CN110544015A - Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis - Google Patents
Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis Download PDFInfo
- Publication number
- CN110544015A CN110544015A CN201910732016.4A CN201910732016A CN110544015A CN 110544015 A CN110544015 A CN 110544015A CN 201910732016 A CN201910732016 A CN 201910732016A CN 110544015 A CN110544015 A CN 110544015A
- Authority
- CN
- China
- Prior art keywords
- carbon
- data
- module
- carbon emission
- enterprise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/84—Greenhouse gas [GHG] management systems
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a carbon data or carbon asset intelligent management and control platform, which comprises: a host computer, wherein the host computer comprises a plurality of servers, and the servers are preferably an application server and a database server; a monitoring module connected with the plurality of servers; a base data maintenance module that includes collecting and processing information related to the carbon data; the carbon emission calculation module mainly comprises the steps of obtaining carbon emission basic data, correspondingly determining a carbon emission value according to the obtained basic data, and calculating the whole carbon emission; and the prediction module is connected with the basic data maintenance module and the carbon emission calculation module, and analyzes and calculates the data by adopting a neural network prediction model and a multivariate linear regression prediction model based on principal component analysis. The invention provides effective support for enterprise carbon emission management from the aspects of carbon data integration, carbon emission/asset business support, mining and analysis of data and the like.
Description
Technical Field
The invention relates to the field of environmental protection, in particular to an enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis.
Background
In 2013, carbon transaction trial work is firstly carried out in 7 provinces and cities in China, and then national carbon transaction is gradually promoted. At present, most enterprises in China pay low attention to the management of carbon emission and carbon assets, most of enterprise carbon inventory check work is still based on the existing energy consumption ledger to carry out data manual input, and a small number of enterprises carry out the check of carbon emission conditions according to the original 'energy monitoring platform'. Enterprises can hardly guarantee the accuracy, integrity and real-time performance of data, so that most enterprises can hardly master the conditions of carbon emission and carbon assets completely, even some enterprises do not know how to calculate and check the carbon emission and how to trade the carbon assets, and professional knowledge, carbon trading value, policy and regulation and market conditions are not known.
For the key carbon emission industry, the national carbon market is both opportunistic and challenging, and enterprises face two options: firstly, passive incorporation results in higher performance cost; and secondly, actively participating in improving the carbon asset management capability of the enterprise and preempting the market opportunity. Due to the current state of enterprises and the current state of technology, most enterprises are difficult to comprehensively master and accurately support data for managing carbon emission and carbon assets, and the potential value of carbon transaction cannot be mined.
Therefore, it is necessary to build an efficient intelligent management and control platform for carbon data or carbon assets of an enterprise.
disclosure of Invention
The invention aims to provide an enterprise carbon data or carbon asset intelligent management and control platform which can realize automatic data acquisition, automatic carbon emission calculation, real-time policy and regulation pushing, real-time carbon transaction price query, real-time carbon asset mastering, carbon emission trend analysis and carbon emission prediction, so that an enterprise can better know the carbon emission and carbon asset conditions, the development of the era and the market demand are met, and the maximum value of the carbon asset is excited.
By analyzing the traditional technical background and the problems existing in the traditional technology, the invention solves the problems of carbon data and carbon asset management of enterprises from a data level, a business support level and a technical level.
and (3) data layer: the method has the advantages that a set of enterprise carbon number data management database is established through an enterprise carbon data or carbon asset intelligent control platform, the acquisition, storage, management and use of enterprise carbon emission and energy data are realized, relevant data can be timely, accurately and efficiently provided, and data support is provided for enterprise carbon emission management business and data mining analysis.
And (3) service layer: 1. the method is characterized in that a set of enterprise carbon data or carbon asset intelligent management and control platform meeting the requirements of various industries is established by combining the requirements of various industries on carbon asset management business, effective support is provided for carbon emission management of enterprises from several aspects of carbon number data acquisition, carbon emission real-time accounting, carbon asset management and carbon number data mining and analysis, the carbon emission condition of the enterprises is mastered in real time, and effective analysis decision is carried out through the carbon data. 2. Aiming at the current situation that the carbon emission management knowledge of enterprises is deficient at present, the system embeds the carbon emission accounting guideline calculation formula of each industry of China into the system, supports flexible configuration, can realize the carbon emission calculation of each industry, and provides an effective tool for the carbon emission self-check of the enterprises. 3. At present, the carbon transaction pilot work is firstly carried out in 7 provinces, and the carbon asset and carbon transaction management is imperative. 4. Most enterprises have insufficient knowledge of the potential value of the carbon emission data, and few enterprises utilize historical data to analyze and mine. The system is based on enterprise historical data, combines a neural network and a regression analysis algorithm, builds an enterprise carbon emission prediction model, realizes the predictability of the carbon emission of the enterprise, plays a certain guiding role in future carbon emission management and carbon transaction of the enterprise, and is convenient for the enterprise to deal with the changes of markets and policies.
the technical level is as follows: 1. the system establishes a standardized data coding standard and an interface specification in a technical level, finally forms a platform and an application mode which can be popularized and copied, and has stronger universality.
The invention provides a carbon data or carbon asset intelligent management and control platform, which comprises: a host computer, wherein the host computer comprises a plurality of servers, and the servers are preferably an application server and a database server; a monitoring module connected with the plurality of servers; a base data maintenance module that includes collecting and processing information related to the carbon data; the carbon emission calculation module mainly comprises the steps of obtaining carbon emission basic data, correspondingly determining a carbon emission value according to the obtained basic data, and calculating the whole carbon emission; and the prediction module is connected with the basic data maintenance module and the carbon emission calculation module, and analyzes and calculates the data by adopting a neural network prediction model and a multivariate linear regression prediction model based on principal component analysis.
the host comprises a master control system and an enterprise subsystem; the main control system is connected with each enterprise subsystem through a communication network, and an application server and a database server of the host are respectively deployed in a main control unit and a pilot plant enterprise.
The monitoring module comprises a comprehensive display submodule and a carbon flow diagram display submodule.
The basic data maintenance module comprises a general configuration submodule and an enterprise configuration submodule.
Wherein preferably the carbon emission module is connected to a monitoring module.
The system further comprises a data statistics module, wherein the data statistics module is mainly used for carrying out statistics on carbon number data and energy data.
The carbon trading price module mainly comprises a carbon trading price and carbon trading management.
Wherein, further comprises a carbon asset management module.
The intelligent management and control platform for the carbon data or the carbon assets of the enterprise provides effective support for the carbon emission management of the enterprise from the aspects of carbon data integration, carbon emission/asset business support, data mining and analysis and the like. The method can uniformly manage and store the carbon data on a data level, check the carbon emission data in time and further mine the potential value of the data; the method is characterized in that accounting is carried out on the carbon emission condition of an enterprise in a business layer, a set of complete enterprise carbon emission management system is formed, related contents of carbon emission and carbon asset management are comprehensively covered, the carbon emission condition of the enterprise is comprehensively reflected, business points such as carbon assets, carbon data statistical analysis and carbon emission prediction are involved, and the management of the carbon assets of the enterprise is centered. The technology can be popularized in different industries, and the technology has universality, practicability and reproducibility; and mining and analyzing the historical data, and predicting the carbon emission of the enterprise based on algorithms such as a neural network algorithm, regression analysis and the like.
drawings
FIG. 1 is a functional exploded view of a management and control platform according to the present invention;
FIG. 2 is a schematic connection diagram of the management and control platform of the present invention;
FIG. 3 is a computational logic diagram of a carbon emissions calculation module;
FIG. 4 is a graph of a sigmoid function;
Fig. 5BP network topology.
Detailed Description
For the purpose of facilitating an understanding of the present invention, reference will now be made to the following descriptions taken in conjunction with the accompanying drawings, and it will be understood by those skilled in the art that the following descriptions are provided for purposes of illustration and description, and are not intended to limit the scope of the invention.
fig. 1 is a schematic structural diagram of a carbon data or carbon asset intelligent management and control platform according to the present invention. The intelligent management and control platform analyzes the problems existing in the traditional technology background and the traditional technology and the purpose of the patent, and is designed and developed according with the actual conditions of enterprises, and the intelligent management and control platform is introduced below.
The intelligence management and control platform includes:
A host computer, wherein the host computer comprises a plurality of servers, and the servers are preferably an application server and a database server; the host comprises a master control system and an enterprise subsystem; the main control system is connected with each enterprise subsystem through a communication network, and an application server and a database server of the host are respectively deployed in a machine room of a main control unit and a machine room of a trial-and-error enterprise. The database server is mainly used for storing the converted and analyzed carbon and energy data, related yield and other data, and part of the functional modules are deployed in the application server;
And the monitoring module is connected with the servers and comprises a comprehensive display submodule and a carbon flow diagram display submodule. The comprehensive display sub-module mainly displays the main information concerned by the enterprise in a one-picture mode, so that management personnel can conveniently check the main information, and the comprehensively displayed data is provided by the database server. The display contents of the comprehensive display submodule can include, but are not limited to, carbon emission of the year, carbon emission of the month, carbon emission percentage of the year, carbon source distribution of the year, carbon emission trend of the year, yield and unit product energy consumption of the year, comprehensive energy consumption of the year, unit product energy consumption of the year, comprehensive energy consumption of key equipment of the year, carbon emission of nearly five years, carbon emission intensity of nearly five years, carbon trading price query, relevant policy information, carbon reduction service provider information and the like. The monitoring module makes full use of the collected data, the calculated data and the historical data to perform multidimensional statistical analysis, shows the trend change of the data and the quantity display through various data charts, and is convenient for users to know important information of carbon emission of enterprises, so that the efficiency of the enterprises in managing the carbon emission is improved. The carbon flow diagram submodule mainly realizes the display of an important process flow diagram of an enterprise and the flow direction display of carbon emission. The enterprise user can upload the process flow chart drawn offline in the enterprise data management module, and the foreground page can automatically call and display the process flow chart. The enterprise can clearly know the flow direction of the carbon of the whole enterprise, and the enterprise can control the whole process flow more conveniently. As a further implementation, the monitoring module may further include a display terminal, and the display terminal displays corresponding content according to the control.
The basic data maintenance module is used for collecting and processing information related to carbon data and mainly comprises a general configuration submodule and an enterprise configuration submodule, wherein the general configuration mainly comprises industry information configuration, material category configuration, carbon emission factor information configuration, activity data information configuration, calculation unit information configuration, carbon emission calculation formula configuration and carbon emission alarm configuration; the enterprise configuration mainly comprises enterprise basic information maintenance, enterprise emission sources, enterprise measured emission factors, enterprise data and policy information maintenance. Through the basic data maintenance function, the functions of increasing, deleting, changing and searching the carbon emission configuration and the enterprise basic information configuration can be realized, and the user can conveniently manage and maintain the basic data of the system.
the carbon emission calculation module mainly comprises the steps of obtaining carbon emission basic data, correspondingly determining a carbon emission value according to the obtained basic data, and calculating the whole carbon emission; and obtaining the carbon emission according to different algorithms and comparing the carbon emission with the carbon emission. By understanding the actual conditions of enterprises and national carbon emission accounting guidelines, there are currently guideline calculation methods, and the system supports the guideline calculation methods. The main data requirements are the average low calorific value of the emission source, the carbon content of a unit calorific value, the carbon oxidation rate, the emission factor and the consumption, and the carbon emission of an enterprise, a production process, a net purchasing power and heat, a carbon fixing process and the whole enterprise are calculated by combining the basic data according to a calculation method of a guideline.
the carbon emission calculation module belongs to the important functions of the management and control platform, the calculation logic is relatively complex, as shown in fig. 3, the calculation logic firstly obtains basic data of carbon emission and a carbon emission target value corresponding to each basic data, then selects a formula expression according to the carbon emission condition, calculates and obtains parameter values of different carbon emission conditions, obtains a calculation result and displays the result, preferably, the carbon emission module is connected with the monitoring module, and the basic data, the carbon emission related parameters, the carbon emission amount and the like of the carbon emission module can be displayed through the comprehensive display sub-module and the carbon flow diagram display sub-module of the monitoring module. The flexible configuration of a calculation formula on a foreground page can be realized through the carbon emission calculation configuration module; the carbon emission calculation module can realize automatic calculation and visual display of the carbon emission of enterprises; the carbon emission comparison module can realize comparison of carbon emission and carbon emission intensity under different accounting methods and multi-form display of calculation results; fully embodies the functional characteristics of universality, flexibility and practicability, and meets the requirement of enterprises on carbon emission calculation. As a further preference, a carbon emission data acquisition terminal may be included
And the data statistics module is mainly used for carrying out statistics on carbon number data and energy data. The carbon data statistical module mainly realizes that enterprises can perform statistical query on carbon emission and carbon emission intensity according to different time granularities and different categories. The energy statistics mainly comprises the statistics and query of comprehensive energy consumption and index energy consumption, and the data trend is visually displayed in a graphical mode. Meanwhile, the energy consumption condition of key equipment of an enterprise can be counted, and the operation condition of the key equipment of the enterprise is shown in a chart.
the carbon asset management module mainly comprises carbon asset personnel management, carbon asset document management, emission reduction project management, quota and performance. The management of the carbon asset personnel is mainly the management of the carbon asset personnel in the enterprise, and can comprehensively master the basic information of related personnel, including the unit, name, division (clear group leader, group member), department, contact telephone, responsibility and management content of the personnel. The carbon asset document management is mainly used for managing related documents of the carbon assets of the enterprises, can realize the uploading and downloading of the related documents and is convenient for users to manage the carbon asset documents of the enterprises. Emission reduction project management is mainly used for managing enterprise emission reduction projects, including the name, the starting time, the ending time, the transformation content, the cost, project responsible persons, the operation effect, the emission reduction amount, whether the project is a CCER project or not and whether carbon assets are formed or not. The quota and the performance are mainly used for maintaining the carbon quota performance of the current year by the enterprise, and the carbon emission performance of the current year is more intuitively understood through a chart.
And the carbon transaction price module mainly comprises carbon transaction price and carbon transaction management. The carbon transaction management realizes management of carbon transaction information of enterprises in a per year fulfillment period, and comprises enterprise names, fulfillment years, transaction types, transaction amounts, transaction prices, transaction time, dealers and remark information. The carbon trading price mainly acquires the carbon trading price information of the regional websites of the enterprises at regular time every day through a crawler technology, and the carbon trading price information is inquired in the module.
The prediction module is an enterprise carbon emission prediction model which is constructed based on enterprise historical data and combined with enterprise production process flow, carbon emission conditions and energy consumption conditions, the prediction model is used as a most core and most innovative functional model of the system, preferably, the prediction module is connected with a basic data maintenance module and a carbon emission calculation module, the prediction module takes data of the database server as an analysis basis, a neural network prediction model and a multivariate linear regression prediction model based on principal component analysis are mainly adopted to analyze and calculate the data of the platform, the neural network model has strong nonlinear fitting capacity, can map any complex nonlinear relation and strong self-learning capacity, and is simple in learning rule and convenient to realize by a computer. The regression analysis is a mainstream prediction method, and has the effects of accurately measuring the correlation degree between all factors and the regression fitting degree and improving the prediction equation. The platform of the invention simultaneously uses two prediction models, thus the user can have certain selectivity, and meanwhile, the comparison between prediction results can increase the reliability of prediction. The derivation processes of the two prediction models are respectively described as follows:
(1) multi-layer neural network BP algorithm derivation
output in the perceptron training law:
since the sign function is a non-continuous function, which makes it non-trivial, the above gradient descent algorithm cannot be used to minimize the loss function.
The output in the incremental rule is:
Each output is a linear combination of inputs, so that when a plurality of linear units are connected together, only linear combinations of inputs can be obtained finally, which is not different from the situation that only one sensor unit node is used.
In order to solve the above problems, on one hand, we cannot directly output in a linear combination mode, and need to add a processing function at the time of output; on the other hand, the added processing function must be differentiable, so that we can use the gradient descent algorithm.
The functions that satisfy the above conditions are very many, but the most classical momorid function is also called Logistic function, which can compress any number within (+ ∞, and infinity) to between (0,1), and thus is also called squeeze function. To further normalize the input to this function, we add a threshold to the linear combination of inputs, such that the linear combination of inputs is bounded by 0.
sigmoid function:
The function curve is shown in fig. 4.
An important characteristic of this function is its derivative:
with this feature, it is much easier to calculate its gradient descent.
In addition, a hyperbolic function tanh can also be used for replacing the sigmoid function, and the graphs of the hyperbolic function tanh and the sigmoid function are similar.
the Back Propagation algorithm is also called BP algorithm (Back Propagation). We can now build a multi-layer neural network with the above-described perceptron using sigmoid functions, where we use a three-layer network for analysis for simplicity. Assume that the network topology is as shown in fig. 5.
the operation flow of the network is as follows: when a sample is input, obtaining a characteristic vector of the sample, obtaining an input value of a sensor according to the weight vector, calculating the output of each sensor by using a sigmoid function, taking the output as the input of the sensor of the next layer, and repeating the steps until the output layer.
How is the weight vector for each perceptron determined? We then need to use a back-propagation algorithm to perform the optimization step by step. We proceed with the analysis before formally introducing the back propagation algorithm.
To obtain the weight vector, we constantly adjust the weight vector by minimizing the loss function. The method is also suitable for solving the weight vector, firstly, a loss function needs to be defined, and since the output layer of the network has a plurality of output nodes, the square sum of the difference value of each output node of the output layer needs to be obtained. The loss function for each training example is then obtained as: (the front is added with 0.5 to facilitate the use of the back seeking guidance)
In a multi-layer neural network, the error surface may have multiple local minima, which means that what is found using the gradient descent algorithm may be a local minimum, rather than a global minimum.
Now we have a loss function, and we can adjust the input weight vector in the output node according to the loss function, which is similar to the stochastic gradient descent algorithm in the perceptron, and then adjust the weight layer by layer from back to front, which is the idea of the back propagation algorithm.
Back propagation algorithm for feedforward network with two layers of sigmoid units:
all weights in the network are initialized randomly. For each training example, the following operations are performed:
A) And calculating sequentially from front to back according to the input of the example to obtain the output of each unit of the output layer. The error term for each cell of each layer is then calculated back starting from the output layer.
B) For each cell k of the output layer, its error term is calculated:
δ＝o(1-o)(t-o)
C) for each hidden unit h in the network, its error term is calculated:
D) Updating each weight:
w＝W+ρδx
Δ wji ═ ρ δ jxji is referred to as the weight update rule. Description of the symbols:
xji: the input of node i to node j, wji, represents the corresponding weight.
outputs: representing the set of output level nodes.
the whole algorithm is similar to the random gradient descent algorithm of delta rule, and the algorithm is analyzed as follows:
the updating aspect of the weight value is similar to the delta rule, and mainly depends on the learning rate, the input corresponding to the weight value and the error item of the unit.
② for the output layer unit, its error term is (t-o) multiplied by the derivative ok (1-ok) of the sigmoid function, which is different from the error term of delta rule, which is (t-o).
For the hidden layer unit, because a direct target value is lacked to calculate the error of the hidden unit, the error term δ k of the hidden layer needs to be calculated in an indirect manner to perform weighted summation on the error δ k of each unit affected by the hidden unit h, the weight of each error δ k is wkh, and wkh is the weight from the hidden unit h to the output unit k.
derivation of the back propagation algorithm. The derivation process of the algorithm is mainly a process of minimizing a loss function by using a gradient descent algorithm, and the loss function is now:
for each weight wji in the network, its derivative is calculated:
If j is the output layer unit of the network
Derivation of netj:
Wherein:
therefore, the method comprises the following steps:
to make the expression concise, we use:
The weight changes towards the negative gradient of the loss function, so there is a weight change:
If j is a hidden unit in the network
since the value of w in the hidden unit indirectly affects the input through the next layer, the derivation is done using a layer-by-layer peeling approach:
Because:
Therefore:
also, we use:
Therefore, the weight variation is:
By utilizing the algorithm, the data in the carbon asset management and control platform can be calculated and simulated, and the predicted trend can be obtained
And (5) improving the algorithm. The application of back-propagation algorithms is very widespread, and many different variants have been created in order to meet various requirements, two variants being described below:
Increasing impulse term
the method mainly modifies the weight updating rule. The main idea is to make the updating part of the weight value in the nth iteration depend on the weight value of the (n-1) th iteration.
△w(n)＝ρδx+αΔw(n-1)
Wherein 0< ═ α < 1: called the coefficient of impulse. The addition of the impulse term has the effect of increasing the search step length to a certain extent, so that the convergence can be performed more quickly. On the other hand, the multilayer network is easy to cause the loss function to converge to a local minimum value, but the impulse term can cross some narrow local minimum values to a smaller place to some extent.
② learning arbitrary depth ring-free network
In the case where the above-described back propagation algorithm has only three layers, i.e. only one hidden layer, if there are many hidden layers that should be handled?
now, suppose that the neural network has m +2 layers, i.e. m hidden layers. In this case, only one place needs to be changed to obtain the back propagation algorithm with m hidden layers. The value of the error of the unit r of the k-th layer is calculated from the error term of the deeper k + 1-th layer:
(2) multivariate linear regression model derivation
for sample data of n-dimensional features, if we decide to use linear regression, the corresponding model is such that:
h(x，x，…，x)＝θ+θx+…+θx
this representation can be simplified by adding a feature x0 equal to 1, which is further expressed in a matrix form as follows:
Further expressed in matrix form, it is more concise as follows:
h(X)＝X·θ
the h theta (X) is assumed to be a vector of mx1, theta is assumed to be a vector of nx1, and n algebraic model parameters are arranged inside the theta (X). X is a matrix of m X n dimensions. m represents the number of samples, and n represents the number of features of the samples. Obtaining a model, we need to find the required loss function, and generally linear regression we use the mean square error as the loss function. The algebraic representation of the loss function is as follows:
The loss function is further expressed in matrix form:
There are two methods commonly used by us to find the θ parameter when the loss function is minimized: one is a gradient descent method and the other is a least square method. If a gradient descent method is used, the iterative formula for θ is such that:
θ＝θ-αX(Xθ-Y)
If the least squares method is used, the resulting equation for θ is as follows:
θ＝(XX)XY
through the algorithm, each change parameter value in the asset management and control platform can be preferably used as a factor by combining various factors related to carbon number data, each factor is used as sample data in a neural network, the sample data is substituted into the function derivation process, and then the related carbon emission information in the future time period can be deduced, the operation function of a prediction module is exerted, and different prediction algorithms are adopted for comparison and verification.
The invention adopts general information configuration and enterprise information configuration, thus realizing the flexibility and easy maintenance of the system; the data storage adopts the coding rule of the national energy standard and the standardized interface specification, so that the system has better expansibility; the carbon emission comprehensive monitoring module can intensively display visual charts of different information, so that a user can conveniently know key information of carbon emission of an enterprise, and the carbon emission management level of the enterprise is improved; through policy and regulation and a carbon trading price function, an enterprise can know national policies and price information in time, and the enterprise can be helped to know market changes in real time; through statistical analysis of system data, multidimensional statistics and visual display of carbon number data can be realized, and potential values of the data are fully reflected; due to the establishment of the enterprise carbon emission prediction model, the system has certain predictability on future carbon emission, and meanwhile, the prediction result has higher precision on the basis of a large amount of historical data and continuous optimization of the model.
It will be understood by those skilled in the art that many possible variations and modifications, or equivalent embodiments, may be made to the present invention using the teachings disclosed above without departing from the scope of the present invention. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention are still within the scope of the protection of the technical solution of the present invention, unless the contents of the technical solution of the present invention are departed.
Claims (8)
1. A carbon data or carbon asset intelligent management and control platform, comprising: a host computer, wherein the host computer comprises a plurality of servers, and the servers are preferably an application server and a database server; a monitoring module connected with the plurality of servers; a base data maintenance module that includes collecting and processing information related to the carbon data; the carbon emission calculation module mainly comprises the steps of obtaining carbon emission basic data, correspondingly determining a carbon emission value according to the obtained basic data, and calculating the whole carbon emission; and the prediction module is connected with the basic data maintenance module and the carbon emission calculation module, and analyzes and calculates the data by adopting a neural network prediction model and a multivariate linear regression prediction model based on principal component analysis.
2. A carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: the host comprises a master control system and an enterprise subsystem; the main control system is connected with each enterprise subsystem through a communication network, and an application server and a database server of the host are respectively deployed in a main control unit and a pilot plant enterprise.
3. A carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: the monitoring module comprises a comprehensive display sub-module and a carbon flow diagram display sub-module.
4. a carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: the basic data maintenance module comprises a general configuration submodule and an enterprise configuration submodule.
5. a carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: preferably, the carbon emission module is connected to a monitoring module.
6. A carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: the system further comprises a data statistics module, wherein the data statistics module is mainly used for carrying out statistics on carbon number data and energy data.
7. a carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: and the carbon transaction price module mainly comprises carbon transaction price and carbon transaction management.
8. A carbon data or carbon asset intelligent management and control platform according to claim 1, wherein: further comprising a carbon asset management module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910732016.4A CN110544015A (en) | 2019-08-09 | 2019-08-09 | Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910732016.4A CN110544015A (en) | 2019-08-09 | 2019-08-09 | Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110544015A true CN110544015A (en) | 2019-12-06 |
Family
ID=68710544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910732016.4A Pending CN110544015A (en) | 2019-08-09 | 2019-08-09 | Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110544015A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113094580A (en) * | 2021-03-24 | 2021-07-09 | 远光软件股份有限公司 | Carbon quota calculation method and device, storage medium and terminal equipment |
| CN113219932A (en) * | 2021-06-02 | 2021-08-06 | 内蒙古自治区计量测试研究院 | Digital analytic system of thermal power trade carbon emission |
| CN113592187A (en) * | 2021-08-06 | 2021-11-02 | 时代云英(重庆)科技有限公司 | Intelligent carbon emission management system and method |
| CN113672663A (en) * | 2021-08-13 | 2021-11-19 | 国网(衢州)综合能源服务有限公司 | Industrial enterprise carbon account system |
| CN113962468A (en) * | 2021-10-29 | 2022-01-21 | 杭州青橄榄网络技术有限公司 | Energy consumption monitoring and statistics-based energy consumption carbon emission management method and system |
| CN114037586A (en) * | 2022-01-07 | 2022-02-11 | 阿里云计算有限公司 | Carbon peak path deduction method, electronic device and storage medium |
| CN114548481A (en) * | 2021-12-26 | 2022-05-27 | 特斯联科技集团有限公司 | Power equipment carbon neutralization processing apparatus based on reinforcement learning |
| CN115759788A (en) * | 2023-01-05 | 2023-03-07 | 碳阻迹（北京）科技有限公司 | Enterprise carbon data comprehensive intelligent management and control system based on big data analysis |
| CN115984069A (en) * | 2022-12-30 | 2023-04-18 | 数字双碳科技(合肥)有限公司 | Carbon emission data processing and analyzing method based on carbon metering edge all-in-one machine |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260836A (en) * | 2015-10-13 | 2016-01-20 | 天津大学 | Automobile manufacture enterprise carbon emission acquisition checking system and method |
| CN109272405A (en) * | 2018-09-30 | 2019-01-25 | 大唐碳资产有限公司 | Carbon transaction in assets method and system |
| CN109740301A (en) * | 2019-03-14 | 2019-05-10 | 华北电力大学 | A kind of accounting method of the Gas Generator Set carbon emission amount based on BP neural network |
| CN110097279A (en) * | 2019-04-29 | 2019-08-06 | 兰州环境能源交易中心有限公司 | The carbon asset management system, enterprise |
-
2019
- 2019-08-09 CN CN201910732016.4A patent/CN110544015A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260836A (en) * | 2015-10-13 | 2016-01-20 | 天津大学 | Automobile manufacture enterprise carbon emission acquisition checking system and method |
| CN109272405A (en) * | 2018-09-30 | 2019-01-25 | 大唐碳资产有限公司 | Carbon transaction in assets method and system |
| CN109740301A (en) * | 2019-03-14 | 2019-05-10 | 华北电力大学 | A kind of accounting method of the Gas Generator Set carbon emission amount based on BP neural network |
| CN110097279A (en) * | 2019-04-29 | 2019-08-06 | 兰州环境能源交易中心有限公司 | The carbon asset management system, enterprise |
Non-Patent Citations (2)
| Title |
|---|
| 张虹: "《辽宁老工业基地住宅产业可持续发展与创新研究》", 30 November 2013 * |
| 郭科: "《2015年中国管理科学与工程研究报告》", 31 October 2015 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113094580A (en) * | 2021-03-24 | 2021-07-09 | 远光软件股份有限公司 | Carbon quota calculation method and device, storage medium and terminal equipment |
| CN113219932A (en) * | 2021-06-02 | 2021-08-06 | 内蒙古自治区计量测试研究院 | Digital analytic system of thermal power trade carbon emission |
| CN113219932B (en) * | 2021-06-02 | 2023-09-05 | 内蒙古自治区计量测试研究院 | Digital analysis system for carbon emission in thermal power generation industry |
| CN113592187A (en) * | 2021-08-06 | 2021-11-02 | 时代云英(重庆)科技有限公司 | Intelligent carbon emission management system and method |
| CN113592187B (en) * | 2021-08-06 | 2024-01-05 | 时代云英(深圳)科技有限公司 | Intelligent carbon emission management system and method |
| CN113672663B (en) * | 2021-08-13 | 2023-07-25 | 国网(衢州)综合能源服务有限公司 | Industrial enterprise carbon account system |
| CN113672663A (en) * | 2021-08-13 | 2021-11-19 | 国网(衢州)综合能源服务有限公司 | Industrial enterprise carbon account system |
| CN113962468A (en) * | 2021-10-29 | 2022-01-21 | 杭州青橄榄网络技术有限公司 | Energy consumption monitoring and statistics-based energy consumption carbon emission management method and system |
| CN114548481A (en) * | 2021-12-26 | 2022-05-27 | 特斯联科技集团有限公司 | Power equipment carbon neutralization processing apparatus based on reinforcement learning |
| CN114037586A (en) * | 2022-01-07 | 2022-02-11 | 阿里云计算有限公司 | Carbon peak path deduction method, electronic device and storage medium |
| CN115984069A (en) * | 2022-12-30 | 2023-04-18 | 数字双碳科技(合肥)有限公司 | Carbon emission data processing and analyzing method based on carbon metering edge all-in-one machine |
| CN115984069B (en) * | 2022-12-30 | 2023-09-05 | 数字双碳科技(合肥)有限公司 | Carbon emission data processing and analyzing method based on carbon metering edge all-in-one machine |
| CN115759788A (en) * | 2023-01-05 | 2023-03-07 | 碳阻迹（北京）科技有限公司 | Enterprise carbon data comprehensive intelligent management and control system based on big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110544015A (en) | Enterprise carbon data or carbon asset intelligent management and control platform based on big data analysis | |
| Mi et al. | An integrated approach to multiple criteria decision making based on the average solution and normalized weights of criteria deduced by the hesitant fuzzy best worst method | |
| Chatterjee et al. | A two-phase fuzzy AHP-fuzzy TOPSIS model for supplier evaluation in manufacturing environment | |
| Dev et al. | Multi-criteria evaluation of real-time key performance indicators of supply chain with consideration of big data architecture | |
| Keshavarz Ghorabaee et al. | Stochastic EDAS method for multi-criteria decision-making with normally distributed data | |
| Kabir et al. | Multi-criteria inventory classification through integration of fuzzy analytic hierarchy process and artificial neural network | |
| Bayram et al. | Comparison of multi layer perceptron (MLP) and radial basis function (RBF) for construction cost estimation: the case of Turkey | |
| Owolabi et al. | Predicting completion risk in PPP projects using big data analytics | |
| Baloui Jamkhaneh et al. | Inspection error and its effects on single sampling plans with fuzzy parameters | |
| Chu et al. | A manufacturing resource allocation method with knowledge-based fuzzy comprehensive evaluation for aircraft structural parts | |
| Boateng et al. | A dynamic framework for managing the complexities of risks in megaprojects | |
| Arredondo et al. | Learning and adaptation of a policy for dynamic order acceptance in make-to-order manufacturing | |
| CN111882157A (en) | Demand prediction method and system based on deep space-time neural network and computer readable storage medium | |
| Xu et al. | Integrated optimisation for production capacity, raw material ordering and production planning under time and quantity uncertainties based on two case studies | |
| Taghizadeh | Utilizing artificial neural networks to predict demand for weather-sensitive products at retail stores | |
| Petkovic et al. | Deep learning for spatio‐temporal supply and demand forecasting in natural gas transmission networks | |
| Li et al. | Machine learning algorithm generated sales prediction for inventory optimization in cross-border E-commerce | |
| Chargui et al. | Coupling the ILS optimisation algorithm and a simulation process to solve the travelling quay-crane worker assignment and balancing problem | |
| CN114154716A (en) | Enterprise energy consumption prediction method and device based on graph neural network | |
| Datta et al. | Comparative study on application of utility concept and VIKOR method for vendor selection | |
| Azadeh et al. | Selecting optimum maintenance activity plans by a unique simulation-multivariate approach | |
| Hassan et al. | A decision support system for subjective forecasting of new product sales | |
| Jalalvand et al. | A multi-objective risk-averse workforce planning under uncertainty | |
| Liu et al. | Application of fuzzy ordered weighted geometric averaging (FOWGA) operator for project delivery system decision-making | |
| Biyeme et al. | An analytical model for analyzing the value of information flow in the production chain model using regression algorithms and neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |