Method and system for monitoring carbon footprint of traditional Chinese medicine pharmaceutical enterprise

Abstract

The invention belongs to the technical field of carbon footprint intelligent monitoring, and provides a traditional Chinese medicine pharmacy enterprise carbon footprint monitoring method and system. The Chinese patent medicine historical data and the Chinese patent medicine monthly data are acquired through the data acquisition robot module, the acquired data are preprocessed through the data management robot module, and Chinese patent medicine standard carbon footprint accounting and monthly whole plant carbon emission accounting are carried out. The data analysis robot module identifies a key carbon emission link, predicts a monthly carbon emission subitem value, monitors an abnormal carbon emission subitem, performs carbon footprint abnormality diagnosis, and determines the carbon footprint abnormality. The collected data, the key carbon emission link, the monthly carbon emission sub-item prediction value, the abnormal carbon emission sub-item and the carbon footprint abnormity diagnosis result are subjected to visual interaction through the man-machine interaction robot module; according to the invention, Chinese patent medicine pharmaceutical enterprises can pay attention to abnormal carbon footprints and abnormal carbon emission events in time, so that low-carbon operation of the enterprises is facilitated, and the enterprises are helped to manage carbon emission of the enterprises.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon footprint intelligent monitoring, and in particular relates to a carbon footprint monitoring method and system for a traditional Chinese medicine pharmaceutical enterprise. Background Art

[0002] In recent years, the Chinese medicine pharmaceutical industry, as an important branch of the manufacturing industry, has played a pivotal role in my country's national economic and social development. However, this traditional characteristic industry is also facing the challenges of high energy consumption and large emissions. Its production process covers multiple links such as the planting of Chinese medicinal materials, the preparation of medicinal pieces, the production of preparations, and the circulation and sales of products. Especially in the industrial production process of Chinese medicine, the consumption of electricity, heat energy and organic solvents is significant, and a large amount of wastewater, waste residue and high carbon emissions are also generated. The traditional carbon footprint accounting method mainly relies on manual data collection and analysis, which is not only time-consuming and labor-intensive, but also prone to data omissions and errors, and it is difficult to meet the needs of modern enterprises for efficient and accurate carbon emission management.

[0003] At present, there have been some studies and applications that attempt to improve carbon emission monitoring through information technology. For example, patent document CN114358668A proposes a carbon footprint big data analysis method and system based on industrial Internet identification. The system can obtain the industrial Internet identification of the product, parse the target identification to obtain the target storage address, and then obtain product information, calculate the carbon footprint of the product, and generate a carbon footprint report. Patent document CN117151960A proposes a big data-driven carbon footprint assessment system, which includes a product cycle management module, a data collection module, an analysis and evaluation module, and a result display module. It can record the carbon emission stage of the entire life cycle of the product, collect carbon emission data, and conduct an assessment.

[0004] Although these patents propose different technical solutions to monitor and calculate carbon footprint, there are still some shortcomings: (1) Lack of comprehensiveness and accuracy in data collection, unable to fully cover all relevant carbon emission sources, or the accuracy of data collection needs to be improved; (2) Lack of real-time performance, unable to meet the demand for real-time data in the production process; (3) Insufficient system integration and compatibility capabilities. Summary of the invention

[0005] The present invention aims to address the problems existing in the prior art and provides a carbon footprint monitoring method and system for traditional Chinese medicine pharmaceutical companies that can comprehensively reflect the carbon emissions of the company, enable traditional Chinese medicine pharmaceutical companies to pay attention to abnormal carbon footprints and abnormal carbon emission events in a timely manner, help companies maintain low-carbon operations in the long term, reduce carbon emissions, and help companies better manage their carbon emissions.

[0006] In order to achieve the above technical objectives, the present invention adopts the following technical solutions: A carbon footprint monitoring method for a traditional Chinese medicine pharmaceutical enterprise comprises the following steps: S1. Collection of historical and monthly data of Chinese patent medicines; S2. Preprocess the collected historical data and monthly data of Chinese patent medicines; S3. Obtain the pre-processed historical data of Chinese patent medicines, perform standard carbon footprint accounting of Chinese patent medicines, and perform monthly carbon emissions accounting for the entire factory; S4. Obtain the pre-processed monthly data of Chinese patent medicines, calculate the impact of improvements in different links of Chinese patent medicines on overall carbon emissions based on the standard carbon footprint of Chinese patent medicines, sort different links according to the impact, and identify links with an impact greater than the preset standard as key carbon emission links; S5, constructing a decision tree model, using the data obtained in step S2 and step S4 to train the decision tree model, and using the decision tree model to obtain the monthly carbon emission sub-item forecast value; S6. Perform carbon footprint anomaly diagnosis based on the deviation model principle according to the monthly carbon emission sub-item forecast values.

[0007] Preferably, in step S1, the collected historical data of Chinese patent medicines include procurement, sales and transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy for each Chinese patent medicine; the collected monthly data of Chinese patent medicines include carbon emission-related items and monthly production data of each Chinese patent medicine product; the historical data of Chinese patent medicines and monthly data of Chinese patent medicines are collected from the information system of Chinese medicine pharmaceutical companies.

[0008] Preferably, in step S2, the process of preprocessing the collected historical data and monthly data of Chinese patent medicines includes: S201. Cleaning the collected historical data and monthly data of Chinese patent medicines by desensitizing, removing duplicates and filling missing values ​​on the collected historical data and monthly data of Chinese patent medicines; S202, standardizing and collating the cleaned historical data and monthly data of Chinese patent medicines, wherein the standardization and collating method is to associate, match and aggregate the historical data and monthly data of Chinese patent medicines from different sources; S203. Conduct quality assessment on the standardized and collated historical data and monthly data of Chinese patent medicines. The quality assessment method is to use a quality assessment mechanism to inspect and manage the quality of the historical data and monthly data of Chinese patent medicines.

[0009] Preferably, in step S3, procurement and sales transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy are obtained, and the carbon emission factor method is used to perform standard carbon footprint accounting of traditional Chinese medicines.

[0010] Preferably, in step S3, the standard carbon footprint accounting of Chinese patent medicines specifically includes: S301. Calculate the carbon emissions from the purchase, sales and transportation of each Chinese patent medicine; S302, calculating the carbon emissions of the storage process of each Chinese patent medicine; S303. Calculate the carbon emissions of each Chinese patent medicine processing step; S304. Calculate the standard carbon footprint of each Chinese patent medicine based on the carbon emissions in the procurement, sales and transportation links, the carbon emissions in the storage link, and the carbon emissions in the processing link.

[0011] Preferably, in step S3, the monthly plant-wide carbon emissions accounting is specifically to calculate the monthly plant-wide carbon emissions based on carbon emission related items, wherein the carbon emission related items include actual product transportation carbon emissions, actual product processing carbon emissions and actual product storage carbon emissions.

[0012] Preferably, in step S4, the monthly data of Chinese patent medicines include the number of product categories of monthly Chinese patent medicine products, the product batches corresponding to each type of product, each type of product in each batch, as well as the number of product categories, product batches, and the actual product transportation carbon emissions corresponding to each type of product, the actual product processing carbon emissions, the actual product storage carbon emissions and the standard carbon emissions.

[0013] Preferably, in step S5, the historical data of Chinese patent medicines and monthly data of Chinese patent medicines preprocessed in step S2, the standard carbon footprint of Chinese patent medicines and monthly carbon emissions of the whole plant obtained in step S3, and the key carbon emission links obtained in step S4 are input into the decision tree model, the decision tree model is trained, and the monthly carbon emission sub-item forecast values ​​are obtained using the decision tree model.

[0014] Preferably, in step S6, the decision tree model trained in step S5 is used to obtain the monthly carbon emission forecast value according to the number of product categories of Chinese patent medicine products in that month, the product batch size corresponding to each type of product, and each batch of each type of product. The forecast value is compared with the standard carbon emissions, and the total monitoring value is calculated. When the total monitoring value exceeds the set threshold, the monthly carbon emission sub-item forecast value is compared with the sub-item standard carbon emissions, and the sub-item monitoring value is calculated. When the sub-item monitoring value exceeds the set sub-item threshold, it is recorded as an abnormality in the carbon emission sub-item.

[0015] The present invention also provides a carbon footprint monitoring system for a Chinese medicine pharmaceutical enterprise, which uses the carbon footprint monitoring method for a Chinese medicine pharmaceutical enterprise as described above, and includes: a data collection robot module, a data governance robot module, a data analysis robot module, and a human-computer interaction robot module; The data collection robot module is used to collect historical data and monthly data of Chinese patent medicines; The data governance robot module is used to pre-process the collected historical data and monthly data of Chinese patent medicines, calculate the standard carbon footprint of Chinese patent medicines, and calculate the monthly carbon emissions of the entire factory; The data analysis robot module is used to obtain pre-processed monthly data of Chinese patent medicines, identify key carbon emission links, and predict monthly carbon emission sub-item values. It also monitors abnormal carbon emission sub-items and performs carbon footprint abnormality diagnosis. The human-computer interaction robot module is used to visualize the collected historical data of Chinese patent medicines, monthly data of Chinese patent medicines, key carbon emission links, monthly carbon emission sub-item forecast values, abnormal carbon emission sub-items and carbon footprint abnormality diagnosis results, and generate a visual user interface.

[0016] Compared with the prior art, the beneficial effects of the present invention are: (1) The present invention can collect basic data from various production links of traditional Chinese medicine pharmaceutical enterprises and analyze and optimize them through a decision tree model, which greatly reduces manual intervention and improves work efficiency; (2) The present invention adopts decision tree model and data analysis technology, and the obtained carbon emission prediction value has high accuracy and operability, and can fully reflect the carbon emission situation of the enterprise; (3) The present invention uses a trained decision tree model to predict carbon emissions, monitors abnormal carbon emission items of Chinese patent medicine pharmaceutical companies, and diagnoses abnormal carbon footprints, thereby achieving continuous improvement in carbon emission management. Chinese patent medicine pharmaceutical companies can pay attention to abnormal carbon footprints and abnormal carbon emission items in a timely manner, which helps companies maintain low-carbon operations in the long term, reduce carbon emissions, and achieve a win-win situation for corporate economic and environmental benefits. (4) The present invention provides a visual user interface, which allows users to view the historical data of Chinese patent medicines, monthly data of Chinese patent medicines, key carbon emission links, monthly carbon emission sub-item forecast values, abnormal carbon emission sub-items and carbon footprint abnormality diagnosis results at any time, so as to help users better manage the company's carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a method for monitoring carbon footprint of a traditional Chinese medicine pharmaceutical enterprise according to an embodiment of the present invention; Figure 2 This is a technical roadmap for the carbon footprint monitoring system for traditional Chinese medicine pharmaceutical companies according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in 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.

[0019] Example 1 Combination Figure 1 As shown, the embodiment of the present invention provides a carbon footprint monitoring method for a traditional Chinese medicine pharmaceutical enterprise, comprising the following steps: S1. Collection of historical and monthly data of Chinese patent medicines; S2. Preprocess the collected historical data and monthly data of Chinese patent medicines; S3. Obtain the pre-processed historical data of Chinese patent medicines, perform standard carbon footprint accounting of Chinese patent medicines, and perform monthly carbon emissions accounting for the entire factory; S4. Obtain the pre-processed monthly data of Chinese patent medicines, calculate the impact of improvements in different links of Chinese patent medicines on overall carbon emissions based on the standard carbon footprint of Chinese patent medicines, sort different links according to the impact, and identify links with an impact greater than the preset standard as key carbon emission links; S5, constructing a decision tree model, using the data obtained in step S2 and step S4 to train the decision tree model, and using the decision tree model to obtain the monthly carbon emission sub-item forecast value; S6. Perform carbon footprint anomaly diagnosis based on the deviation model principle according to the monthly carbon emission sub-item forecast values.

[0020] Example 2 On the basis of Example 1, in this embodiment, in step S1, the collected historical data of Chinese patent medicines include the procurement and sales transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy for each Chinese patent medicine; the collected monthly data of Chinese patent medicines include carbon emission related items and monthly production data of each Chinese patent medicine product; the historical data of Chinese patent medicines and monthly data of Chinese patent medicines are collected from the information system of Chinese medicine pharmaceutical companies; specifically, the information system mainly includes EMS system, EF database, eAM system, TMS system, CRM system, ERP system, MES system, WMS system, SRM system, SCADA system, etc.; For the common information systems of traditional Chinese medicine pharmaceutical companies, the collected data are mainly various data sources closely related to carbon footprint accounting. For example, Table 1 shows some common tables and their data fields as well as possible data management source systems: Table 1

[0021] Furthermore, in step S2, the process of preprocessing the collected historical data and monthly data of Chinese patent medicines includes: S201. Cleaning the collected historical data and monthly data of Chinese patent medicines by desensitizing, removing duplicates and filling missing values ​​on the collected historical data and monthly data of Chinese patent medicines; S202, standardizing and collating the cleaned historical data and monthly data of Chinese patent medicines, wherein the standardization and collating method is to associate, match and aggregate the historical data and monthly data of Chinese patent medicines from different sources; S203. Conduct quality assessment on the standardized and collated historical data and monthly data of Chinese patent medicines. The quality assessment method is to use a quality assessment mechanism to inspect and manage the quality of the historical data and monthly data of Chinese patent medicines.

[0022] Example 3 On the basis of Example 2, in this embodiment, in step S3, the procurement and sales transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy are obtained. Specifically, the procurement and sales transportation energy consumption data include the diesel energy consumption and gasoline energy consumption generated in the procurement and sales transportation links of each Chinese patent medicine, the storage energy consumption data include the electric energy consumption generated in the storage link of each Chinese patent medicine, the processing energy consumption data include the electric energy consumption, water energy consumption and raw material consumption generated in the processing link of each Chinese patent medicine, and the carbon emission factors of different types of energy include the carbon emission factor of diesel, the carbon emission factor of gasoline, the carbon emission factor of electricity, the carbon emission factor of water energy and the carbon emission factor of raw materials; The carbon emission factor method is used to calculate the standard carbon footprint of Chinese patent medicines. Specifically, the formula of the carbon emission factor method is: , in, represents carbon emissions, Represents energy consumption data, represents the carbon emission factor; Since it is difficult to obtain carbon footprint information for upstream raw material production and downstream drug use, and it is difficult to apply optimization control, the scope of carbon footprint accounting and optimization is generally limited to the range from material procurement and transportation, in-plant storage, in-plant processing to product sales and transportation. Detailed mapping of the entire production process of Chinese patent medicines from raw material acquisition to product delivery. Identify each step in the process, including raw material procurement, processing, packaging, transportation, distribution, etc. Collect energy consumption data for each process step, including electricity, water, steam, fuel, etc. Collect transportation data for raw materials and products, including transportation methods, distances, loads, etc. Determine the carbon emission factors for each energy source and raw material, which can be based on IPCC guidelines or industry standards. For the transportation link, determine the carbon emission factors for different modes of transportation. Using the collected data and carbon emission factors, calculate the carbon emissions of each process step. Further, summarize the carbon emissions of the entire value stream to obtain the total carbon footprint of the product. The established standard carbon footprint of Chinese patent medicines can be used as a basis for carbon emission optimization and as a benchmark for future comparison and improvement; Furthermore, the standard carbon footprint accounting of Chinese patent medicines is carried out, including: S301. Calculate the carbon emissions from the purchase, sales and transportation of each Chinese patent medicine. The specific calculation formula is as follows: , in, Indicates the carbon emissions from the purchase and sales transportation process. Indicates the carbon emissions generated when diesel is used in transportation. Represents the carbon emissions generated when gasoline is used in transportation. , in, Indicates the diesel energy consumption in the procurement and sales transportation links, represents the carbon emission factor of diesel,

[0023] in, Indicates the gasoline energy consumption in the purchase and sales transportation process, represents the carbon emission factor of gasoline; S302. Calculate the carbon emissions of the storage process of each Chinese patent medicine. The specific calculation formula is as follows: , in, represents the carbon emissions in the storage process, Indicates the carbon emissions generated by the electricity energy consumption in the storage stage, It indicates the amount of electricity consumption in the storage process due to the need for refrigerated storage of raw materials, intermediates and products. represents the carbon emission factor of electricity; S303. Calculate the carbon emissions of each Chinese patent medicine processing step. The specific calculation formula is as follows: , in, Indicates the carbon emissions of each Chinese patent medicine processing link, Indicates the carbon emissions generated by the water and energy consumption in each Chinese patent medicine processing link, Indicates the carbon emissions generated by the electricity consumption of each Chinese patent medicine in the processing stage. It represents the carbon emissions generated by all raw materials consumed in the processing of each Chinese patent medicine. , in, Indicates the water energy consumption generated by each Chinese patent medicine in the processing link, represents the carbon emission factor of hydropower; , in, Indicates the electricity energy consumption generated by each Chinese patent medicine in the processing link, represents the carbon emission factor of electricity; , in, Indicates The consumption of raw materials, Indicates Carbon emission factor of each raw material; S304. Calculate the standard carbon footprint of each Chinese patent medicine based on the carbon emissions in the procurement, sales and transportation stages, the carbon emissions in the storage stage, and the carbon emissions in the processing stage. The specific calculation formula is as follows: ; Furthermore, the monthly carbon emission accounting of the whole plant is specifically to calculate the monthly carbon emission of the whole plant according to the carbon emission related items, where the carbon emission related items include the carbon emission of actual product transportation, the carbon emission of actual product processing and the carbon emission of actual product storage. Specifically, the real-time water energy consumption of the water meter, the real-time electric energy consumption of the electric meter, gasoline consumption, diesel consumption and raw material consumption in the actual product transportation, processing and storage sub-items can be counted, as shown in Table 2: Table 2

[0024] The specific formula for calculating monthly carbon emissions for the entire plant is: , in, Indicates the actual total carbon emissions of the entire plant in a month, Indicates actual monthly transportation Carbon emissions generated by each product, Indicates actual monthly processing Carbon emissions generated by each product, Indicates actual monthly storage Carbon emissions generated by each product, Indicates products The carbon emissions generated in actual transportation, Indicates products The carbon emissions generated in actual processing, Indicates products The carbon emissions generated during actual storage, Indicates products Carbon emission factors for carbon emissions in actual transportation, Indicates products The carbon emission factor of carbon emissions in actual processing, Indicates products Carbon emission factor for carbon emissions in actual storage.

[0025] Example 4 On the basis of Example 2, in this embodiment, in step S4, the monthly data of Chinese patent medicines include the number of product categories of monthly Chinese patent medicine products, the product batches corresponding to each type of product, each batch of each type of product, and the number of product categories, product batches, and the actual product transportation carbon emissions corresponding to each type of product, the actual product processing carbon emissions, the actual product storage carbon emissions, and the standard carbon emissions; the pre-processed monthly data of Chinese patent medicines are obtained, and the degree of impact of improvements in different links of Chinese patent medicines on the overall carbon emissions is calculated according to the standard carbon footprint of Chinese patent medicines (for example, Table 3), and different links are sorted according to the degree of impact, and the links with an impact greater than the preset standard are identified as key carbon emission links, and carbon emission improvements are implemented preferentially for the key carbon emission links; Table 3

[0026] Further, in step S5, the historical data of Chinese patent medicines and monthly data of Chinese patent medicines preprocessed in step S2, the standard carbon footprint of Chinese patent medicines and the monthly carbon emissions of the whole plant obtained in step S3, and the key carbon emission links obtained in step S4 are input into the decision tree model, the decision tree model is trained, and the monthly carbon emission sub-item forecast value is obtained using the decision tree model; Specifically, the decision tree model is a classification and regression method based on a tree structure. Here, a regression variant of the decision tree is used to predict carbon emissions; the decision tree divides the data through a series of questions, each internal node judges a feature, and according to the judgment result, the data is assigned to the next level node until it reaches the leaf node, which gives the predicted value; The decision tree model inputs the data matrix X (e.g. Table 4) (monthly product production); Table 4

[0027] The decision tree model prediction result data matrix Y (e.g. Table 5) (monthly carbon emission sub-item prediction value); Table 5

[0028] The random forest decision tree method was used. The number of decision trees in the random forest was set to 50. The Bootsrap sampling method was used to extract multiple subsamples from the original data. A decision tree was trained for each subsample. The prediction results of all decision trees were combined to obtain the final prediction value. During the model training process, different decision tree depths (such as 5, 10, 15, etc.) were set, cross-validation was used to evaluate model performance, and the depth with the best performance was selected as the depth of the final model; different numbers of small leaf nodes (such as 5, 10, 20, etc.) were set, cross-validation was used to evaluate model performance, and the small leaf nodes with the best performance were selected as the parameters of the final model; different splitting criteria such as Gini impurity and information gain were used to evaluate model performance, and the splitting criteria with the best performance were selected as the splitting criteria of the final model.

[0029] The decision tree model training results are as follows: Training set shape: (46, 15) Test set shape: (13, 15) Optimizing decision tree depth: Depth 5: Average MSE = 2.3829 Depth 10: Average MSE = 2.3558 Depth 15: Average MSE = 2.3557 Depth 20: Average MSE = 2.3557 Depth 25: Average MSE = 2.3557 DepthNone: Average MSE = 2.3557 Optimal depth: 15 Optimize the number of leaf nodes: Minimum number of leaf nodes 1: Average MSE = 2.3557 Minimum number of leaf nodes 5: Average MSE = 4.8115 Minimum number of leaf nodes 10: Average MSE = 7.8475 Minimum number of leaf nodes 15: Average MSE = 13.2646 Minimum number of leaf nodes 20: Average MSE = 13.2646 Minimum number of leaf nodes 25: Average MSE = 13.2646 Optimal number of leaf nodes: 1 Optimize the splitting criteria: Split criterion squared_error: Average MSE = 2.3557 Splitting criterion friedman_mse: Average MSE = 2.4344 Split criterion absolute_error: Average MSE = 2.6747 Split standard poisson: Average MSE = 2.7192 Optimal splitting criterion: squared_error Training set performance evaluation: Transport Carbon Emissions Summary - MSE: 0.1851, R2: 0.9668 Processing carbon emissions summary - MSE: 1.2104, R2: 0.9611 Storage Carbon Emissions Summary - MSE: 0.0645, R2: 0.9691 Test set performance evaluation: Transport Carbon Emissions Summary - MSE: 0.0563, R2: 0.9514 Processing Carbon Emissions Summary - MSE: 0.6765, R2: 0.8686 Storage Carbon Emissions Summary - MSE: 0.0348, R2: 0.9050 Further, in step S6, the decision tree model trained in step S5 is used to obtain the monthly carbon emission forecast value according to the number of product categories of Chinese patent medicine products in that month, the product batches corresponding to each type of product, and each batch of each type of product. The forecast value is compared with the standard carbon emission amount, and the total monitoring value is calculated. When the total monitoring value exceeds the set threshold, the monthly carbon emission sub-item forecast value is compared with the sub-item standard carbon emission amount, and the sub-item monitoring value is calculated. When the sub-item monitoring value exceeds the set sub-item threshold, it is recorded as an abnormality in the carbon emission sub-item. The specific calculation process is as follows: , in, It represents the predicted carbon emission value of the whole plant for that month. It is a product Standard carbon emissions; When the total monitoring value exceeds the set threshold a, the sub-item monitoring value is calculated. The calculation formula is as follows: , , , in, represents the monthly transportation carbon emission forecast value, represents the monthly standard carbon emissions of transportation, represents the monthly storage carbon emission forecast value, represents the monthly storage standard carbon emissions, represents the monthly processing carbon emission forecast value, Indicates the monthly processing standard carbon emissions, the set sub-item threshold of the transportation monitoring value is b, the set sub-item threshold of the storage monitoring value is c, and the set sub-item threshold of the processing monitoring value is d; Compare the predicted value of each sub-item with the standard value of each sub-item (e.g. Table 6), calculate the difference between the predicted value and the standard value of each sub-item, and record it as the sub-item monitoring value. When the sub-item monitoring value exceeds the set sub-item threshold, it is recorded as an abnormality in the carbon emission sub-item. Set the sub-item thresholds b, c, and d to 20%. When the difference value of a sub-item exceeds 20%, it is considered that the sub-item is abnormal, and a detailed deviation investigation is conducted.

[0030] Table 6

[0031] Example 5 Combination Figure 2 As shown, an embodiment of the present invention further provides a carbon footprint monitoring system for a Chinese medicine pharmaceutical enterprise, which uses the carbon footprint monitoring method for a Chinese medicine pharmaceutical enterprise as described above, including: a data collection robot module, a data governance robot module, a data analysis robot module and a human-computer interaction robot module; The data collection robot module is used to collect historical data and monthly data of Chinese patent medicines; The data governance robot module is used to pre-process the collected historical data and monthly data of Chinese patent medicines, calculate the standard carbon footprint of Chinese patent medicines, and calculate the monthly carbon emissions of the entire factory; The data analysis robot module is used to obtain pre-processed monthly data of Chinese patent medicines, identify key carbon emission links, and predict monthly carbon emission sub-item values. It also monitors abnormal carbon emission sub-items and performs carbon footprint abnormality diagnosis. The human-computer interaction robot module is used to visualize the collected historical data of Chinese patent medicines, monthly data of Chinese patent medicines, key carbon emission links, monthly carbon emission sub-item forecast values, abnormal carbon emission sub-items and carbon footprint abnormality diagnosis results, and generate a visual user interface.

[0032] The above description is only an embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement and improvement made within the application scope of the present invention shall be included in the protection scope of the present invention.

Claims

1. A carbon footprint monitoring method for a traditional Chinese medicine pharmaceutical enterprise, characterized in that: The following steps are involved: S1. Collection of historical data and monthly data of Chinese patent medicines; S2. Preprocess the collected historical data and monthly data of Chinese patent medicines; S3. Obtain the pre-processed historical data of Chinese patent medicines, perform standard carbon footprint accounting of Chinese patent medicines, and perform monthly carbon emissions accounting for the entire factory; S4. Obtain the pre-processed monthly data of Chinese patent medicines, calculate the impact of improvements in different links of Chinese patent medicines on overall carbon emissions based on the standard carbon footprint of Chinese patent medicines, sort different links according to the impact, and identify links with an impact greater than the preset standard as key carbon emission links; S5, constructing a decision tree model, using the data obtained in step S2 and step S4 to train the decision tree model, and using the decision tree model to obtain the monthly carbon emission sub-item forecast value; S6. Perform carbon footprint anomaly diagnosis based on the deviation model principle according to the monthly carbon emission sub-item forecast values.

2. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 1, characterized in that: In step S1, the collected historical data of Chinese patent medicines include the procurement, sales and transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy for each Chinese patent medicine; the collected monthly data of Chinese patent medicines include carbon emission-related items and monthly production data of each Chinese patent medicine product; the historical data of Chinese patent medicines and monthly data of Chinese patent medicines are collected from the information system of Chinese medicine pharmaceutical companies.

3. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 2, characterized in that: In step S2, the process of preprocessing the collected historical data and monthly data of Chinese patent medicines includes: S201. Cleaning the collected historical data and monthly data of Chinese patent medicines by desensitizing, removing duplicates and filling missing values ​​on the collected historical data and monthly data of Chinese patent medicines; S202, standardizing and collating the cleaned historical data and monthly data of Chinese patent medicines, wherein the standardization and collating method is to associate, match and aggregate the historical data and monthly data of Chinese patent medicines from different sources; S203. Conduct quality assessment on the standardized and collated historical data and monthly data of Chinese patent medicines. The quality assessment method is to use a quality assessment mechanism to inspect and manage the quality of the historical data and monthly data of Chinese patent medicines.

4. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 3 is characterized in that: In step S3, the procurement and sales transportation energy consumption data, storage energy consumption data, processing energy consumption data and carbon emission factors of different types of energy are obtained, and the carbon emission factor method is used to calculate the standard carbon footprint of traditional Chinese medicine.

5. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 4, characterized in that: In step S3, the standard carbon footprint accounting of Chinese patent medicines specifically includes: S301. Calculate the carbon emissions from the purchase, sales and transportation of each Chinese patent medicine; S302, calculating the carbon emissions of the storage process of each Chinese patent medicine; S303. Calculate the carbon emissions of each Chinese patent medicine processing step; S304. Calculate the standard carbon footprint of each Chinese patent medicine based on the carbon emissions in the procurement, sales and transportation links, the carbon emissions in the storage link, and the carbon emissions in the processing link.

6. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 5, characterized in that: In step S3, the monthly carbon emissions accounting for the entire plant is specifically to calculate the monthly carbon emissions for the entire plant based on carbon emission related items, wherein the carbon emission related items include the actual product transportation carbon emissions, the actual product processing carbon emissions and the actual product storage carbon emissions.

7. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 6, characterized in that: In step S4, the monthly data of Chinese patent medicines include the number of product categories of monthly Chinese patent medicine products, the product batches corresponding to each type of product, each type of product in each batch, as well as the number of product categories, product batches, and the actual product transportation carbon emissions corresponding to each type of product, the actual product processing carbon emissions, the actual product storage carbon emissions and the standard carbon emissions.

8. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 7, characterized in that: In step S5, the historical data of Chinese patent medicines and monthly data of Chinese patent medicines preprocessed in step S2, the standard carbon footprint of Chinese patent medicines and the monthly carbon emissions of the whole factory obtained in step S3, and the key carbon emission links obtained in step S4 are input into the decision tree model, the decision tree model is trained, and the monthly carbon emission sub-item forecast values ​​are obtained using the decision tree model.

9. The carbon footprint monitoring method for traditional Chinese medicine pharmaceutical enterprises according to claim 8, characterized in that: In step S6, the decision tree model trained in step S5 is used to obtain the monthly carbon emission forecast value based on the number of product categories of Chinese patent medicine products in that month, the product batch size corresponding to each type of product, and each batch of each type of product. The forecast value is compared with the standard carbon emissions, and the total monitoring value is calculated. When the total monitoring value exceeds the set threshold, the monthly carbon emission sub-item forecast value is compared with the sub-item standard carbon emissions, and the sub-item monitoring value is calculated. When the sub-item monitoring value exceeds the set sub-item threshold, it is recorded as an abnormality in the carbon emission sub-item.

10. A carbon footprint monitoring system for traditional Chinese medicine pharmaceutical enterprises, characterized in that: A carbon footprint monitoring method for a traditional Chinese medicine pharmaceutical enterprise as described in any one of claims 1 to 9, comprising: a data collection robot module, a data governance robot module, a data analysis robot module and a human-computer interaction robot module; The data collection robot module is used to collect historical data and monthly data of Chinese patent medicines; The data governance robot module is used to pre-process the collected historical data and monthly data of Chinese patent medicines, calculate the standard carbon footprint of Chinese patent medicines, and calculate the monthly carbon emissions of the entire factory; The data analysis robot module is used to obtain pre-processed monthly data of Chinese patent medicines, identify key carbon emission links, and predict monthly carbon emission sub-item values. It also monitors abnormal carbon emission sub-items and performs carbon footprint abnormality diagnosis. The human-computer interaction robot module is used to visualize the collected historical data of Chinese patent medicines, monthly data of Chinese patent medicines, key carbon emission links, monthly carbon emission sub-item forecast values, abnormal carbon emission sub-items and carbon footprint abnormality diagnosis results, and generate a visual user interface.

Images