Project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by using electric energy

Abstract

The invention relates to the technical field of energy management, in particular to a high-frequency project-level energy consumption monitoring model for calculating carbon by using electric energy. Comprising the following specific steps: (1) a matching technology based on multi-dimensional data keyword extraction; (2) a decision tree-based project power consumption subject classification algorithm; (3) a key project energy consumption progress model based on a regression algorithm; (4) an energy consumption monitoring model based on multiple early warning states; and (5) an electricity-energy-carbon analysis model based on an ARDL algorithm. The invention provides a high-frequency project-level energy consumption monitoring model for calculating energy and carbon by using power, and precise measurement and real-time monitoring of project energy consumption are realized through technical means such as multi-dimensional data fusion, decision tree classification, a regression algorithm, multi-early warning state monitoring and an ARDL algorithm. The model can significantly reduce the manpower and time cost of energy consumption supervision, improves the supervision efficiency, and provides powerful energy consumption management support for the government and enterprises.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and in particular to a project-level energy consumption monitoring model that realizes high-frequency calculation of energy and carbon by electricity. Background Art

[0002] Against the backdrop of the accelerated advancement of new industrialization, informatization, urbanization, and agricultural modernization, there are still many challenges in achieving a comprehensive green transformation. In the actual implementation process, energy consumption supervision has encountered the following specific problems:

[0003] 1. On-site inspection tasks are arduous and manual verification is difficult:

[0004] With the continuous development of the economy, the number of energy-saving review projects approved each year increases year by year, resulting in a large amount of project data that front-line staff need to collect and verify, and the inspection and verification tasks are extremely heavy.

[0005] The projects are distributed throughout the province, and the construction and operation of the projects begin after approval, involving multiple links such as bidding, construction, completion and commissioning. There are many uncertainties in the operation, which further increases the difficulty of manual verification.

[0006] 2. The project data type is complex and energy consumption calculation is difficult:

[0007] The number of projects is increasing, covering a wide range of industries and a wide variety of energy types.

[0008] Different industries have different reporting situations for energy consumption types. There are also differences in reported energy types and energy consumption due to factors such as differences in process levels or size of construction units in the same industry.

[0009] The approved annual design comprehensive energy consumption among various projects varies greatly, and the difference in energy consumption volume can reach thousands of times, making the scientific and efficient completion of energy consumption measurement and supervision of various projects one of the issues that need to be urgently addressed.

[0010] 3. Lack of timeliness in energy consumption supervision and insufficient real-time perception capabilities:

[0011] After the energy-saving review and approval authority was decentralized to local levels, the supervision of energy-saving projects has always lacked effective means due to factors such as the low degree of automation of the energy metering system and limited development conditions in some cities and prefectures.

[0012] The professional capabilities of grassroots regulatory forces are limited, and they can only rely on regular inspections every year for post-management, which makes it difficult to prevent the risk of excessive energy consumption in projects in advance.

[0013] More convenient and efficient regulatory means are needed to assist energy consumption supervision work to move from post-event to in-event management, and to achieve the transformation from scheduled passive inspections to active intervention supervision.

[0014] In summary, the current energy consumption supervision work faces specific problems such as heavy on-site inspection tasks, complex project data types, and lack of timeliness in energy consumption supervision. In order to solve these problems, the present invention proposes a project-level energy consumption monitoring model that realizes high-frequency calculation of energy and carbon by electricity, and carries out key technical research such as electricity subject matching, key project type identification, energy consumption progress model, energy consumption monitoring and early warning, energy-electricity-carbon calculation, etc., aiming to build an electricity subject identification framework, key project carbon emission measurement and early warning model system, so as to maximize the value of energy big data mining and realize accurate control of energy and carbon emissions of key projects. Summary of the invention

[0015] The technical problems solved by the present invention are that the on-site inspection tasks are heavy, manual verification is difficult, the project data types are complex, the energy consumption calculation is difficult, the energy consumption supervision lacks timeliness, and the real-time perception capability is insufficient.

[0016] The technical solution adopted by the present invention is: a project-level energy consumption monitoring model that realizes high-frequency calculation of energy and carbon by electricity, including the following specific steps:

[0017] (1) A matching technology based on keyword extraction from multidimensional data is used to perform preliminary management and cleaning of multidimensional data. The monitoring subjects of each project are preliminarily screened based on the keyword extraction algorithm of statistical characteristics. A directional task script is established to match the multidimensional data to obtain a preliminary project list. The preliminary project list includes "matched projects" and "projects to be verified". After manual screening, a complete list of project electricity monitoring subjects is obtained, and the accurate verification matching rate reaches 90% or above;

[0018] (2) A decision tree-based classification algorithm for project electricity users, which is used to define and identify matched projects, and to classify single / comprehensive projects according to whether they apply for business expansion services or install additional monitoring and collection devices. By using internal and external data cross-analysis technology and multi-data source residual noise and abnormal information elimination methods, relevant information on project electricity users is obtained, and the identification and classification of project electricity users is completed, resulting in a set of energy-saving review project monitoring lists that meet the algorithm requirements;

[0019] (3) A key project energy consumption progress model based on a regression algorithm, which is used to calculate the comprehensive energy consumption of the project based on the approved energy consumption data of the energy-saving review project and the annual production and operation energy consumption data, and calculate the comprehensive energy consumption of the project with the approved energy consumption value to obtain the energy consumption progress of the project;

[0020] (4) An energy consumption monitoring model based on multiple early warning states is used to calculate the historical monthly electricity consumption of the project, the project energy approval data, and the energy conversion coefficient to obtain the historical monthly comprehensive energy consumption, and then divide it by the annual approved value to obtain the monthly cumulative energy consumption early warning value. A static / dynamic energy consumption monitoring model is further constructed to obtain the monthly status and static early warning results of the project, as well as the annual cumulative status and dynamic early warning results;

[0021] (5) An electricity-energy-carbon analysis model based on the ARDL algorithm is used to further integrate the electricity consumption data collected in the project electricity user identification module with influencing factors such as energy consumption type, and use the ARDL model to perform an autoregressive distributed lag model of the time series to achieve accurate calculation of project-level carbon emissions and carbon footprints.

[0022] As a further solution of the present invention: the relevant information is the power user file, the user's business expansion application, the project bidding and construction status to determine whether the metering point ID is correct, the construction and operation status, and the project approval time.

[0023] As a further solution of the present invention: the electricity user identification and determination process:

[0024] Start from the "Start" stage, first, enter the "Verify Successful Project" stage;

[0025] Next, the process enters a decision point, namely "whether the user is a non-temporary user of electricity". The system or operator needs to determine whether the user associated with the project is a non-temporary user of electricity. If the user is determined to be a non-temporary user of electricity, the process continues to proceed downward. If the user is determined to be a temporary user of electricity, other processing processes may be triggered.

[0026] For non-temporary electricity users, the process enters the "project approval time" stage; at this stage, the system or operator needs to pay attention to the project approval time and check the following three conditions in turn:

[0027] 1) “Whether it is after the user’s first power supply date”: Determine whether the project approval time is later than the user’s first power supply date;

[0028] 2) “Whether it meets the project requirements”: assess whether the project meets the established requirements or standards;

[0029] 3) “Whether the power supply time requirements are met”: Determine whether the power supply time of the project complies with relevant regulations or plans;

[0030] If all three conditions above are met, the process will continue. Otherwise, it is necessary to return to the previous step to make corrections or take other measures;

[0031] Next, the process enters the "Project under review, unreviewed can be considered as review results" stage. At this stage, if the project is under review but has not yet completed the review, it is considered to have obtained some form of review results; at this time, the system or operator needs to check the following two conditions again:

[0032] 1) "Whether it is before the user's first power supply date": Determine whether the current status of the project is still earlier than the user's first power supply date; this condition is used to ensure that the project approval progress does not lag behind the user's electricity demand;

[0033] 2) “Whether it meets the power supply time requirements”: reconfirm whether the power supply time of the project meets the requirements;

[0034] If both of the above conditions are met, or because the project is under review but is considered reviewed, the process will continue to the last stage - "Close".

[0035] As a further solution of the present invention: the energy consumption progress model of key projects based on regression algorithm can be combined with annual production and operation energy consumption data, and utilize the computing resources of the data center to obtain the project energy consumption progress.

[0036] As a further solution of the present invention: the energy consumption monitoring model based on multiple warning states can construct a static / dynamic energy consumption monitoring model, calculate the project's monthly comprehensive energy consumption and the approved monthly average of energy consumption, and obtain the project's monthly status and static warning results; calculate the project's cumulative comprehensive energy consumption for the year and the monthly cumulative energy consumption warning value to obtain the project's annual cumulative status and dynamic warning results.

[0037] As a further solution of the present invention: the electricity-energy-carbon analysis model based on the ARDL algorithm can fully consider the relationship between variables and their own lags and other factors, is suitable for small sample data, can better capture the trend changes of variables, and realize the accurate calculation of project-level carbon emissions and carbon footprints.

[0038] Beneficial effects of the present invention:

[0039] Government level:

[0040] 1. Improve supervision efficiency and accuracy: With the automated targeted supervision data support provided by the system algorithm, the supervision department can quickly lock in the over-standard early warning analysis projects of key concern, so that supervision resources can be efficiently allocated. For risk-free projects, unnecessary manpower input is reduced, thereby greatly reducing the manpower cost of verification work and improving overall supervision efficiency.

[0041] 2. Optimize the allocation of regulatory resources: The application of system algorithms enables regulatory authorities to dynamically adjust the intensity and frequency of supervision based on data analysis results, ensuring that regulatory resources can be accurately invested in where they are most needed, effectively avoiding waste of resources.

[0042] 3. Strengthen policy implementation: Through real-time monitoring and analysis of energy consumption data by system algorithms, regulatory authorities can promptly detect and correct violations, ensuring the smooth implementation of relevant policies such as the country's "3060 dual carbon targets".

[0043] Enterprise level:

[0044] 1. Reduce the risk of excessive energy consumption: The automated in-process early warning capability at the project level enables enterprises to promptly detect and resolve problems of excessive energy consumption, thereby avoiding the risk of additional expenditures caused by fines for exceeding the standard and reducing the overall cost of the enterprise.

[0045] 2. Improve project management level: Through the real-time energy consumption data and analysis results provided by the system algorithm, enterprises can formulate and adjust energy consumption management strategies more scientifically and improve project management level and energy efficiency level.

[0046] 3. Enhance market competitiveness: Under the background of "dual carbon", enterprises can actively respond to the national call, reduce carbon emissions and energy consumption levels through intelligent transformation and energy efficiency improvement, thereby enhancing market competitiveness and winning more market share.

[0047] Social level:

[0048] 1. Promote the realization of the dual-carbon vision: By implementing energy consumption control strategies, the implementation of this system algorithm and strategy will help promote the realization of the country's dual-carbon vision and promote sustainable economic and social development.

[0049] 2. Promote industrial upgrading and transformation: Research on energy consumption measurement, analysis and monitoring of key energy-consuming projects can deepen the importance of power data in energy consumption supervision and promote the development of the industry towards a more intelligent and green direction.

[0050] 3. Provide demonstration effect: The successful application of the system algorithm and strategy in a certain region can provide useful reference and demonstration effect for other regions, and promote greater progress in energy consumption management and carbon emission control across the country. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 The present invention provides a flow chart for identifying and determining electricity users of a project-level energy consumption monitoring model that implements high-frequency calculation of energy and carbon using electricity.

[0052] Figure 2 The invention discloses an energy-saving review project progress and status model for a project-level energy consumption monitoring model that realizes high-frequency calculation of energy and carbon by electricity.

[0053] Figure 3 The invention discloses an energy consumption monitoring and early warning model for a project-level energy consumption monitoring model that realizes high-frequency calculation of energy and carbon by electricity.

[0054] Figure 4 The invention discloses an ARDL algorithm-based electricity-energy-carbon analysis model for realizing a high-frequency project-level energy consumption monitoring model for calculating carbon by calculating energy using electricity. DETAILED DESCRIPTION

[0055] The present invention will be further described below.

[0056] A project-level energy consumption monitoring model that realizes high-frequency calculation of carbon by electricity is characterized by comprising the following specific steps:

[0057] (1) A matching technology based on keyword extraction from multidimensional data is used to perform preliminary management and cleaning of multidimensional data. A keyword extraction algorithm based on statistical characteristics is used to perform preliminary screening of each project monitoring subject. A directional task script is established to match the multidimensional data to obtain a preliminary project list. The preliminary project list includes "matched projects" and "projects to be verified". After manual screening based on the preliminary project list, a complete list of project electricity monitoring subjects is obtained. The accurate verification matching rate reaches 90% or more, thereby realizing targeted management of project energy consumption.

[0058] Table 1. Example of matching results

[0059]

[0060] (2) A decision tree-based classification algorithm for project electricity users is used to define and identify matched projects, and to classify single / comprehensive projects based on whether to apply for business expansion services or install additional monitoring and collection devices, namely, the "case analysis method". By utilizing internal and external data cross-analysis technology and multi-data source residual noise and abnormal information elimination methods (to improve the accuracy and reliability of monitoring results), relevant information about project electricity users is obtained. Based on the relevant information such as power user files, user business expansion applications, project bidding and construction conditions, it is determined whether the metering point ID is correct, the construction and operation conditions, and the project approval time, and the identification and classification of project electricity users is completed, resulting in a set of energy-saving review project monitoring lists that meet the algorithm requirements. After on-site verification, among the 141 projects tested, there were 76 single projects and 65 comprehensive projects, with an accuracy rate of 95%.

[0061] Electricity user identification and determination process (such as Figure 1 shown):

[0062] Starting from the "Start" stage, first, enter the "Verify Successful Project" stage; this stage means that the project has passed a certain verification and is qualified for further processing.

[0063] Next, the process enters a decision point, namely, "whether the user is a non-temporary user of electricity". The system or operator needs to determine whether the user associated with the project belongs to the non-temporary user type. If the user is determined to be a non-temporary user of electricity (i.e., "yes"), the process continues to proceed downward. If the user is determined to be a temporary user of electricity (i.e., "no"), other processing flows may be triggered.

[0064] For non-temporary electricity users, the process enters the "project approval time" stage; at this stage, the system or operator needs to pay attention to the project approval time and check the following three conditions in turn:

[0065] 1) “Whether it is after the user’s first power supply date”: Determine whether the project approval time is later than the user’s first power supply date;

[0066] 2) “Whether it meets the project requirements”: assess whether the project meets the established requirements or standards;

[0067] 3) “Whether the power supply time requirements are met”: Determine whether the power supply time of the project complies with relevant regulations or plans;

[0068] If all three conditions above are met, the process will continue. Otherwise, it is necessary to return to the previous step to make corrections or take other measures;

[0069] Next, the process enters the "Project under review, unreviewed can be considered as review results" stage. At this stage, if the project is under review but has not yet completed the review, it is considered to have obtained some form of review results (although this result may not be final); at this time, the system or operator needs to check the following two conditions again:

[0070] 1) "Whether it is before the user's first power supply date": Determine whether the current status of the project (even if it has not been fully reviewed) is still earlier than the user's first power supply date; this condition is used to ensure that the project approval progress does not lag behind the user's electricity demand;

[0071] 2) “Whether it meets the power supply time requirements”: reconfirm whether the power supply time of the project meets the requirements;

[0072] If both of the above conditions are met, or because the project is under review but is considered reviewed, the process will continue to the last stage - "Close".

[0073] (3) A model for the energy consumption progress of key projects based on a regression algorithm is used to obtain the monthly electricity consumption based on the approved energy-saving review project data and the annual production and operation energy consumption data using the data center. The data center computing resources are used to calculate the specific energy consumption of each project and the comprehensive energy consumption of the project. The comprehensive energy consumption of the project is calculated with the approved energy consumption value to obtain the energy consumption progress of the project (e.g. Figure 2 Provides a visual display of the project’s energy consumption to regulatory authorities.

[0074] (4) An energy consumption monitoring model based on multiple warning states, relying on the computing resources of the data center, calculates the historical monthly electricity consumption of the project with the project energy approval data and the energy conversion coefficient to obtain the historical monthly comprehensive energy consumption, and divides it with the annual approval value to obtain the monthly cumulative energy consumption warning value. The monthly electricity consumption data of each project is processed by the same technical means to obtain the comprehensive energy consumption of the current month and the cumulative comprehensive energy consumption of the year. Finally, a static / dynamic energy consumption monitoring model is constructed, and the monthly comprehensive energy consumption of the project is calculated with the monthly average of the energy consumption approval to obtain the monthly status and static warning results of the project; the cumulative comprehensive energy consumption of the project in the year is calculated with the monthly cumulative energy consumption warning value to obtain the annual cumulative status and dynamic warning results of the project; the timeliness and accuracy of energy consumption supervision are improved (such as Figure 3 shown).

[0075] (5) An electricity-energy-carbon analysis model based on the ARDL algorithm is used to further integrate the electricity consumption data collected in the project electricity user identification module with influencing factors such as energy consumption types, and use the ARDL model to perform an autoregressive distributed lag model for time series. This model can fully consider the relationship between the variable and its own lag and other factors, is suitable for small sample data, and can better capture the trend changes of variables. Industry and agriculture are important industries, and the electricity-energy-carbon analysis model of the ARDL algorithm is constructed for verification testing. The error rate between the predicted value and the actual value in 2020 is 0.76% and -0.32%. Accurate calculation of carbon emissions and carbon footprint at the project level (such as Figure 4 shown).

[0076] In response to the existing energy consumption supervision issues of key projects, the research team intends to take advantage of the wide coverage and high timeliness of power data, build a scientific model, calculate energy with electricity, measure carbon with electricity, create an automated early warning system, and then realize real-time perception of project energy consumption, providing convenient and efficient auxiliary tools for regulatory departments to manage digital energy consumption. To this end, the research team specifically proposed the following six research points:

[0077] The first is to carry out multi-data fusion and matching research, by fusing and matching various types of data collected in prior preparation to ensure the source and reliability of data acquisition throughout the research process.

[0078] The second is to carry out single / comprehensive project identification and analysis, improve the accuracy and reliability of monitoring results, and enhance the efficiency of on-site energy-saving reviews by supervisors.

[0079] The third is to carry out comprehensive energy consumption research based on electricity data measurement, realize project energy consumption calculation and analysis based on electricity data, and form a management and control list so that supervisors can directly observe the energy consumption of the project.

[0080] Fourth, build an energy consumption early warning system for energy-saving review projects, establish and improve a scientific, complete and unified project indicator system and monitoring system, realize an automated early warning system for project energy consumption saturation reminders and over-limit alarms, and provide digital support for regulators to shift from post-site visits to in-process early warning management.

[0081] Fifth, we will build an electricity-energy-carbon model system for energy-saving review projects, realize the statistics of carbon emissions and traceability of carbon footprints in various energy-saving review projects, make project-level carbon emissions and carbon footprints observable, measurable and controllable, provide government regulatory agencies with project-level granular production carbon emission data, and help the government promote refined carbon management.

[0082] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A project-level energy consumption monitoring model that implements high-frequency calculation of energy and carbon by electricity, characterized by: The specific steps are as follows: (1) A matching technology based on keyword extraction from multidimensional data is used to perform preliminary management and cleaning of multidimensional data. The monitoring subjects of each project are preliminarily screened based on the keyword extraction algorithm of statistical characteristics. A directional task script is established to match the multidimensional data to obtain a preliminary project list. The preliminary project list includes "matched projects" and "projects to be verified". After manual screening, a complete list of project electricity monitoring subjects is obtained, and the accurate verification matching rate reaches 90% or above; (2) A decision tree-based classification algorithm for project electricity users, which is used to define and identify matched projects, and to classify single / comprehensive projects according to whether they apply for business expansion services or install additional monitoring and collection devices. By using internal and external data cross-analysis technology and multi-data source residual noise and abnormal information elimination methods, relevant information on project electricity users is obtained, and the identification and classification of project electricity users is completed, resulting in a set of energy-saving review project monitoring lists that meet the algorithm requirements; (3) A key project energy consumption progress model based on a regression algorithm, which is used to calculate the comprehensive energy consumption of the project based on the approved energy consumption data of the energy-saving review project and the annual production and operation energy consumption data, and calculate the comprehensive energy consumption of the project with the approved energy consumption value to obtain the energy consumption progress of the project; (4) An energy consumption monitoring model based on multiple early warning states is used to calculate the historical monthly electricity consumption of the project, the project energy approval data, and the energy conversion coefficient to obtain the historical monthly comprehensive energy consumption, and then divide it by the annual approved value to obtain the monthly cumulative energy consumption early warning value. A static / dynamic energy consumption monitoring model is further constructed to obtain the monthly status and static early warning results of the project, as well as the annual cumulative status and dynamic early warning results; (5) An electricity-energy-carbon analysis model based on the ARDL algorithm is used to further integrate the electricity consumption data collected in the project electricity user identification module with the influencing factors of energy consumption types, and use the ARDL model to perform an autoregressive distributed lag model of the time series to achieve accurate calculation of project-level carbon emissions and carbon footprints.

2. According to claim 1, a project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by electricity is characterized in that: The relevant information includes electricity user files, user business expansion registration, project bidding and construction status to determine whether the metering point ID is correct, construction and operation status, and project approval time.

3. According to claim 1, a project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by electricity is characterized in that: The electricity user identification and determination process: Start from the "Start" stage, first enter the "Verify Successful Project" stage; Next, the process enters a decision point, namely "whether the user is a non-temporary user of electricity". The system or operator needs to determine whether the user associated with the project is a non-temporary user of electricity. If the user is determined to be a non-temporary user of electricity, the process continues to proceed downward. If the user is determined to be a temporary user of electricity, other processing flows are triggered. For non-temporary electricity users, the process enters the "project approval time" stage; at this stage, the system or operator needs to pay attention to the project approval time and check the following three conditions in turn: 1) "Whether it is after the user's first power supply date": Determine whether the project approval time is later than the user's first power supply date; 2) "Whether it meets the project requirements": assess whether the project meets the established requirements or standards; 3) "Whether it complies with the power supply time requirements": determine whether the power supply time of the project complies with relevant regulations or plans; If all three conditions above are met, the process will continue; otherwise, it is necessary to return to the previous step to make corrections or take other measures; Next, the process enters the "project under review, unreviewed is considered as review result" stage; At this stage, the project is under review but has not yet been reviewed. It is considered to have obtained some form of review results. At this point, the system or operator needs to check the following two conditions again: 1) "Whether it is before the user's first power supply date": Determine whether the current status of the project is still earlier than the user's first power supply date; this condition is used to ensure that the project approval progress does not lag behind the user's electricity demand; 2) "Whether it meets the power supply time requirements": reconfirm whether the power supply time of the project meets the requirements; If both of the above conditions are met, or because the project is under review but is considered reviewed, the process will continue to the last stage - "Close".

4. According to claim 1, a project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by electricity is characterized in that: The energy consumption progress model for key projects based on the regression algorithm can be combined with annual production and operation energy consumption data and utilize the computing resources of the data center to obtain the project energy consumption progress.

5. According to claim 1, a project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by electricity is characterized in that: The energy consumption monitoring model based on multiple early warning states can construct a static / dynamic energy consumption monitoring model, calculate the monthly comprehensive energy consumption of the project and the monthly average of the energy consumption approval, and obtain the monthly status and static early warning results of the project; Calculate the project's cumulative comprehensive energy consumption for the year and the monthly cumulative energy consumption warning value to obtain the project's annual cumulative status and dynamic warning results.

6. A project-level energy consumption monitoring model for realizing high-frequency calculation of carbon by electricity according to claim 1, characterized in that: The electricity-energy-carbon analysis model based on the ARDL algorithm can fully consider the relationship between variables and their own lags and other factors, is suitable for small sample data, can better capture the trend changes of variables, and realize the accurate calculation of project-level carbon emissions and carbon footprints.

Images