Abstract

The patent designs an evaluation method and a platform for the full life cycle carbon neutralization performance of a manufacturing enterprise, which are evaluated from two dimensions of enterprise organization full life cycle carbon management evaluation and product full life cycle carbon management evaluation, comprehensively reflect the carbon neutralization performance of the manufacturing enterprise, and perform the grounding of an evaluation method model through a big data technology and an artificial intelligence technology, and perform professional and accurate evaluation on the performance of the manufacturing enterprise on a low-carbon transformation road in a digital and intelligent mode. The patent aims at effectively evaluating characteristic points of the manufacturing industry and solves the problem that the carbon neutralization performance evaluation standard of the current manufacturing industry is absent. Meanwhile, through an intelligent platform for data mining treatment, fusion analysis and calculation application, the defects of long time consumption, low data utilization rate, subjective prejudice in scoring, incomplete evaluation caused by data deletion, inaccuracy and the like of manual information collection are overcome.

Description

Method and platform for evaluating full life cycle carbon neutralization performance of manufacturing enterprise

Technical Field

The invention relates to the technical field of carbon neutralization evaluation, in particular to an evaluation method and platform for full life cycle carbon neutralization performance of manufacturing enterprises.

Background

Climate change severely threatens our lives society, economy and environment. As a typical representative of high carbon emissions, the emitted greenhouse gases exacerbate climate change, and thus there is a need to achieve low carbon conversion. However, none of the existing evaluation methods in the carbon neutral market are set for manufacturing enterprises, and are used for evaluating the carbon neutral performance of the enterprises, and the data-driven evaluation standards and systems are covered for organizing and product-level full life cycle carbon neutral performance;

the current problems of evaluating carbon neutralization performance of manufacturing enterprises mainly comprise the following four points:

(1) At present, an evaluation standard for the carbon neutralization performance of an enterprise is not specific to a manufacturing enterprise, and effective evaluation cannot be performed on characteristic points of the manufacturing industry;

(2) The data loss causes incomplete and inaccurate evaluation;

(3) Comprehensive digitization is not realized, big data and artificial intelligence technology are not well applied, time and labor are consumed, and the data utilization rate is low;

(4) In the scoring process, subjective judgment is mainly performed by people, subjective prejudice exists, and fairness of evaluation are difficult to embody.

Specifically, the current evaluation criteria for the carbon neutralization performance of enterprises do not aim at the characteristic points of the manufacturing industry, such as raw material purchase, product design, production process and the like, and cannot be effectively evaluated. Meanwhile, there is a great limitation in carbon neutralization performance evaluation on the aspect of the company that the carbon neutralization evaluation target range does not cover both organization and product layers. Most of the existing accounting standards aiming at the carbon emission of enterprises pay attention to the greenhouse gas emission numerical accounting of enterprises or products, and the carbon neutralization transformation measures of the enterprises at the strategic, talent, management and feedback and efficiency benefit organization levels, and the contents of target setting, technical innovation and management implementation in the whole life cycle process of the products are not fully considered, so that the evaluation contents are lacking. In addition, the implementation means of the evaluation method is more traditional, and the acquisition of industry data, public data and enterprise internal data is insufficient depending on the manual or enterprise self-provided mode to acquire data. The serious loss of data results in incomplete and inaccurate evaluation. The existing method also does not realize digitization or use some technologies in big data and machine learning, and is time-consuming and labor-consuming, and low in data utilization rate. Meanwhile, in the existing evaluation standard of the carbon neutralization performance of enterprises or products, the subjective judgment is mainly used in the scoring process, subjective prejudice exists, and fairness of evaluation are difficult to embody. The lack of content and logic of the evaluation criteria makes it difficult for manufacturing enterprises to clearly recognize their own carbon neutralization performance levels and where they should be improved.

Disclosure of Invention

The invention aims to solve the technical problems, and provides a method and a platform for evaluating the full life cycle carbon neutralization performance of a manufacturing enterprise, wherein the unique evaluation system and method are utilized to evaluate the carbon neutralization performance of the manufacturing enterprise in two dimensions of organization and product, the grounding of the evaluation method is performed through a big data technology and an artificial intelligence technology, and the performance of the manufacturing enterprise on a low-carbon transformation road is evaluated in a professional and accurate way through a digital and intelligent mode.

In order to solve the technical problems, the technical scheme provided by the invention is as follows:

a method for evaluating full life cycle carbon neutralization performance of manufacturing enterprises comprises the following steps:

s1: information is dug, and various information data related to enterprise carbon neutralization and carbon emission are imported into a data lake through recognition, extraction and cleaning;

s2: information classification, namely classifying the collected evaluated enterprise data stored in the data lake through a text classification model and combining with an evaluation method to organize full-life-cycle carbon management evaluation and product full-life-cycle carbon management evaluation, forming and constructing an enterprise knowledge graph, wherein the enterprise full-life-cycle carbon neutralization performance evaluation method integrates the ideas of Mobius loop circulation and closed loop and the ideas of life-cycle evaluation LCA into evaluation, and evaluates the two dimensions of the organization full-life-cycle evaluation and the product full-life-cycle evaluation;

s3: classifying and grading, namely extracting information from the enterprise knowledge graph according to the evaluation indexes in the evaluation method by using a semantic analysis technology to generate an evaluation base table, wherein the evaluation base table covers all evaluation indexes evaluated by an evaluated company and collected grading bases corresponding to all the evaluation indexes, and the evaluation bases are enterprise data and information acquired from the enterprise knowledge graph through text matching;

(1) Comparing and scoring the data of enterprises similar to the industries in the data lake;

(2) Using a natural language processing model, scoring an evaluation base table through a carbon neutralization measurement algorithm and a classification model, and obtaining a score vector through multiple sampling;

s4: the method comprises the steps of total grading, splicing score vectors to form a score matrix, fusing the score matrix of each index by using a carbon neutralization quantitative grading model to obtain a final score matrix, calculating the average number and standard deviation of all elements in the score matrix to obtain a final grading table, wherein the grading table covers carbon neutralization performance of the whole life cycle of a manufacturing enterprise, simultaneously considering mutual linkage influence of each index in an organization and a product layer, fully considering the relationship between the enterprise and society and environment in a low-carbon transformation process, taking economic benefit, environmental benefit, social benefit and industry benefit generated in the transformation process into investigation, comprehensively evaluating the carbon neutralization performance of the enterprise comprehensively, optimizing a scheme, using case data in a data lake and case data in an ancestor intelligent library according to the total grading table and expert experience, deducing best practices suitable for improving the effective method of the carbon neutralization performance of the enterprise, and helping the enterprise to determine the carbon neutralization lifting direction. Meanwhile, through data feedback of the whole life cycle of the enterprise product, further evaluation and deduction are carried out regularly, and a continuously-lifted closed loop is constructed.

After adopting the structure, the invention has the following advantages:

1. when the method and the platform for evaluating the full life cycle carbon neutralization performance of the manufacturing enterprise are used, digitization and intellectualization are realized, and the defects of data loss, low data utilization rate, long time consumption for manually collecting information, waste manpower, subjective bias in scoring and the like in the evaluation process can be overcome through a data acquisition management module, a data fusion scoring module and a data mining and artificial intelligence algorithm based on big data in a data calculation module;

2. the evaluation method and the platform can bring each link of enterprise organization carbon management and product carbon management into a research range aiming at manufacturing enterprise design, so that whether enterprises perform adequate preparation for carbon neutralization and low-carbon transformation from the aspects of enterprise strategy and management, a method and a system related to carbon neutralization are formulated, whether closed-loop design exists on top layer design or not, and the actual operation of enterprises is researched by the aspect of enterprise actual operation, and the carbon emission in the production process of the products is researched;

3. and the evaluation from top to bottom is carried out from top to bottom in the two dimensions of enterprise organization full cycle carbon management evaluation and product full cycle carbon management evaluation, so that the carbon neutralization performance of manufacturing enterprises is comprehensively reflected. Compared with other evaluation standards, the evaluation method and the platform can evaluate the carbon neutralization performance of the manufacturing enterprises more comprehensively and comprehensively, and can effectively enable the manufacturing enterprises to clearly recognize the carbon neutralization performance level and future stress improvement points of the enterprises;

4. the evaluation method and the platform are based on a deep learning model constructed by massive data, cases and actual combat experience, deduce best practices suitable for improving the carbon neutralization of enterprises and represent effective methods, and assist enterprises in determining the carbon neutralization improving direction. Meanwhile, through data feedback of the whole life cycle of the enterprise product, further evaluation and deduction are carried out regularly, and a continuously-lifted closed loop is constructed.

As an improvement, the product full life cycle carbon management evaluation in the step S1 is a closed loop evaluation model formed by research and development design, raw material purchase and transportation, production and packaging, distribution and use, and scrapped recovery full life cycle links, and the organization full life cycle carbon management evaluation is a closed loop evaluation model formed by examining strategic, talent, management and operation, evaluation and feedback mechanisms and efficiency benefit links of enterprises.

As an improvement, the evaluation index in the step S3 includes an organization full life cycle carbon management evaluation index and a product full life cycle carbon management evaluation index.

As an improvement, the organization full life cycle carbon management evaluation index comprises: the method comprises the steps of quantifying carbon reduction targets and promises of enterprises, scientifically quantifying the carbon reduction targets and promises of the enterprises, planning time of low-carbon transformation of the enterprises, planning fund input of low-carbon transformation of the enterprises, measuring and calculating cost benefits of the low-carbon transformation, low-carbon transformation risk cognition, establishment conditions of carbon neutralization departments, recruitment plans of carbon neutralization talents, recruitment conditions of carbon neutralization talents, cultivation and development of carbon neutralization talents, low-carbon offices, low-carbon consumption, carbon ticket mechanisms, execution conditions of enterprise carbon investigation, establishment conditions of carbon budget, participation of carbon trade conditions, carbon asset management performance evaluation system construction conditions, carbon risk management system, capacity and level of the enterprises for coping with carbon risks, partner management system, enterprise exposure conditions, low-carbon authentication acquisition conditions, carbon verification conditions, carbon public benefits, carbon neutral stage evaluation, knowledge precipitation, feedback mechanisms, enterprise organization level carbon reduction conditions, enterprise organization level energy saving conditions, carbon neutralization topic degrees, social reputation and industry contribution.

As an improvement, the product full life cycle carbon management evaluation index comprises: the method comprises the steps of product research and development design, raw material purchase transportation, production and packaging, distribution and use, carbon reduction target of individual links of scrapped recovery product life cycle, management system, budget setting, technical innovation quantity, self-grinding degree, carbon reduction technology and technology fund investment, carbon reduction effect and overall efficiency benefit, and comprises product level carbon reduction, product organization level energy saving, total asset return rate, sales net benefit rate, cost benefit ratio and industry contribution.

An evaluation platform for full life cycle carbon neutralization performance of manufacturing enterprises comprises a data source, a data acquisition and management module, a data fusion scoring module and a data calculation module from bottom to top,

the data acquisition management module is used for identifying, disassembling, classifying, storing, analyzing and managing various data acquired from the data source, and disambiguating the data to construct carbon neutralization metadata and identify carbon neutralization main data;

the data fusion scoring module uses two main data of product full life cycle carbon management data and organization full life cycle carbon management data, and calculates enterprise carbon neutralization evaluation base table scoring and each classification scoring item by combining a quantitative scoring model and a carbon neutralization measuring algorithm;

the data calculation module can calculate a final scoring table of enterprise carbon neutralization performance based on the treated main data and carbon neutralization measurement algorithm in the platform and the carbon neutralization quantitative scoring model to determine the current carbon neutralization capacity of the enterprise, and is used for analyzing and predicting carbon data and providing optimization suggestions for the carbon neutralization direction of the enterprise.

As an improvement, the data sources are divided into internal data and external data.

Drawings

FIG. 1 is a schematic diagram of the scoring flow chart of the evaluation method of the full life cycle carbon neutralization performance of the manufacturing enterprise according to the invention.

FIG. 2 is a schematic diagram of the architecture of the full life cycle carbon neutralization performance evaluation platform of the manufacturing enterprise of the present invention.

Detailed Description

The present invention will be described in further detail in connection with the following.

With reference to figures 1 and 2 of the drawings,

the method for evaluating the full-Life cycle carbon neutralization performance of the manufacturing enterprise integrates the ideas of the mobius loop circulation and the closed loop and the ideas of Life-cycle assessment (LCA) into an evaluation, and evaluates from top to bottom from shallow to deep from two dimensions of enterprise organization full-cycle carbon management evaluation and product full-cycle carbon management evaluation to comprehensively reflect the carbon neutralization performance of the manufacturing enterprise. In terms of the product level, the evaluation is carried out from the links of research and development design, raw material purchase and transportation, production and packaging, distribution and use, scrapping and recovery, and meanwhile, in each link, the evaluation from the goal and management to the result is considered to form a closed-loop evaluation model. And in the organization level, the closed loop evaluation model formed by enterprises in strategic, talent, management and operation, evaluation and feedback mechanism links is mainly examined.

The evaluation model is shown in fig. 1, and comprises the following steps:

s1: the information is dug, and related information and data of carbon neutralization provided by enterprises such as news, reports, regulation and history documents, carbon neutralization information disclosure information, carbon management modes, carbon neutralization department information, low-carbon conversion strategy, low-carbon technology of products and the like related to the evaluated enterprises are obtained from news, reports, regulation and history documents in news and official websites, internal document libraries of the evaluated enterprises, reference data and ancestor intelligent libraries through a crawler technology; if the obtained documents and data have image data, the OCR technology is used for identifying the related data in the image documents and extracting the data in the carbon emission report; finally, various information data related to enterprise carbon neutralization and carbon emission are imported into a data lake through a non-negative matrix factorization NMF algorithm and recognition, extraction and cleaning of other information;

s2: classifying information, namely classifying the collected evaluated enterprise data stored in the data lake through a text classification model and combining the whole period carbon management evaluation of an evaluation method and the whole period carbon management evaluation of a product, mining the knowledge structure of unstructured data in the data lake through a path sequencing algorithm PRA, extracting related entities and relations, carrying out logic attribution, redundancy and error filtering on the data, and forming and constructing an enterprise knowledge graph; the patent evaluation method defines a classification method, namely, the classification method is evaluated by taking an organization full life cycle evaluation and a product full life cycle evaluation as two major dimensions, wherein the product full life cycle evaluation is a closed loop evaluation model formed by research and development design, raw material purchase and transportation, production and packaging, distribution and use, scrapping and recovery full life cycle links, and the organization full life cycle evaluation is a closed loop evaluation model formed by examining strategic, talent, management and operation, evaluation and feedback mechanisms and efficiency benefit links of enterprises.

S3: and (3) classifying and grading, namely extracting information from the knowledge graph according to the evaluation indexes in the enterprise full life cycle carbon neutralization performance evaluation model by a semantic analysis technology to generate an evaluation base table, wherein the evaluation base table covers all evaluation indexes evaluated by an evaluated company and collected grading basis corresponding to all the evaluation indexes, and the evaluation basis is enterprise data and information obtained from the enterprise knowledge graph through text matching.

According to the subdivision industry of enterprises, data to be compared with other enterprises of the industry, such as the carbon neutralization target year, the carbon emission and the carbon emission reduction ratio, are aimed. This patent proposes to extract carbon emissions and carbon management data for the same industry enterprises in the data lake. And grabbing similar enterprises through multi-dimensional comparison, acquiring carbon emission and carbon management data information data, taking the carbon emission and the carbon management data information data as a comparison data set, and comparing to obtain the section of the enterprise as an evaluation result.

The method comprises the steps of circulating a neural network (recurrent neural network, RNN) model by using a natural language processing (Natural language processing, NLP) technology, carrying out semantic extraction, extracting key information in collected evaluation basis information, scoring an evaluation base table by using a carbon neutralization measurement algorithm, intelligently matching qualitative and quantitative characteristics of evaluation dimensions in the evaluation base table, drawing an enterprise carbon neutralization capacity image, carrying out scoring and classification by using a two-class or multi-class model, carrying out multiple sampling, and recording data of each sampling to obtain a score vector from low to high.

The above-mentioned organization full life cycle carbon management evaluation index includes: the method comprises the steps of quantifying carbon reduction targets and promises of enterprises, scientifically quantifying the carbon reduction targets and promises of the enterprises, planning time of low-carbon transformation of the enterprises, planning fund input of low-carbon transformation of the enterprises, measuring and calculating cost benefits of the low-carbon transformation, low-carbon transformation risk cognition, establishment conditions of carbon neutralization departments, recruitment plans of carbon neutralization talents, recruitment conditions of carbon neutralization talents, cultivation and development of carbon neutralization talents, low-carbon offices, low-carbon consumption, carbon ticket mechanisms, execution conditions of enterprise carbon investigation, establishment conditions of carbon budget, participation of carbon trade conditions, carbon asset management performance evaluation system construction conditions, carbon risk management system, capacity and level of the enterprises for coping with carbon risks, partner management system, enterprise exposure conditions, low-carbon authentication acquisition conditions, carbon verification conditions, carbon public benefits, carbon neutral stage evaluation, knowledge precipitation, feedback mechanisms, enterprise organization level carbon reduction conditions, enterprise organization level energy saving conditions, carbon neutralization topic degrees, social reputation and industry contribution.

The whole life cycle carbon management evaluation index of the product comprises the following components: the method comprises the steps of product research and development design, raw material purchase transportation, production and packaging, distribution and use, carbon reduction target of individual links of scrapped recovery product life cycle, management system, budget setting, technical innovation quantity, self-grinding degree, carbon reduction technology and technology fund investment, carbon reduction effect and overall efficiency benefit, and comprises product level carbon reduction, product organization level energy saving, total asset return rate, sales net benefit rate, cost benefit ratio and industry contribution.

S4: and (3) total scoring, namely splicing vectors under the same primary evaluation index to form a score matrix, wherein the columns of the score matrix are index score vectors, obtaining the score matrix of the primary index by using a carbon neutralization quantization scoring model, covering the possibility that the primary index score is from high to low, calculating the average number and standard deviation of all elements in the score matrix of the primary index, and obtaining the average number and standard deviation of all the primary indexes to obtain a final scoring table. The evaluation table covers carbon neutralization performance of the whole life cycle of the manufacturing industry enterprise, simultaneously considers that all indexes in the organization and product layers have mutual linkage influence, fully considers the relationship between the enterprise and society and environment in the low-carbon transformation process, and takes into consideration economic benefit, environmental benefit, social benefit and industry benefit generated in the transformation process, thereby comprehensively evaluating the carbon neutralization performance of the enterprise.

According to the optimization scheme, according to the total scoring table and expert experience, the data in the data lake and the case data in the ancestor intelligent library are used for constructing a deep learning model, so that best practices suitable for improving the carbon neutralization expression effective method of enterprises are deduced, and the enterprises are assisted to determine the carbon neutralization improvement direction. Meanwhile, through data feedback of the whole life cycle of the enterprise product, further evaluation and deduction are carried out regularly, and a continuously-lifted closed loop is constructed.

The invention provides an evaluation platform for the full life cycle carbon neutralization performance of a manufacturing enterprise, which is composed of a data source, a data acquisition management module, a data fusion scoring module and a data calculation module from bottom to top as shown in fig. 2.

The data source is divided into internal data and external data, and the data source also comprises an ancestor intelligence library.

The internal data comprise data obtained through consultation and investigation, and meanwhile, various digitization systems of an enterprise are subjected to data capture, such as capturing production energy consumption and carbon emission data from an EMS (energy consumption management system), capturing product BOM (bill of materials) from an MES (manufacturing execution system), producing bill of materials data, capturing employee office, employee traffic, travel data, various contract and bill data from an OA (office automation system), capturing product sales data, product distribution data, product use and waste process data from a CRM (customer relationship management system), capturing energy consumption related financial data, purchasing data, fund input plan, department setting conditions, material data, product data, supply chain data and various reports and bill material data from an internal database. The ancestor intelligent warehouse is manufacturing industry accumulated by the operation of ancestor science and technology for many years and carbon neutralization related data of each subdivision industry. External data includes internet (industry analysis, report disclosure, public relations, competitors), social media data.

The data acquisition and management module comprises a task distribution queue based on an Apache Kafka distributed message system and an object holder suitable for general data, and is used for identifying, disassembling, classifying, storing, analyzing, managing and disambiguating various data acquired from a data source to construct carbon neutralization metadata and identification carbon neutralization main data. The specific embodiments are as follows:

the data management task distribution queue is developed through the message queue of the existing Kafka, receives data which is obtained from each data source and is preprocessed into the data which can be identified by the platform, and can distribute different data to different processing task flows according to system configuration for corresponding processing.

The generic data storage is to provide different storage capacities for data to be processed and processed so as to store various types of data, including numbers, texts, pictures, images and binary files.

The data acquisition process comprises the following functions:

1. multidimensional disassembly: and disassembling various data acquired from the data source according to different dimensionalities of the related indexes.

Ocr (optical character recognition) recognition: for various non-data content such as paper materials, digital images, various documents can be identified with valid data items therein by OCR technology.

Dama classification: a method of classifying data in terms of data management in DAMA international (DAMA International).

4. Semantic recognition: for consultation studies, internal databases, image data and non-digital class data acquired on the internet and social media, it is necessary to identify the type of data, purpose or object model described by analyzing semantics in the corresponding text.

5. Domain oriented crawler: relevant information of social media resources is crawled in a targeted manner aiming at the relevant Internet of each subdivision field of the manufacturing industry.

The data management process comprises the following functions:

nmf (non-negative matrix factorization) identification: a non-negative matrix factorization method is used to determine the validity of the input data and determine whether to record/employ the input data.

2. Full-chain analysis: and (3) carrying out full-chain hierarchical clustering analysis by using the similarity matrix, and comparing and correlating the input data to form the relation among all pieces of main data.

3. Blood margin analysis: the relationship between the main data is analyzed to find and track the process from the source to the final formation of the data, so that the traceability of the data is formed to ensure the quality of the data.

4. Metadata construction: metadata refers to data used to describe entities and relationships related to metrology algorithms in enterprise carbons. Metadata construction refers to the discovery of attributes and events of an entity for guiding subsequent primary data identification.

5. Data disambiguation: the method based on the graph, the probability generation model and the topic model, and the method based on the deep learning identifies the data entity and the association relation. And extracting the entities in the data and linking the entities.

6. And (3) main data identification: and (3) comparing and judging whether the data is carbon neutral or not and identifying core data in a measurement algorithm through analysis of the data and the data chain. If identified, the data enters the master data management to construct a carbon neutral metric algorithm.

The data fusion scoring module uses two main data of the whole life cycle carbon management data of the product and the whole life cycle carbon management data of the organization, and calculates an enterprise carbon neutralization evaluation base table and each classification scoring item by combining a carbon neutralization measurement algorithm and a carbon neutralization quantitative scoring model.

The carbon neutralization measurement algorithm is applied to classification scoring, and is characterized in that the key information in the extracted and collected evaluation basis information is matched with the evaluation description in the attached table 1 and the key words, and the classification model is used for scoring and classifying for multiple times through qualitative and quantitative evaluation results of each dimension defined in the table to form a scoring result of carbon neutralization expression.

The carbon neutralization quantization scoring model is used in the total scoring, defines the weight of indexes of each dimension referring to the attached table 1 in the scoring, and is a weight matrix formed by splicing a plurality of groups of weight vectors, wherein the weight matrix is formed by integrating massive carbon emission data operated for years by ancestor science and technology, carbon neutralization methods in various carbon neutralization stages and the measurement of the results of the execution of the adopted carbon neutralization methods. The relevance and complexity of the enterprise internal data are considered in the carbon neutralization quantization scoring model, and in order to avoid the influence of extreme data and to cover the mutual influence of indexes under different conditions, a weight matrix is formed by splicing a plurality of groups of weight vectors, so that the objectivity caused by using a single group of weights is greatly avoided.

The data calculation module can calculate a final scoring table of enterprise carbon neutralization performance to determine the current carbon neutralization capacity of an enterprise based on the treated main data (including product full life cycle carbon management data and organization full life cycle carbon management data) and the ancestral carbon neutralization measurement algorithm and the carbon neutralization quantitative scoring model in the platform, can be used for analyzing and predicting carbon data, and provides optimization suggestions for the carbon neutralization direction of the enterprise.

The data calculation module comprises:

multiple vector matrix calculation: and the multiple vector matrix algorithm is used for enterprise carbon neutralization evaluation, and various carbon related index data of an enterprise are spliced and multiplied by the weight matrix to obtain the average number, standard deviation and score interval of the score index, so as to evaluate the enterprise carbon neutralization capability. The algorithm is applied to the total scoring step in the patent evaluation model.

Cyclic neural network (recurrent neural network, RNN) model deduction based on natural language processing (Natural language processing, NLP) techniques: and (3) carrying out semantic extraction by using an RNN-based model, extracting key information in the collected evaluation basis information, and scoring an evaluation base table through a carbon neutralization measurement algorithm. The algorithm is applied to the classification scoring step in the patent evaluation model. Meanwhile, the model is also used for analyzing and predicting carbon emission, recycling data to evaluate the improvement and help of various carbon neutralization optimization methods to the carbon neutralization measurement algorithm of the enterprise, and the model is used for selecting a better method to help the enterprise to achieve the carbon neutralization target early.

The evaluation model and the platform are based on mass data, a deep learning model constructed by cases and actual combat experience, and the best practice suitable for improving the carbon neutralization performance effective method of the enterprise is deduced, so that the enterprise is assisted to determine the carbon neutralization improvement direction. Meanwhile, through data feedback of the whole life cycle of the enterprise product, further evaluation and deduction are carried out regularly, and a continuously-lifted closed loop is constructed.

Table 1 is a detailed table of evaluation methods and systems for the full life cycle carbon neutralization performance of the manufacturing enterprise

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

The invention and its embodiments have been described above with no limitation, and the actual construction is not limited to the embodiments of the invention as shown throughout. In summary, if one of ordinary skill in the art is informed by this disclosure, a structural manner and an embodiment similar to the technical solution should not be creatively devised without departing from the gist of the present invention.

Thanks for the team work and contribution from Yunjin Tong and Shengtao Gao at CarbonSense Technologies.