbon factor calculation unit, a current electricity-carbon factor prediction unit, and a current carbon emission prediction unit.

[0206] The historical electricity carbon factor calculation unit calculates the historical electricity carbon factor of each enterprise based on the historical electricity consumption and carbon emission data of each enterprise using the following formula:

[0207] ;

[0208] In the above formula, is the electricity carbon factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t respectively.

[0209] The current electricity-carbon factor prediction unit is used to predict the current electricity-carbon factor of each enterprise based on the historical electricity-carbon factor of each enterprise using the autoregressive difference moving average ARIMA model.

[0210] The current carbon emission prediction unit is used to calculate the current carbon emission prediction value of each enterprise through the current electricity carbon factor:

[0211] ;

[0212] In the above formula, is the predicted value of carbon emissions for the current period, is the current electric carbon factor, The current electricity consumption.

[0213] The deviation rate calculation module is used to calculate the deviation rate between the actual value of the current carbon emissions of each benchmark enterprise and the current carbon emissions forecast value. :

[0214] ;

[0215] In the above formula, 、 are the measured value and predicted value of carbon emissions for the current period, respectively.

[0216] The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the cluster space distance, the current carbon emission forecast values ​​of each non-benchmark enterprise in the enterprise cluster are calibrated as follows:

[0217] ;

[0218] ;

[0219] In the above formula, is the calibrated predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of the current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the carbon emission calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the base ratio, take 0.5.

[0220] The current carbon emission forecast value after differentiated calibration of each non-benchmark enterprise is the corresponding carbon emission accounting result.

Claims

1. A method for calculating enterprise energy consumption based on differentiated calibration, characterized in that: The method comprises: A1. Collect historical production and operation information of enterprises in the same industry within the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output; A2. Conduct cluster analysis on all enterprises in the same industry within the region based on production and operation information to obtain multiple enterprise clusters. Select the enterprise at the center of each cluster as the benchmark enterprise. A3. For benchmark enterprises, their current energy consumption values ​​are obtained through their installed energy consumption monitoring systems. For all enterprises in each enterprise cluster, their current energy consumption forecast values ​​are obtained based on historical data. A4. Calculate the deviation rate between the actual measured energy consumption of each benchmark enterprise and the current energy consumption forecast value. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current energy consumption forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding energy consumption accounting result.

2. The enterprise energy consumption accounting method based on differentiated calibration according to claim 1 is characterized in that: In the A4, the deviation rate The calculation formula is: ； In the above formula, 、 are the measured value and predicted value of current energy consumption respectively; The current energy consumption forecast values ​​of each non-benchmark enterprise are calibrated differently according to the following formula: ； ； In the above formula, is the calibrated current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the current energy consumption forecast value of the i-th non-benchmark enterprise in the enterprise cluster, is the energy consumption calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

3. The enterprise energy consumption accounting method based on differentiated calibration according to claim 1 or 2, characterized in that: A2 uses the weighted K-Means++ clustering algorithm for cluster analysis. The weighted K-Means++ clustering algorithm uses electricity consumption, energy consumption, carbon emissions, production output value, and product output as clustering features and uses the following distance calculation formula: ； In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

4. The enterprise energy consumption accounting method based on differentiated calibration according to claim 1 or 2, characterized in that: In A3, the current energy consumption forecast values ​​of each enterprise obtained based on historical data include: A31. Based on the historical electricity consumption and energy consumption data of each enterprise, the historical electricity factor of each enterprise is calculated using the following formula: ； In the above formula, is the electricity factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively; A32. Based on the historical electricity factors of each enterprise, the ARIMA model is used to predict the current electricity factors of each enterprise; A33. Calculate the current energy consumption forecast for each enterprise using the current electricity factor: ； In the above formula, is the current energy consumption forecast value, is the current electric energy factor, The current electricity consumption.

5. A method for calculating corporate carbon emissions based on differentiated calibration, characterized in that: The method comprises: B1. Collect historical production and operation information of enterprises in the same industry within the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output; B2. Conduct cluster analysis on all enterprises in the same industry within the region based on production and operation information to obtain multiple enterprise clusters. Select the enterprise at the center of each cluster as the benchmark enterprise. B3. For benchmark enterprises, obtain their current measured energy consumption values ​​through their installed energy consumption monitoring systems, and calculate their current measured carbon emissions values ​​based on these values. For all enterprises in each enterprise cluster, obtain their current predicted carbon emissions values ​​based on historical data. B4. Calculate the deviation rate between the actual carbon emissions of each benchmark enterprise and the current carbon emissions forecast value. For each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster , considering the clustering space distance, the current carbon emission forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current carbon emission forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding carbon emission accounting result.

6. The enterprise carbon emission accounting method based on differentiated calibration according to claim 5 is characterized in that: In B4, the deviation rate The calculation formula is: ； In the above formula, 、 are the measured and predicted values ​​of carbon emissions for the current period, respectively; The current carbon emissions forecast values ​​of each non-benchmark enterprise are calibrated differently according to the following formula: ； ； In the above formula, is the calibrated predicted value of current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the predicted value of the current carbon emissions of the i-th non-benchmark enterprise in the enterprise cluster, is the carbon emission calibration rate of the i-th non-benchmark enterprise in the enterprise cluster, is the clustering space distance from the i-th non-benchmark enterprise to the benchmark enterprise in the enterprise cluster, is the clustering space distance between the non-benchmark enterprise farthest from the benchmark enterprise in the enterprise cluster and the benchmark enterprise, As the basic ratio.

7. The enterprise carbon emission accounting method based on differentiated calibration according to claim 5 or 6, characterized in that: B2 uses the weighted K-Means++ clustering algorithm to perform cluster analysis. The weighted K-Means++ clustering algorithm uses electricity consumption, energy consumption, carbon emissions, production output value, and product output as clustering features and uses the following distance calculation formula: ； In the above formula, is the distance between the i-th enterprise and the j-th enterprise, is the weight of the k-th cluster feature, is the kth cluster eigenvalue of the i-th enterprise, is the kth cluster eigenvalue of the jth enterprise.

8. The enterprise carbon emission accounting method based on differentiated calibration according to claim 5 or 6, characterized in that: In B3, the current carbon emissions forecast values ​​of each enterprise obtained based on historical data include: B31. Based on the historical electricity consumption and carbon emissions data of each enterprise, the following formula is used to calculate the historical electricity carbon factor of each enterprise: ； In the above formula, is the electricity carbon factor of the enterprise in period t, 、 are the energy consumption and electricity consumption of the enterprise in period t, respectively; B32. Based on the historical electricity carbon factor of each enterprise, the ARIMA model is used to predict the current electricity carbon factor of each enterprise; B33. Calculate the current carbon emissions forecast for each enterprise using the current electricity carbon factor: ； In the above formula, is the predicted value of carbon emissions for the current period, is the current electric carbon factor, The current electricity consumption.

9. An enterprise energy consumption accounting system based on differentiated calibration, characterized in that: The system includes an information acquisition module, a cluster analysis module, an energy consumption monitoring module, an energy consumption prediction module, a deviation rate calculation module, and a differential calibration module; The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output; The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise located at the cluster center in each enterprise cluster as the benchmark enterprise; The energy consumption monitoring module is used to obtain the actual value of the current energy consumption of the benchmark enterprise through the energy consumption monitoring system installed; The energy consumption prediction module is used to obtain the current energy consumption prediction value of all enterprises in each enterprise cluster based on historical data; The deviation rate calculation module is used to calculate the deviation rate between the current energy consumption measured value and the current energy consumption forecast value of each benchmark enterprise. ; The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the cluster space distance, the current energy consumption forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current energy consumption forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding energy consumption accounting result.

10. An enterprise carbon emission accounting system based on differentiated calibration, characterized in that: The system includes an information collection module, a cluster analysis module, a carbon emission monitoring module, a carbon emission prediction module, a deviation rate calculation module, and a differential calibration module; The information collection module is used to collect historical production and operation information of enterprises in the same industry in the region, including electricity consumption, energy consumption, carbon emissions, production output value, and product output; The cluster analysis module is used to perform cluster analysis on all enterprises in the same industry in the region based on production and operation information to obtain multiple enterprise clusters, and select the enterprise located at the cluster center in each enterprise cluster as the benchmark enterprise; The carbon emission monitoring module is used to obtain the actual value of the current energy consumption of the benchmark enterprise through the energy consumption monitoring system installed by the benchmark enterprise, and calculate the actual value of the current carbon emission based on the actual value of the current energy consumption; The carbon emission prediction module is used to obtain the current carbon emission prediction value of each enterprise in each enterprise cluster based on historical data; The deviation rate calculation module is used to calculate the deviation rate between the actual value of the current carbon emissions of each benchmark enterprise and the current carbon emissions forecast value. ; The differentiation calibration module is used for each enterprise cluster, according to the deviation rate of the benchmark enterprise in the enterprise cluster. , considering the clustering space distance, the current carbon emission forecast value of each non-benchmark enterprise in the enterprise cluster is differentially calibrated, and the current carbon emission forecast value of each non-benchmark enterprise after differential calibration is used as the corresponding carbon emission accounting result.

Images