CN117726053A - Carbon emission monitoring method and system applied to digital platform system - Google Patents
Carbon emission monitoring method and system applied to digital platform system Download PDFInfo
- Publication number
- CN117726053A CN117726053A CN202410178403.9A CN202410178403A CN117726053A CN 117726053 A CN117726053 A CN 117726053A CN 202410178403 A CN202410178403 A CN 202410178403A CN 117726053 A CN117726053 A CN 117726053A
- Authority
- CN
- China
- Prior art keywords
- emission
- rule
- data
- deviation
- target
- 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
- 229910052799 carbon Inorganic materials 0.000 title claims abstract description 307
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 title claims abstract description 306
- 238000012544 monitoring process Methods 0.000 title claims abstract description 127
- 238000000034 method Methods 0.000 title claims abstract description 58
- 230000010354 integration Effects 0.000 claims abstract description 140
- 238000001514 detection method Methods 0.000 claims abstract description 99
- 238000005265 energy consumption Methods 0.000 claims abstract description 18
- 238000005516 engineering process Methods 0.000 claims abstract description 13
- 230000003993 interaction Effects 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims description 137
- 238000011156 evaluation Methods 0.000 claims description 71
- 238000001914 filtration Methods 0.000 claims description 18
- 238000013075 data extraction Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 description 17
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 229910002092 carbon dioxide Inorganic materials 0.000 description 4
- 239000001569 carbon dioxide Substances 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 150000001721 carbon Chemical class 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 239000005431 greenhouse gas Substances 0.000 description 1
- 238000010921 in-depth analysis Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 230000029305 taxis Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application provides a carbon emission monitoring method and system applied to a digital platform system, which collect and generate candidate carbon emission data through interconnection interaction with various energy utilization devices and energy consumption devices, and store the candidate carbon emission data in the digital platform system based on a blockchain technology, so that the safety and the non-tamper property of the data can be ensured. And then, analyzing the candidate carbon emission data according to the customized carbon emission deviation detection model so as to generate a carbon emission deviation detection result. The method and the device can be used for customizing the emission control system according to actual conditions by generating alternative emission deviation list rules based on a predefined emission scene set, so that the applicability and flexibility of a customized carbon emission deviation detection model are enhanced. Moreover, the customized carbon emission deviation detection model is generated by adopting an integration strategy searching method, so that the accuracy and stability of the customized carbon emission deviation detection model can be further improved, and the accuracy and reliability of carbon emission deviation detection are ensured.
Description
Technical Field
The application relates to the technical field of computers, in particular to a carbon emission monitoring method and system applied to a digital platform system.
Background
In order to effectively control and reduce carbon emissions, various energy and consumption devices that produce carbon emissions need to be monitored and analyzed accurately in real time.
However, existing carbon emission information collection and analysis techniques have some problems. First, errors or holes may occur during the data collection process, resulting in an inability to guarantee the quality and integrity of the acquired candidate carbon emission information stream. Second, systems that store such information often lack the necessary security precautions, are vulnerable to tampering and attack, and thus affect the authenticity and reliability of the data. Again, most of existing carbon emission deviation detection models are based on unified standards, lack pertinence, and cannot meet specific requirements under different situations. In addition, the generation of the carbon emission deviation detection model generally lacks a mechanism of dynamic adjustment and optimization, so that the detection effect and accuracy thereof often have difficulty in reaching an ideal state in the face of a complex variation of emission conditions.
Therefore, a new technical solution is urgently needed, a high-quality candidate carbon emission information stream can be acquired and stored, a customized and targeted carbon emission deviation detection model can be generated, dynamic optimization can be performed according to actual conditions, and therefore efficiency and accuracy of carbon emission monitoring are improved, and management and control work of carbon emission is effectively supported.
Disclosure of Invention
In order to overcome at least the above-mentioned shortcomings in the prior art, an object of the present application is to provide a carbon emission monitoring method and system applied to a digital platform system.
In a first aspect, the present application provides a carbon emission monitoring method applied to a digital platform system, the method applied to a carbon emission monitoring system, the method comprising:
acquiring candidate carbon emission information streams stored in a digital platform system based on a blockchain technology, wherein the candidate carbon emission information streams are carbon emission data generated by carrying out interconnection and interaction acquisition with various energy utilization devices and energy consumption devices;
analyzing the candidate carbon emission information stream according to a customized carbon emission deviation detection model to generate a carbon emission deviation detection result; the generating step of the customized carbon emission deviation detection model comprises the following steps:
generating an alternative emission deviation list rule corresponding to the emission scene set based on emission data points corresponding to set emission variation vectors in the predefined emission scene set and emission classification attributes corresponding to the emission data points;
and carrying out integration strategy searching on the alternative emission deviation list rules corresponding to the emission scene set, and generating a customized carbon emission deviation detection model according to the integration strategy searching data.
In a possible implementation manner of the first aspect, the generating the alternative emission deviation list rule corresponding to the emission scenario set based on the emission data point corresponding to the set emission variation vector in the predefined emission scenario set and the emission classification attribute corresponding to the emission data point includes:
acquiring a plurality of set emission variation vectors; the set emission variation vector corresponds to an emission data point carrying an emission classification attribute;
performing emission scene division on a plurality of the set emission variation vectors according to a set emission theme to generate a predefined emission scene set;
and generating a target deviation matching rule corresponding to the set emission variation vector in the emission scene set based on the emission data point corresponding to the set emission variation vector in the emission scene set and the emission classification attribute corresponding to the emission data point, and generating an alternative emission deviation list rule corresponding to the emission scene set according to the target deviation matching rule.
In a possible implementation manner of the first aspect, the generating, based on the emission data point corresponding to the set emission variation vector in the emission scenario set and the emission classification attribute corresponding to the emission data point, a target deviation matching rule corresponding to the set emission variation vector in the emission scenario set, and generating, according to the target deviation matching rule, an alternative emission deviation list rule corresponding to the emission scenario set includes:
Acquiring an initial judgment chart network corresponding to a set emission variation vector in the emission scene set;
taking the emission data points corresponding to the set emission variation vectors in the emission scene set as network learning data corresponding to the set emission variation vectors in the emission scene set, and performing parameter learning on an initial judgment graph network corresponding to the set emission variation vectors in the emission scene set according to the network learning data and the emission classification attribute corresponding to the network learning data to generate a target judgment graph network corresponding to the set emission variation vectors in the emission scene set;
taking the deviation matching rule of the network members in the target judgment graph network as a target deviation matching rule corresponding to a set emission variation vector in the emission scene set;
and generating an alternative discharge deviation list rule corresponding to the discharge scenario set according to the target deviation matching rule.
In one possible implementation manner of the first aspect, the performing, as the network learning data corresponding to the set emission variation vector in the emission scenario set, parameter learning on the initial decision graph network corresponding to the set emission variation vector in the emission scenario set according to the network learning data and the emission classification attribute corresponding to the network learning data, to generate the target decision graph network corresponding to the set emission variation vector in the emission scenario set includes:
Taking the emission data points corresponding to the set emission variation vectors in the emission scene set as network learning data corresponding to the set emission variation vectors in the emission scene set, loading the network learning data corresponding to the set emission variation vectors in the emission scene set into an initial judgment graph network corresponding to the set emission variation vectors in the emission scene set to make a decision, and generating initial emission classification attributes output by the initial judgment graph network;
updating the deviation matching rule of the network members in the initial judgment graph network according to the emission classification attribute corresponding to the network learning data until the weight of the preset emission data point in the initial emission classification attribute meets the target requirement, and generating a target judgment graph network corresponding to the set emission variation vector in the emission scene set; and the attribute type of the emission classification attribute corresponding to the preset emission data point is a set type.
In a possible implementation manner of the first aspect, the generating, according to the target deviation matching rule, an alternative discharge deviation list rule corresponding to the discharge scenario set includes:
integrating the set emission variation vector in the emission scenario set and the target deviation matching rule corresponding to the set emission variation vector in the emission scenario set to generate an alternative emission deviation list rule corresponding to the emission scenario set.
In a possible implementation manner of the first aspect, the performing an integration policy search on the alternative emission deviation list rule corresponding to the emission scenario set, generating a customized carbon emission deviation detection model according to the integration policy search data includes:
performing integration strategy searching on the alternative discharge deviation list rules corresponding to the discharge scenario set, and determining target alternative discharge deviation rules corresponding to the discharge scenario set according to the integration strategy searching data;
and generating the customized carbon emission deviation detection model according to a merging rule sequence of target alternative emission deviation rules corresponding to the emission scene set.
In a possible implementation manner of the first aspect, the performing an integration policy search on the candidate emission deviation list rule corresponding to the emission scenario set, and determining, according to the integration policy search data, the target candidate emission deviation rule corresponding to the emission scenario set includes:
performing integration strategy search on the alternative discharge deviation list rules corresponding to the discharge scenario set to generate integration strategy search data; the integration strategy searching data comprises an alternative discharge deviation list rule and one or more target integration rules corresponding to the discharge scenario set;
Taking the alternative discharge deviation list rule corresponding to the discharge scenario set and one or more target integration rules as the alternative discharge deviation rule corresponding to the discharge scenario set, and determining preset rule evaluation data of the alternative discharge deviation rule corresponding to the discharge scenario set; the preset rule evaluation data reflects the discharge deviation detection effect of the alternative discharge deviation rule corresponding to the discharge scenario set;
and outputting the rule of which the preset rule evaluation data is larger than a set value in the alternative discharge deviation rules corresponding to the discharge scene set as a target alternative discharge deviation rule corresponding to the discharge scene set.
In a possible implementation manner of the first aspect, the performing an integration policy search on the alternative emission deviation list rule corresponding to the emission scenario set, to generate integration policy search data includes:
taking the alternative discharge deviation list rule corresponding to the discharge scenario set as a current undetermined rule, and constructing a current undetermined rule sequence corresponding to the discharge scenario set;
integrating any two current undetermined rules in the current undetermined rule sequence to generate integrated rule data; the integration rule data comprises one or more currently pending integration rules corresponding to the currently pending rule sequence;
Determining a target integration rule corresponding to the currently pending rule sequence from one or more currently pending integration rules corresponding to the currently pending rule sequence; the target integration rule is a rule that the current rule evaluation data accords with target requirements in one or more current integration rules corresponding to the current rule sequence; the current rule evaluation data reflects the emission deviation detection effect of the current undetermined integration rule;
taking a target integration rule corresponding to the currently pending rule sequence as a currently pending rule in the currently pending rule sequence;
circularly executing the step of integrating any two current undetermined rules in the current undetermined rule sequence to a target integration rule corresponding to the current undetermined rule sequence as the current undetermined rule in the current undetermined rule sequence until the current undetermined rule which accords with the preset length exists in the current undetermined rule sequence, and generating scene candidate integration search data corresponding to the emission scene set;
the step of determining the current rule evaluation data includes:
acquiring an emission variation evaluation data sequence; the emission variation evaluation data sequence comprises at least two groups of emission variation evaluation data and evaluation feature vectors corresponding to each group of emission variation evaluation data;
Searching in the emission variation evaluation data sequence according to the currently pending integration rule to obtain target emission variation evaluation data;
determining an accuracy index and a coverage index corresponding to the currently pending integration rule according to the evaluation feature vector corresponding to the target emission variation evaluation data;
and determining the current rule evaluation data of the current integration rule according to the accuracy index and the coverage index corresponding to the current integration rule.
In a possible implementation manner of the first aspect, the candidate carbon emission information stream includes a candidate emission variation vector and a candidate emission data point corresponding to the candidate emission variation vector;
the method further comprises the steps of:
when the carbon emission deviation detection result reflects that a target candidate emission variation vector matched with a set emission variation vector in the customized carbon emission deviation detection model exists in the candidate emission variation vector, and a candidate emission data point corresponding to the target candidate emission variation vector is matched with a target deviation matching rule, determining that the candidate carbon emission information flow is deviation carbon emission data;
loading the deviation carbon emission data to a significance data extraction network for feature filtering to generate a feature filtering result output by the significance data extraction network;
Wherein the saliency data extraction network comprises at least one feature filtering network; the characteristic filtering result comprises target carbon emission variation data and a target characteristic filtering network corresponding to the target carbon emission variation data; the target carbon emission variation data is data conforming to at least one characteristic filter network in the deviation carbon emission data.
In a second aspect, embodiments of the present application also provide a carbon emission monitoring system including a processor and a machine-readable storage medium having stored therein a computer program loaded and executed in accordance with the processor to implement the carbon emission monitoring method of the first aspect above applied to a digital platform system.
According to the technical scheme in any aspect, through interconnection interaction with various energy consumption equipment and energy consumption equipment, candidate carbon emission data is collected and generated, and the candidate carbon emission data is stored in a digital platform system based on a blockchain technology, so that the safety and the non-falsifiability of the data can be ensured, and the trust and the effectiveness of the data are enhanced. And then, analyzing the candidate carbon emission data according to the customized carbon emission deviation detection model, so as to generate a carbon emission deviation detection result. Setting emission data points corresponding to emission variation vectors and emission classification attributes corresponding to the emission data points according to the predefined emission scene set, and generating alternative emission deviation list rules corresponding to the emission scene set; and then, carrying out integration strategy searching on the alternative emission deviation list rules, and finally generating a customized carbon emission deviation detection model according to the integration strategy searching data. The use of such customized carbon emission deviation detection models can provide more accurate and targeted carbon emission deviation detection results, which can help to more effectively find and treat carbon emission problems. Meanwhile, by generating the alternative emission deviation list rule based on the predefined emission scene set, customization setting can be performed according to actual conditions, and applicability and flexibility of the customized carbon emission deviation detection model are enhanced. Moreover, the model is generated by adopting an integration strategy searching method, so that the accuracy and stability of the customized carbon emission deviation detection model can be further improved, and the accuracy and reliability of carbon emission deviation detection are ensured.
Drawings
For a clearer description of the technical solutions of the embodiments of the present application, reference will be made to the accompanying drawings, which are needed to be activated, for the sake of simplicity, and it should be understood that the following drawings only illustrate some embodiments of the present application and should therefore not be considered as limiting the scope, and that other related drawings can be obtained according to these drawings without the need for inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a carbon emission monitoring method applied to a digital platform system according to an embodiment of the present application;
fig. 2 is a schematic functional block diagram of a carbon emission monitoring system for implementing the carbon emission monitoring method applied to the digital platform system according to the embodiment of the present application.
Detailed Description
The following description is presented to enable one of ordinary skill in the art to make and use the application and is provided in the context of a particular application and its requirements. It will be apparent to those having ordinary skill in the art that various changes can be made to the disclosed embodiments and that the general principles defined herein may be applied to other embodiments and applications without departing from the principles and scope of the present application. Thus, the present application is not limited to the embodiments described, but is to be accorded the widest scope consistent with the claims.
Referring to fig. 1, the present application provides a carbon emission monitoring method applied to a digital platform system, comprising the following steps.
Step S110, candidate carbon emission information flows stored in the digital platform system based on the blockchain technology are obtained, wherein the candidate carbon emission information flows are carbon emission data generated by carrying out interconnection and interaction collection with various energy utilization devices and energy consumption devices.
In this embodiment, the carbon emission monitoring system is connected to a digital platform system, and the digital platform system stores carbon emission data generated by interconnection and interaction collection of various energy consumption devices and various energy consumption devices by using a blockchain technology.
The digital platform system is a platform integrating various information technologies and tools and is used for achieving the functions of data acquisition, processing, analysis, display and the like. For example, an enterprise energy management system is a digital platform system, various energy use data can be collected through sensors and intelligent meters, then the data are processed and analyzed, finally the data are displayed in the form of charts and reports, and related personnel are helped to make more intelligent energy management decisions.
Blockchain technology is a de-centralized, de-trusted, collective-maintenance, reliable database technology that allows participants in a network to share and verify an ever-growing list of records (called "blocks") and to use encryption techniques to ensure data integrity and non-tamper-ability. In carbon emission data management, blockchain techniques may ensure the authenticity and non-tamper-ability of each carbon emission data point. For example, carbon emissions data generated by a power plant may be recorded and verified by blockchain technology, and any attempt to tamper with the data may be discovered and prevented by other participants in the network.
The candidate carbon emission information stream refers to a series of carbon emission data collected by a specific means that has not been finally confirmed and processed. These data are typically in the form of information streams containing various carbon emission related information. For example, traffic authorities in a city can collect a large amount of vehicle exhaust emission data via sensors mounted on buses and taxis, which form candidate carbon emission information streams. These data require further processing and analysis to derive accurate carbon emissions and emission trends.
The energy-consuming devices refer to devices that consume energy in production, living, etc., such as motors, boilers, air conditioners, etc. For example, an electric motor in a factory is a typical energy-consuming device that drives various mechanical devices on a production line by consuming electrical energy. The factory can reduce production costs and reduce carbon emissions by monitoring and managing the energy consumption of the motor.
By energy consuming devices are meant those devices which produce energy consumption during operation, typically various electricity, gas or oil consuming devices. For example, refrigerators and washing machines in homes are energy consuming devices that consume electrical energy during operation. By monitoring and managing the energy consumption of these devices, the household can use energy more reasonably, reducing waste and carbon emissions.
In the carbon emission monitoring system, various energy consumption devices and energy consumption devices can realize interconnection interaction through the Internet or the Internet of things technology. For example, an intelligent home system can connect various energy consumption devices in a home through a wireless network to realize remote control and energy consumption monitoring. Thus, the user can check and control the energy consumption equipment in the home at any time through the mobile phone or the computer, thereby realizing a more energy-saving and environment-friendly life style.
Thus, the carbon emission monitoring system sends a request to the digital platform system to obtain candidate carbon emission information streams stored based on the blockchain technology. These candidate carbon emission information streams include carbon emission data, such as emissions of carbon dioxide, methane, etc. from various equipment (e.g., power plants, plant machinery, vehicles, etc.) collected in real time. The digitized platform system responds to the request to transmit an encrypted and non-tamperable candidate carbon emission information stream to the carbon emission monitoring system.
Step S120, analyzing the candidate carbon emission information stream according to the customized carbon emission deviation detection model to generate a carbon emission deviation detection result.
The method mainly uses a customized carbon emission deviation detection model to analyze the collected candidate carbon emission information flow so as to detect any potential abnormality or deviation.
The customized carbon emission deviation detection model is a pre-designed analysis model and is used for identifying abnormal modes or deviations in carbon emission data based on historical data, industry standards and expert knowledge. The model may include a series of algorithms, rules, and thresholds for defining what is the "normal" carbon emission behavior, and what is the potential deviation.
The carbon emission deviation detection result is an output of the customized carbon emission deviation detection model after analysis, and generally includes a detailed description of any deviation identified, such as the type, size, time and frequency of occurrence, etc. This information is critical to subsequent investigation and corrective action.
Assume that a city has a carbon emission monitoring system that continuously collects carbon emission data from various plants within the city. These data form a candidate carbon emission information stream. For example, carbon emission data including the emission amount of greenhouse gases such as carbon dioxide and methane, and conditions such as temperature and pressure at the time of emission are first collected from the discharge ports of the respective factories. Next, these data are analyzed using a custom carbon emission deviation detection model. For example, a series of thresholds and rules may be set based on historical average emissions, emissions criteria, and the like. For example, if the emissions from a plant increase dramatically over a short period of time, the model identifies it as a potential deviation beyond a set threshold. During the analysis, it may be found that the emission of a certain factory is abnormally high between 10 pm and 2 am. This period is typically a low peak period of the plant, so the high emissions are not consistent with the expected normal mode, and are modeled as a bias. Finally, a carbon emission deviation detection report is generated, and the deviation is detailed, including information such as occurrence time, specific value of emission, comparison with normal mode, etc. This report will be sent to the relevant manager or environmental authorities so that they will further investigate the cause and take the necessary corrective action.
Through such a process, the carbon emission monitoring system can help the manager to discover and address potential carbon emission problems in time, thereby supporting a more environmentally friendly and sustainable production approach.
The generating step of the customized carbon emission deviation detection model comprises the following steps:
step S101, based on emission data points corresponding to set emission variation vectors in a predefined emission scenario set and emission classification attributes corresponding to the emission data points, generating alternative emission deviation list rules corresponding to the emission scenario set.
Wherein the predefined set of emission scenarios refers to a series of emission conditions or scenarios that have been determined and set up prior to the establishment of the carbon emission deviation detection model. These scenarios may include a combination of different emissions sources, emissions intensities, emissions times, etc. For example, in the power industry, the predefined emission scenario set may include a reference scenario (emission amount at normal operation), a peak scenario (emission amount at peak operation increases), a fault scenario (emission amount abnormality caused by equipment failure), and the like.
The set emission variation vector refers to a quantization index for describing the direction and magnitude of variation in the carbon emission amount that may occur in a specific emission scenario, and may be a multi-dimensional vector, each dimension representing a different influencing factor. For example, in a traffic emission scenario, setting an emission variation vector may include a variation in the number of vehicles, travel speed, degree of road congestion, and the like. Variations in these factors can result in a corresponding increase or decrease in carbon emissions.
The emission data points refer to specific carbon emission values obtained by monitoring and measuring. Each data point is typically associated with a particular time, location, and emission source. For example, a Continuous Emission Monitoring System (CEMS) is installed at the plant's emissions, which records the carbon dioxide emissions at regular intervals (e.g., hourly). These recorded emissions are emission data points.
The emissions classification attribute refers to a feature or label that classifies and describes the emissions data points. These attributes may include the type of emissions source (e.g., coal-fired power plants, gasoline vehicles, etc.), the type of emissions materials (e.g., carbon dioxide, methane, etc.), the location and time of emissions, etc. For example, in urban traffic emissions, emission classification attributes may include vehicle type (private car, bus, truck, etc.), fuel type (gasoline, diesel, electric, etc.), and travel area (urban, suburban, highway, etc.).
The alternative emission deviation sheet rule refers to a candidate rule for detecting carbon emission deviation generated for each emission data point and its corresponding emission classification attribute based on a predefined emission scenario set and a set emission variation vector. These rules describe which data points may be considered as deviations under different scenarios. For example, assume that in a predefined set of emissions scenarios in the power industry, one scenario is "night low load operation". For this scenario, an alternative emission deviation list rule may be generated, specifying that during a night low load period, an emission of a certain power plant is considered to be a deviation if it exceeds a certain percentage (e.g. 10%) of the historical average emission for that period.
Thus, a series of emission scenario sets and corresponding rules need to be defined prior to generating the customized carbon emission deviation detection model. For example, a series of representative emissions scenarios (e.g., normal emissions, high emissions, low emissions, etc.) may be selected based on historical data and future predictions. For each scenario, a set of set emission variation vectors may be defined that describe the changes that may occur to emission data at different points in time and conditions. Based on each emission data point and its corresponding classification attribute (e.g., emission source type, emission species type, etc.), an alternative emission deviation sheet rule associated with each variation vector is generated. Thus, an alternative set of discharge deviation sheet rules is completed, which provide the basis for the generation of subsequent models.
Step S102, carrying out integration strategy searching on the alternative emission deviation list rules corresponding to the emission scene set, and generating a customized carbon emission deviation detection model according to the integration strategy searching data.
In this embodiment, the integrated policy search refers to a process of comprehensively analyzing and optimizing a plurality of alternative emission deviation list rules, so as to find one or more policies, and effectively combine these rules to improve accuracy and efficiency of carbon emission deviation detection. For example, assume that there are three alternative emission deviation sheet rules A, B and C, each applicable to a different emission scenario. By means of the integration strategy search, it may be found that the use of the combination of rules a and B enables a more accurate detection of carbon emission deviations in one scenario, while in another scenario the combination of rules B and C is more efficient. This process is just like finding an optimal pairing scheme, so that various rules can work cooperatively to exert the maximum performance.
The integration policy lookup data may be a conclusion or a scheme of values that are derived after an integration policy lookup of the alternative placement deviation sheet rules. This integration strategy search data determines how to effectively combine these single rules to improve the accuracy and efficiency of carbon emission deviation detection, which may include rule selection, weight assignment, decision logic determination, and the like. For example, assume that there are multiple alternative discharge bias sheet rules, each rule having different detection effects under different conditions. By integrating policy exploration, it may be found that combining certain rules together, or giving certain rules higher weights, may improve the accuracy of bias detection as a whole. The result of this process is an optimized rule set, i.e., an integration strategy search result, that will guide how these rules are applied in actual applications to detect carbon emission deviations.
In conducting an integration policy search, a large amount of data needs to be collected and analyzed. For example, traffic emission data for a region over the past year under different weather conditions may be collected and then initially detected using alternative emission deviation sheet rules. By analyzing the performances (such as detection accuracy, false alarm rate and the like) of the rules under different situations, valuable integrated strategy searching data can be obtained, and basis is provided for subsequent model customization.
The customized carbon emission deviation detection model refers to integrating and optimizing a plurality of alternative emission deviation rules according to a certain strategy according to the result of searching an integration strategy (the data of searching the integration strategy), so as to form a carbon emission deviation detection model specific to a specific emission scene set. The customized carbon emission deviation detection model can more accurately identify deviation in emission data, and help managers to timely find and cope with potential carbon emission problems. For example, assuming that rules A, B and C are combined in a particular manner by an integration strategy search, carbon emission bias can be effectively detected in a complex emission scenario. Then, based on this combination, a specific carbon emission deviation detection model can be customized. In practical application, the model can automatically monitor and analyze the emission data in the scene in real time, and timely discover and report any potential deviation condition.
That is, after the alternative emission deviation sheet rule set has been generated, the alternative emission deviation sheet rules may be integrated to generate a final custom carbon emission deviation detection model. For example, an integrated policy search is performed on alternative discharge bias sheet rules with the goal of finding a policy that can effectively incorporate these rules to maximize the accuracy and efficiency of the detection model. Based on the searched optimal integration strategy, a customized carbon emission deviation detection model is generated, and the model is ready to be deployed into a carbon emission monitoring system for practical application.
Based on the steps, the method and the device collect and generate candidate carbon emission data through interconnection interaction with various energy consumption devices and energy consumption devices, and store the candidate carbon emission data in a digital platform system based on a blockchain technology, so that the safety and the non-falsifiability of the data can be ensured, and the trust and the effectiveness of the data are enhanced. And then, analyzing the candidate carbon emission data according to the customized carbon emission deviation detection model, so as to generate a carbon emission deviation detection result. Setting emission data points corresponding to emission variation vectors and emission classification attributes corresponding to the emission data points according to the predefined emission scene set, and generating alternative emission deviation list rules corresponding to the emission scene set; and then, carrying out integration strategy searching on the alternative emission deviation list rules, and finally generating a customized carbon emission deviation detection model according to the integration strategy searching data. The use of such customized carbon emission deviation detection models can provide more accurate and targeted carbon emission deviation detection results, which can help to more effectively find and treat carbon emission problems. Meanwhile, by generating the alternative emission deviation list rule based on the predefined emission scene set, customization setting can be performed according to actual conditions, and applicability and flexibility of the customized carbon emission deviation detection model are enhanced. Moreover, the model is generated by adopting an integration strategy searching method, so that the accuracy and stability of the customized carbon emission deviation detection model can be further improved, and the accuracy and reliability of carbon emission deviation detection are ensured.
In one possible implementation, step S101 may include:
in step S1011, a plurality of set emission variation vectors are acquired. The set emission variation vector corresponds to an emission data point carrying an emission classification attribute.
In this embodiment, the carbon emission monitoring system first collects real-time emission data from each emission source (e.g., factory, power plant, vehicle, etc.) to which it is connected, which may include, for example, emission amount, emission substance type, emission time, etc., and automatically carries corresponding emission classification attributes (e.g., emission source type, fuel type, etc.). The data is processed and analyzed to extract a plurality of set emission variation vectors, each set emission variation vector representing the direction and magnitude of variation in the amount of emission that may occur under the influence of different factors (e.g., time, type of emission source, etc.).
Step S1012 performs emission scenario division on the plurality of the set emission variation vectors according to the set emission theme, and generates a predefined emission scenario set.
In this embodiment, the carbon emission monitoring system classifies and divides the plurality of set emission variation vectors according to the set emission subject (e.g., industry type, emission intensity, etc.). For example, the carbon emission monitoring system may categorize a variation vector associated with the power industry into one emission scenario set and a variation vector associated with the traffic industry into another emission scenario set. Each emission scenario set internally contains various emissions and modes of variation that may occur under a particular theme.
Step S1013, generating a target deviation matching rule corresponding to the set emission variation vector in the emission scenario set based on the emission data point corresponding to the set emission variation vector in the emission scenario set and the emission classification attribute corresponding to the emission data point, and generating an alternative emission deviation list rule corresponding to the emission scenario set according to the target deviation matching rule.
In this embodiment, the carbon emission monitoring system uses its built-in algorithm and model to further analyze the emission data points corresponding to each set emission variation vector within each predefined emission scenario set. The carbon emission monitoring system will compare these data points to normal emission patterns or historical data differences, identifying any potential deviations or anomalies. Based on these analyses, the carbon emission monitoring system generates a series of target bias matching rules that define which data points may be considered bias under different scenarios, and give corresponding judgment conditions and thresholds.
Finally, the carbon emission monitoring system generates a set of alternative emission deviation sheet rules for each predefined emission scenario set according to the generated target deviation matching rules. These alternative emission deviation sheet rules are specific and operable and directly correspond to emission data points and potential deviation situations in actual monitoring. In practical applications, real-time emissions data can be automatically monitored and judged according to the single rules, and any potential carbon emission deviation can be timely found and reported.
Thus, the present embodiment is able to generate a series of alternative emission deviation sheet rules for a particular scenario based on a predefined emission scenario set and a set emission variation vector. These rules provide important basis and support for subsequent carbon emission deviation detection.
In one possible implementation, step S1013 may include:
step S1013-1, obtaining an initial decision graph network corresponding to the set emission variation vector in the emission scenario set.
In this embodiment, the carbon emission monitoring system first accesses its internal storage or external database from which to retrieve an initial decision map network associated with a particular emission scenario set. The initial decision map network is a pre-built model that contains a series of nodes and connections corresponding to the set emission variation vector. Each node represents a particular emission data point or emission feature, and the connection represents the relationship between these data points or features.
Step S1013-2, using the emission data points corresponding to the set emission variation vectors in the emission scenario set as the network learning data corresponding to the set emission variation vectors in the emission scenario set, and performing parameter learning on the initial determination map network corresponding to the set emission variation vectors in the emission scenario set according to the network learning data and the emission classification attribute corresponding to the network learning data, so as to generate the target determination map network corresponding to the set emission variation vectors in the emission scenario set.
In this embodiment, the carbon emission monitoring system next extracts emission data points corresponding to the set emission variation vector from its actual or historical database. These data points contain information about the actual emissions, emission time, emission source type, etc., and have been labeled according to emission classification attributes. The carbon emission monitoring system uses these data points as network learning data for training and optimizing the initial decision map network.
The carbon emission monitoring system then processes and analyzes the network learning data using advanced machine learning algorithms, such as deep learning or graph neural networks. The carbon emissions monitoring system minimizes the difference between the network output and the actual emissions data by continuously adjusting the parameters and connection weights in the initial decision map network. In this process, the carbon emission monitoring system may be particularly concerned with data points and features closely related to emission classification attributes. Through multiple iterations and optimizations, the carbon emission monitoring system ultimately produces a more accurate and reliable target decision graph network that better captures and represents the relationship between the set emission variation vector and the emission data points.
Step S1013-3, using the deviation matching rule of the network member in the target judgment chart network as the target deviation matching rule corresponding to the set emission variation vector in the emission scene set.
After the target decision graph network is generated, the carbon emission monitoring system further analyzes the various nodes and connections in the network. For example, network members that have significant deviations from the normal emission pattern or the expected emission range may be identified and the deviation matching rules they represent extracted. These rules describe which emission data points may be considered anomalies or deviations under different circumstances and give corresponding judgment conditions and thresholds. Then, these rules are set as target deviation matching rules corresponding to the set emission variation vector.
Step S1013-4, generating an alternative discharge deviation list rule corresponding to the discharge scene set according to the target deviation matching rule.
For example, the carbon emission monitoring system generates a set of alternative emission deviation sheet rules for each emission scenario set based on the generated target deviation matching rules. These alternative emission deviation sheet rules are specific, operational, and directly correspond to emission data points and potential deviation situations in the actual monitoring. In practical applications, real-time emissions data can be automatically monitored and judged according to the single rules, and any potential carbon emission deviation can be timely found and reported.
Through the above steps, the carbon emission monitoring system is able to generate a series of alternative emission deviation sheet rules for a particular scenario based on the set emission variation vectors and emission data points in the emission scenario set. The rules provide important basis and support for subsequent carbon emission deviation detection, and are helpful for improving the accuracy and efficiency of carbon emission monitoring.
In one possible implementation, step S1013-2 may include:
1. and taking the emission data points corresponding to the set emission variation vectors in the emission scene set as network learning data corresponding to the set emission variation vectors in the emission scene set, loading the network learning data corresponding to the set emission variation vectors in the emission scene set into an initial judgment graph network corresponding to the set emission variation vectors in the emission scene set to make a decision, and generating initial emission classification attributes output by the initial judgment graph network.
2. Updating the deviation matching rule of the network members in the initial judgment graph network according to the emission classification attribute corresponding to the network learning data until the weight of the preset emission data point in the initial emission classification attribute meets the target requirement, and generating a target judgment graph network corresponding to the set emission variation vector in the emission scene set. And the attribute type of the emission classification attribute corresponding to the preset emission data point is a set type.
In this embodiment, the carbon emission monitoring system first determines a particular emissions profile that includes a series of emission data points associated with a set emission variation vector. These data points may come from different emission sources, such as factories, power plants, or vehicles, and have been labeled according to their emission classification attributes. These emission data points are then used as network learning data to train and optimize the initial decision graph network associated with the emission scenario set.
The carbon emission monitoring system then loads the network learning data into the initial decision map network. The initial decision map network is a pre-built model that contains a series of nodes and connections corresponding to the set emission variation vector. The carbon emission monitoring system processes and analyzes the data through each node in the network and makes decisions according to the connection relation among the nodes. Finally, the initial decision graph network outputs a set of initial emissions classification attributes that represent the network's preliminary classification of data points.
After the initial emissions classification attribute is obtained, the carbon emissions monitoring system compares it to the actual emissions classification attribute corresponding to the network learning data. The carbon emission monitoring system finds that there is some deviation between the initial classification result and the actual attributes, which indicates that certain parameters or rules in the initial decision map network need to be adjusted. Thus, the carbon emission monitoring system begins to update the bias matching rules in the network. These rules describe which emission data points may be considered anomalies or deviations under different circumstances and give corresponding judgment conditions and thresholds. The carbon emission monitoring system reduces the deviation between the initial classification result and the actual attribute by adjusting the parameters and thresholds of the rules.
The carbon emission monitoring system repeatedly performs the above-described updating process, recalculates the initial emission classification attribute after each update, and compares it with the actual attribute. The carbon emission monitoring system is particularly concerned with preset emission data points, and emission classification attributes corresponding to the preset emission data points belong to set categories, so that the carbon emission monitoring system is important for accurately judging emission deviation. Thus, the weights and parameters associated with these points in the network are continually adjusted to ensure that their classification results are more accurate. This process continues until the weights of the preset emission data points meet the preset target requirements.
After multiple iterations and optimizations, the carbon emission monitoring system eventually generates a more accurate and reliable target decision graph network. This network is better able to capture and represent the relationship between the set emission variation vector and the emission data point, and the classification result for the preset emission data point is more accurate. The target judgment chart network becomes an important basis and support for subsequent carbon emission deviation detection.
Through the above steps, the carbon emission monitoring system can generate an optimized target decision graph network for a specific scenario based on the set emission variation vector and emission data point in the emission scenario set. The process is beneficial to improving the accuracy and efficiency of carbon emission monitoring, and provides important guarantee for subsequent carbon emission deviation detection.
In one possible implementation, step S1013-4 includes: integrating the set emission variation vector in the emission scenario set and the target deviation matching rule corresponding to the set emission variation vector in the emission scenario set to generate an alternative emission deviation list rule corresponding to the emission scenario set.
The carbon emission monitoring system has generated a target decision map network for a particular emission scenario set for which an emission variation vector is set, and has extracted a target deviation matching rule therefrom. These rules describe which emission data points may be considered anomalies or deviations under different circumstances, as well as the specific characteristics and judgment conditions of these deviations.
These target deviation matching rules now need to be further integrated into alternative emission deviation sheet rules for the entire emission scenario set. To this end, a review of the set emission variation vector in the emission scenario set is first made, which is a key indicator describing how the emission data points within the scenario vary and correlate.
Next, integration of the set emission variation vector with the target deviation matching rule is started. This process is analogous to integrating a specific, local set of rules (i.e., target bias matching rules) into a more comprehensive, more general set of rules (i.e., alternate discharge bias sheet rules). The carbon emission monitoring system ensures that these single rules can cover all possible emissions variations in the scenario set and remain consistent with the set emissions variation vector.
Finally, through integration and optimization, the carbon emission monitoring system generates a set of alternative emission deviation sheet rules for a specific emission scenario set. The single rules embody the core characteristics of the set emission variation vector, and also comprise key judgment conditions and threshold values in the target deviation matching rules, thereby providing direct and operable basis for subsequent carbon emission deviation detection. In practical applications, the carbon emission monitoring system may automatically monitor and determine real-time emission data according to these single rules to discover and report any potential carbon emission bias in time.
In one possible implementation, step S102 may include:
and S1021, carrying out integration strategy search on the alternative discharge deviation list rules corresponding to the discharge scene set, and determining the target alternative discharge deviation rules corresponding to the discharge scene set according to the integration strategy search data.
In this embodiment, the carbon emission monitoring system has generated a series of alternative emission deviation sheet rules for a particular emission scenario set. These single rules describe which emission data points may be considered anomalies or deviations under different circumstances and give specific judgment conditions and thresholds. Today, carbon emission monitoring systems need to explore an effective integration strategy to integrate these single rules into a more powerful and efficient target alternative emission deviation rule.
For the integration strategy search, the carbon emission monitoring system first analyzes the characteristics and application scope of each alternative emission deviation list rule. The carbon emission monitoring system considers similarities and complementarity between the regulations, as well as their performance in different emission scenarios. By analyzing this data, the carbon emission monitoring system can determine which rules have higher priority when integrated and determine the order and manner of merging between rules.
Step S1022, generating the customized carbon emission deviation detection model according to the merging rule sequence of the target alternative emission deviation rules corresponding to the emission scenario set.
After the integration strategy search, the carbon emission monitoring system obtains a series of key data and insight as to how the alternative emission deviation sheet rules are integrated. The carbon emission monitoring system uses this data to determine a final target alternative emission deviation rule.
For example, it is possible to combine individual single rules using weighted averaging, voting mechanisms, or rule-based reasoning. In this process, the carbon emission monitoring system ensures that the combined target alternative emission deviation rule can cover various potential emission deviation situations and can maintain higher accuracy and efficiency.
Once the target alternative emission deviation rule is determined, the carbon emission monitoring system may begin building a custom carbon emission deviation detection model. This custom carbon emission deviation detection model will be built based on the merged rule sequence of target alternative emission deviation rules and used for subsequent real-time emission data monitoring and deviation detection.
For example, each of the merging rules in the target candidate discharge deviation rule is first arranged in a determined sequence. The carbon emission monitoring system then utilizes advanced machine learning algorithms or model fusion techniques to integrate these rules into a unified detection model. This model can automatically process and analyze real-time emissions data and detect potential carbon emissions deviations based on judgment conditions and thresholds in target alternative emissions deviation rules.
Through the steps, the carbon emission monitoring system can generate a customized and efficient carbon emission deviation detection model based on a specific emission scene set, which is beneficial to improving the accuracy and efficiency of carbon emission monitoring and provides powerful support for timely finding and coping with potential carbon emission problems.
In one possible implementation, step S1021 may include:
And S1021-1, performing integration strategy search on the alternative discharge deviation list rule corresponding to the discharge scene set to generate integration strategy search data. The integration policy lookup data includes an alternative discharge deviation sheet rule and one or more target integration rules corresponding to the discharge scenario set.
In this embodiment, the carbon emission monitoring system has generated a series of alternative emission deviation sheet rules for a particular emission scenario set. These rules describe which emission data points may be considered anomalies or deviations under different circumstances, as well as the specific characteristics and judgment conditions of these deviations. Today, carbon emission monitoring systems need to explore an effective integration strategy to integrate these single rules into a more powerful and efficient target alternative emission deviation rule.
For the integration strategy search, the carbon emission monitoring system first analyzes the characteristics and application scope of each alternative emission deviation list rule. The carbon emission monitoring system considers similarities and complementarity between the regulations, as well as their performance in different emission scenarios. The carbon emission monitoring system also evaluates potential conflicts and consistency between different rules to determine their interactions at the time of integration. By analyzing these data, the carbon emission monitoring system generates integration strategy search data including an alternative emission deviation sheet rule and one or more target integration rules corresponding to the emission scenario set. These target integration rules describe how different single rules can be effectively combined together to improve the accuracy and efficiency of bias detection.
And S1021-2, taking the alternative discharge deviation list rule corresponding to the discharge scenario set and one or more target integration rules as the alternative discharge deviation rule corresponding to the discharge scenario set, and determining preset rule evaluation data of the alternative discharge deviation rule corresponding to the discharge scenario set. The preset rule evaluation data reflects an emission deviation detection effect of an alternative emission deviation rule corresponding to the emission scenario set.
The carbon emission monitoring system then combines the alternative emission deviation policy rules corresponding to the emission scenario set with one or more target integration rules to form an alternative emission deviation rule corresponding to the emission scenario set. These rules combine the features of a single rule with the advantages of target integration rules to provide more accurate, comprehensive deviation detection.
To evaluate the effectiveness of these alternative emission deviation rules, the carbon emission monitoring system determines a set of preset rule evaluation data. These data reflect the effect of the rules on emissions deviation detection, which may include indicators of detection accuracy, false positive rate, false negative rate, etc. The carbon emission monitoring system collects these data through simulation or actual testing to objectively evaluate alternative emission deviation rules.
And S1021-3, outputting a rule that the preset rule evaluation data is larger than a set value in the alternative discharge deviation rules corresponding to the discharge scene set as a target alternative discharge deviation rule corresponding to the discharge scene set.
After the preset rule evaluation data are collected, the carbon emission monitoring system screens the alternative emission deviation rules corresponding to the emission scenario set. The carbon emission monitoring system sets one or more thresholds, and only those rules for which the preset rule evaluation data is greater than the set values are considered valid and output as target alternative emission deviation rules for the emission scenario set.
This process ensures that only well-behaved rules are preserved for subsequent carbon emission deviation detection. In this way, the carbon emission monitoring system can automatically optimize and improve its deviation detection capability, increasing accuracy and efficiency. Finally, the carbon emission monitoring system outputs a set of target alternative emission deviation rules aiming at a specific emission scenario set, and powerful support is provided for subsequent carbon emission monitoring and deviation detection.
In one possible embodiment, step S1021-1 may include:
1. And taking the alternative discharge deviation list rule corresponding to the discharge scenario set as a current undetermined rule, and constructing a current undetermined rule sequence corresponding to the discharge scenario set.
In this embodiment, the carbon emission monitoring system first identifies an alternate emission deviation sheet rule corresponding to a particular emission scenario set. The carbon emission monitoring system marks these alternative emission deviation sheet rules as currently pending rules, meaning that they are currently being considered for integration into more complex rules.
Next, the carbon emission monitoring system arranges the currently pending rules into a sequence, i.e., a "currently pending rule sequence," in some logical order (e.g., time of generation, importance, or relevance of the rules). This sequence provides the basis for subsequent integration strategy search.
2. And integrating any two currently pending rules in the currently pending rule sequence to generate integrated rule data. The integration rule data comprises one or more currently pending integration rules corresponding to the currently pending rule sequence.
For example, the carbon emission monitoring system selects any two rules from the currently pending rule sequence for integration. The way of integration may be to merge the conditions of two rules, or one rule as a pre-condition for the other rule, etc. This integration is heuristic in order to explore whether the combination between different rules can improve the accuracy of carbon emission deviation detection.
After integration, the carbon emission monitoring system generates a new set of rules, referred to as currently pending integration rules, which form the integration rule data. These data are temporary and require further evaluation of their effectiveness.
3. And determining a target integration rule corresponding to the currently pending rule sequence from one or more currently pending integration rules corresponding to the currently pending rule sequence. The target integration rule is a rule that the current rule evaluation data accords with target requirements in one or more current integration rules corresponding to the current rule sequence. The current rule evaluation data reflects an emission deviation detection effect of the current pending integration rule.
The generated currently pending integration rules are then evaluated. The basis of the evaluation is a set of indicators called "current rule evaluation data" reflecting the effect of the integration rules on detecting carbon emission deviations, such as accuracy, recall, etc.
Comparing the indexes with preset target requirements, and finding out the currently pending integration rules meeting or exceeding the target requirements. These selected rules are called "target integration rules" and they are the basis for the next integration.
4. And taking the target integrated rule corresponding to the currently pending rule sequence as a currently pending rule in the currently pending rule sequence.
The carbon emission monitoring system updates the currently pending rule sequence, adds the selected target integration rule into the sequence, and replaces or expands the original currently pending rule. In this way, rules in the sequence evolve gradually from simple single rules to more complex, more efficient integration rules.
5. And circularly executing the step of integrating any two current undetermined rules in the current undetermined rule sequence to a target integration rule corresponding to the current undetermined rule sequence as the current undetermined rule in the current undetermined rule sequence until the current undetermined rule which accords with the preset length exists in the current undetermined rule sequence, and generating scene candidate integration search data corresponding to the emission scene set.
The above-described integration and evaluation process is repeated, continually combining and optimizing rules in the currently pending rule sequence. This process is iterative, each iteration generating a new target integration rule and updating the currently pending rule sequence.
This process continues until the complexity of one or more rules (e.g., length of rule, number of conditions involved, etc.) present in the currently pending rule sequence meets or exceeds a preset length criterion. At this point, the carbon emission monitoring system considers that a sufficiently complex and effective integration rule has been found, and then stops iterating.
Finally, the carbon emission monitoring system generates a set of scenario candidate integrated search data for a particular emission scenario set. The data contains all effective integration rules and evaluation data thereof generated in the integration process, and provides a rich candidate rule set for subsequent carbon emission deviation detection.
The step of determining the current rule evaluation data includes:
step a110, an emission variation evaluation data sequence is acquired. The emission fluctuation evaluation data sequence comprises at least two groups of emission fluctuation evaluation data and evaluation feature vectors corresponding to each group of emission fluctuation evaluation data.
In this embodiment, the carbon emission monitoring system first collects a series of emission variation evaluation data. These emissions variation assessment data may be extracted from historical records or may be monitored in real time. Each set of emissions variation assessment data includes the amount of carbon emissions and their variation for a particular emission source at a certain point in time or period of time.
In addition to the emission variation evaluation data itself, an evaluation feature vector corresponding to each set of data is acquired. These evaluation feature vectors include various factors that affect carbon emission variation, such as the type of emission source, operating conditions, environmental conditions, and the like. Evaluating the feature vector provides additional information for analyzing the cause of emissions variation.
And step A120, searching in the emission variation evaluation data sequence according to the currently pending integration rule to obtain target emission variation evaluation data.
The carbon emission monitoring system may then search through the emission variation assessment data sequence using the currently pending integration rules. These currently pending integration rules are generated based on previous integration strategy searching steps that describe specific emissions deviation patterns or anomalies.
By applying these rules to the emission variation assessment data sequence, those data points that meet the rules are identified. These data points are referred to as target emissions variation assessment data, which may be indicative of potential carbon emission deviations or anomalies.
And step A130, determining an accuracy index and a coverage index corresponding to the currently pending integration rule according to the evaluation feature vector corresponding to the target emission variation evaluation data.
After identifying the target emission variation evaluation data, the carbon emission monitoring system further analyzes the evaluation feature vectors corresponding to these data. The accuracy of the currently pending integration rules in identifying potential carbon emission deviations is assessed by comparing various factors in the assessment feature vectors.
The accuracy index may be determined in a number of ways, such as calculating a percentage of the rule's correct recognition bias, or using a more complex statistical model to evaluate the predictive performance of the rule.
In addition to the accuracy index, a coverage index is calculated to assess how well the rules cover all possible carbon emission bias conditions. The coverage index may be calculated based on the distribution of the target emissions variation assessment data in the overall data sequence.
For example, a particular implementation of determining the accuracy index and coverage index corresponding to the currently pending integration rule may involve multiple steps and computations. The following is one possible implementation procedure, including the associated calculation formula:
and (3) a step of: determining an accuracy index
The accuracy index is used for evaluating the accuracy degree of the currently pending integration rule when identifying the target emission variation evaluation data. The accuracy index may be defined by calculating a true case rate (True Positive Rate, TPR) and a false case rate (False Positive Rate, FPR).
Real and false positive examples are calculated:
true examples (True posives, TP): is correctly identified by the rules as the number of samples of the target emission variation evaluation data.
False Positives (FP): the number of samples that are falsely identified as target emission variation evaluation data by the rule.
Calculating the true case rate and the false case rate:
true example rate (TPR): TP/(tp+fn), where FN is the number of samples that are regularly and erroneously identified as non-target emission variation assessment data (false counter-example).
False Positive Rate (FPR): FP/(fp+tn), where TN is the number of samples correctly identified by the rule as non-target emission variation evaluation data (true counter example).
Defining an accuracy index:
some combination of TPR and FPR may be used to define an accuracy index, such as an F1 score: f1 =2× (tpr×precision)/(tpr+precision), where precision=tp/(tp+fp).
And II: determining coverage index
The coverage index is used to evaluate the ability of the currently pending integration rules to cover all possible target emissions variation evaluation data.
Calculating the number of samples covered by the rule:
number of Samples Covered by rule (Covered Samples): the number of samples that are identified by the rules as target emissions variation assessment data, whether or not the identification is correct (i.e., tp+fp).
Calculating a coverage index:
coverage Index (Coverage Index): covered Samples/Total Samples, where Total Samples is the number of Samples of all possible target emission variation assessment data.
Finally, the accuracy index and the coverage index can be combined to form a comprehensive evaluation index for evaluating the overall performance of the currently pending integration rule. For example, the two indices may be combined using a weighted sum or product.
It should be noted that the above steps and calculation formulas provide only one possible implementation. In practical application, the system can be adjusted and optimized according to specific requirements and data characteristics. In addition, other evaluation metrics and methods are contemplated for more fully evaluating the performance of the currently pending integration rules.
And step A140, determining the current rule evaluation data of the current integration rule according to the accuracy index and the coverage index corresponding to the current integration rule.
And finally, comprehensively considering the accuracy index and the coverage index of the currently pending integrated rule by the carbon emission monitoring system to generate current rule evaluation data. These data provide a comprehensive assessment of the performance of the rules, helping the carbon emission monitoring system determine which rules may be more effective when integrated into the final carbon emission deviation detection model.
The current rule evaluation data may be a composite score or a set of specific indicators, such as accuracy, coverage, rule complexity, etc. The carbon emission monitoring system screens out the currently pending integration rule with the best performance according to the evaluation data, and provides powerful support for subsequent carbon emission monitoring and deviation detection.
In one possible embodiment, the candidate carbon emission information stream includes a candidate emission variation vector and a candidate emission data point corresponding to the candidate emission variation vector.
The method further comprises the steps of:
and step S130, when the carbon emission deviation detection result reflects that a target candidate emission variation vector matched with a set emission variation vector in the customized carbon emission deviation detection model exists in the candidate emission variation vector, and a candidate emission data point corresponding to the target candidate emission variation vector is matched with a target deviation matching rule, determining that the candidate carbon emission information flow is deviation carbon emission data.
In this embodiment, the candidate carbon emission information stream collected in real time by the carbon emission monitoring system includes candidate emission variation vectors, each of which represents a variation of the carbon emission over a period of time. At the same time, each candidate emission variation vector corresponds to one or more candidate emission data points that record specific carbon emissions and associated additional information such as time stamp, emission source, etc.
Thus, the carbon emission monitoring system analyzes the candidate carbon emission information stream using the customized carbon emission deviation detection model. The custom carbon emission deviation detection model includes a series of set emission variation vectors representing what is considered a normal or abnormal carbon emission variation pattern.
First, the candidate emission variation vector is compared with the set emission variation vector to find whether there is a matching target candidate emission variation vector. If there is a match, it is further checked whether the candidate emission data points corresponding to these target candidate emission variation vectors satisfy the target bias match rule. These rules may be set based on historical data, expert knowledge, or regulatory requirements for further confirming the existence of the deviation.
The candidate carbon emission information stream is determined to be offset carbon emission data only if the candidate emission variation vector matches the set emission variation vector and the corresponding candidate emission data point also satisfies the target offset matching rule.
And step S140, loading the deviation carbon emission data to a significance data extraction network for feature filtering, and generating a feature filtering result output by the significance data extraction network.
Wherein the saliency data extraction network comprises at least one feature filtering network. The characteristic filtering result comprises target carbon emission variation data and a target characteristic filtering network corresponding to the target carbon emission variation data. The target carbon emission variation data is data conforming to at least one characteristic filter network in the deviation carbon emission data.
In this embodiment, once the offset carbon emission data is determined, the carbon emission monitoring system loads the data into a significance data extraction network. This network contains at least one feature filter network for more in-depth analysis and processing of the offset carbon emission data.
The function of the feature filter network is to extract the most important features from the offset carbon emission data that are most directly related to the carbon emission offset. This may provide a better understanding of the nature and cause of the deviation and may also provide more useful information for subsequent processing and decision making.
After feature filtering, the saliency data extraction network generates feature filtering results. This result includes target carbon emission variation data and a target feature filter network corresponding to the variation data. The target carbon emission variation data are data conforming to at least one characteristic filter network among the deviated carbon emission data, which are considered to be the most relevant and important parts with respect to the deviation of carbon emission.
Thus, through the feature filtering function of the significance data extraction network, the carbon emission monitoring system can more accurately identify and understand carbon emission deviation, and more reliable support is provided for subsequent countermeasures and corrective measures.
In the embodiment of fig. 2, a carbon emission monitoring system 100 is provided, which includes a processor 1001 and a memory 1003, and program code stored on the memory 1003, wherein the processor 1001 executes the program code to implement the steps of the carbon emission monitoring method applied to the digital platform system.
The carbon emission monitoring system 100 shown in fig. 2 includes: a processor 1001 and a memory 1003. The processor 1001 is coupled to the memory 1003, such as via a bus 1002. Optionally, the carbon emission monitoring system 100 may further include a transceiver 1004, the transceiver 1004 may be used for data interaction between the carbon emission monitoring system 100 and other carbon emission monitoring systems 100, such as transmission of data and/or reception of data, etc. It should be noted that, the transceiver 1004 is not limited to one in actual schedule, and the structure of the carbon emission monitoring system 100 is not limited to the embodiment of the present application.
The processor 1001 may be a CPU (Central Processing Unit ), general purpose processor, DSP (Digital Signal Processor, data signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic device, transistor logic device, hardware components, or any combination thereof. Which may implement or perform the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. The processor 1001 may also be a combination that implements computing functionality, such as a combination comprising one or more microprocessors, a combination of a DSP and a microprocessor, or the like.
Bus 1002 may include a path to transfer information between the components. Bus 1002 may be a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (ExtendedIndustry Standard Architecture ) bus, among others. The bus 1002 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in fig. 2, but not only one bus or one type of bus.
The Memory 1003 may be, but is not limited to, ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, RAM (Random Access Memory ) or other type of dynamic storage device that can store information and instructions, EEPROM (Electrically ErasableProgrammable Read Only Memory ), CD-ROM (Compact DiscRead Only Memory, compact disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store program code and that can be Read by a computer.
The memory 1003 is used for storing program codes for executing the embodiments of the present application, and is controlled to be executed by the processor 1001. The processor 1001 is configured to execute the program code stored in the memory 1003 to implement the steps shown in the foregoing method embodiment.
The embodiments of the present application provide a computer readable storage medium having a program code stored thereon, which when executed by a processor, implements the steps of the foregoing method embodiments and corresponding content.
The foregoing is merely an optional implementation manner of the implementation scenario of the application, and it should be noted that, for those skilled in the art, other similar implementation manners according to the technical ideas of the application are adopted without departing from the technical ideas of the application, and also belong to the protection scope of the embodiments of the application.
Claims (10)
1. A carbon emission monitoring method for a digital platform system, the method comprising:
acquiring candidate carbon emission information streams stored in a digital platform system based on a blockchain technology, wherein the candidate carbon emission information streams are carbon emission data generated by carrying out interconnection and interaction acquisition with various energy utilization devices and energy consumption devices;
Analyzing the candidate carbon emission information stream according to a customized carbon emission deviation detection model to generate a carbon emission deviation detection result; the generating step of the customized carbon emission deviation detection model comprises the following steps:
generating an alternative emission deviation list rule corresponding to the emission scene set based on emission data points corresponding to set emission variation vectors in the predefined emission scene set and emission classification attributes corresponding to the emission data points;
and carrying out integration strategy searching on the alternative emission deviation list rules corresponding to the emission scene set, and generating a customized carbon emission deviation detection model according to the integration strategy searching data.
2. The method for carbon emission monitoring applied to a digital platform system according to claim 1, wherein the generating an alternative emission deviation list rule corresponding to the emission scenario set based on emission data points corresponding to set emission variation vectors in the predefined emission scenario set and emission classification attributes corresponding to the emission data points comprises:
acquiring a plurality of set emission variation vectors; the set emission variation vector corresponds to an emission data point carrying an emission classification attribute;
Performing emission scene division on a plurality of the set emission variation vectors according to a set emission theme to generate a predefined emission scene set;
and generating a target deviation matching rule corresponding to the set emission variation vector in the emission scene set based on the emission data point corresponding to the set emission variation vector in the emission scene set and the emission classification attribute corresponding to the emission data point, and generating an alternative emission deviation list rule corresponding to the emission scene set according to the target deviation matching rule.
3. The method for monitoring carbon emissions applied to a digital platform system according to claim 2, wherein the generating a target deviation matching rule corresponding to a set emission variation vector in the emission scenario set based on an emission data point corresponding to the set emission variation vector in the emission scenario set and an emission classification attribute corresponding to the emission data point, and generating an alternative emission deviation list rule corresponding to the emission scenario set according to the target deviation matching rule, comprises:
acquiring an initial judgment chart network corresponding to a set emission variation vector in the emission scene set;
taking the emission data points corresponding to the set emission variation vectors in the emission scene set as network learning data corresponding to the set emission variation vectors in the emission scene set, and performing parameter learning on an initial judgment graph network corresponding to the set emission variation vectors in the emission scene set according to the network learning data and the emission classification attribute corresponding to the network learning data to generate a target judgment graph network corresponding to the set emission variation vectors in the emission scene set;
Taking the deviation matching rule of the network members in the target judgment graph network as a target deviation matching rule corresponding to a set emission variation vector in the emission scene set;
and generating an alternative discharge deviation list rule corresponding to the discharge scenario set according to the target deviation matching rule.
4. The carbon emission monitoring method as defined in claim 3, wherein the generating the target determination map network for the set emission variation vector in the emission scenario set by using the emission data point for the set emission variation vector in the emission scenario set as the network learning data for the set emission variation vector in the emission scenario set, and performing parameter learning on the initial determination map network for the set emission variation vector in the emission scenario set according to the network learning data and the emission classification attribute for the network learning data comprises:
taking the emission data points corresponding to the set emission variation vectors in the emission scene set as network learning data corresponding to the set emission variation vectors in the emission scene set, loading the network learning data corresponding to the set emission variation vectors in the emission scene set into an initial judgment graph network corresponding to the set emission variation vectors in the emission scene set to make a decision, and generating initial emission classification attributes output by the initial judgment graph network;
Updating the deviation matching rule of the network members in the initial judgment graph network according to the emission classification attribute corresponding to the network learning data until the weight of the preset emission data point in the initial emission classification attribute meets the target requirement, and generating a target judgment graph network corresponding to the set emission variation vector in the emission scene set; and the attribute type of the emission classification attribute corresponding to the preset emission data point is a set type.
5. The method for monitoring carbon emissions applied to a digital platform system according to claim 3, wherein the generating the alternative emission deviation list rule corresponding to the emission scenario set according to the target deviation matching rule comprises:
integrating the set emission variation vector in the emission scenario set and the target deviation matching rule corresponding to the set emission variation vector in the emission scenario set to generate an alternative emission deviation list rule corresponding to the emission scenario set.
6. The method for monitoring carbon emissions applied to a digital platform system according to claim 1, wherein the performing an integration strategy search on the alternative emission deviation list rule corresponding to the emission scenario set, generating a customized carbon emission deviation detection model according to the integration strategy search data, comprises:
Performing integration strategy searching on the alternative discharge deviation list rules corresponding to the discharge scenario set, and determining target alternative discharge deviation rules corresponding to the discharge scenario set according to the integration strategy searching data;
and generating the customized carbon emission deviation detection model according to a merging rule sequence of target alternative emission deviation rules corresponding to the emission scene set.
7. The method for monitoring carbon emissions applied to a digital platform system according to claim 6, wherein the performing an integration policy search on the candidate emission deviation list rule corresponding to the emission scenario set, and determining the target candidate emission deviation rule corresponding to the emission scenario set according to the integration policy search data, comprises:
performing integration strategy search on the alternative discharge deviation list rules corresponding to the discharge scenario set to generate integration strategy search data; the integration strategy searching data comprises an alternative discharge deviation list rule and one or more target integration rules corresponding to the discharge scenario set;
taking the alternative discharge deviation list rule corresponding to the discharge scenario set and one or more target integration rules as the alternative discharge deviation rule corresponding to the discharge scenario set, and determining preset rule evaluation data of the alternative discharge deviation rule corresponding to the discharge scenario set; the preset rule evaluation data reflects the discharge deviation detection effect of the alternative discharge deviation rule corresponding to the discharge scenario set;
And outputting the rule of which the preset rule evaluation data is larger than a set value in the alternative discharge deviation rules corresponding to the discharge scene set as a target alternative discharge deviation rule corresponding to the discharge scene set.
8. The method for monitoring carbon emissions applied to a digital platform system according to claim 7, wherein the performing an integration policy search on the alternative emission deviation list rule corresponding to the emission scenario set, generating integration policy search data comprises:
taking the alternative discharge deviation list rule corresponding to the discharge scenario set as a current undetermined rule, and constructing a current undetermined rule sequence corresponding to the discharge scenario set;
integrating any two current undetermined rules in the current undetermined rule sequence to generate integrated rule data; the integration rule data comprises one or more currently pending integration rules corresponding to the currently pending rule sequence;
determining a target integration rule corresponding to the currently pending rule sequence from one or more currently pending integration rules corresponding to the currently pending rule sequence; the target integration rule is a rule that the current rule evaluation data accords with target requirements in one or more current integration rules corresponding to the current rule sequence; the current rule evaluation data reflects the emission deviation detection effect of the current undetermined integration rule;
Taking a target integration rule corresponding to the currently pending rule sequence as a currently pending rule in the currently pending rule sequence;
circularly executing the step of integrating any two current undetermined rules in the current undetermined rule sequence to a target integration rule corresponding to the current undetermined rule sequence as the current undetermined rule in the current undetermined rule sequence until the current undetermined rule which accords with the preset length exists in the current undetermined rule sequence, and generating scene candidate integration search data corresponding to the emission scene set;
the step of determining the current rule evaluation data includes:
acquiring an emission variation evaluation data sequence; the emission variation evaluation data sequence comprises at least two groups of emission variation evaluation data and evaluation feature vectors corresponding to each group of emission variation evaluation data;
searching in the emission variation evaluation data sequence according to the currently pending integration rule to obtain target emission variation evaluation data;
determining an accuracy index and a coverage index corresponding to the currently pending integration rule according to the evaluation feature vector corresponding to the target emission variation evaluation data;
And determining the current rule evaluation data of the current integration rule according to the accuracy index and the coverage index corresponding to the current integration rule.
9. The carbon emission monitoring method for use in a digital platform system according to any one of claims 1-8, wherein the candidate carbon emission information stream includes a candidate emission variation vector and a candidate emission data point corresponding to the candidate emission variation vector;
the method further comprises the steps of:
when the carbon emission deviation detection result reflects that a target candidate emission variation vector matched with a set emission variation vector in the customized carbon emission deviation detection model exists in the candidate emission variation vector, and a candidate emission data point corresponding to the target candidate emission variation vector is matched with a target deviation matching rule, determining that the candidate carbon emission information flow is deviation carbon emission data;
loading the deviation carbon emission data to a significance data extraction network for feature filtering to generate a feature filtering result output by the significance data extraction network;
wherein the saliency data extraction network comprises at least one feature filtering network; the characteristic filtering result comprises target carbon emission variation data and a target characteristic filtering network corresponding to the target carbon emission variation data; the target carbon emission variation data is data conforming to at least one characteristic filter network in the deviation carbon emission data.
10. A carbon emissions monitoring system comprising a processor and a machine-readable storage medium having stored therein machine-executable instructions loaded and executed by the processor to implement the carbon emissions monitoring method of any of claims 1-9 applied to a digital platform system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410178403.9A CN117726053A (en) | 2024-02-09 | 2024-02-09 | Carbon emission monitoring method and system applied to digital platform system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410178403.9A CN117726053A (en) | 2024-02-09 | 2024-02-09 | Carbon emission monitoring method and system applied to digital platform system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117726053A true CN117726053A (en) | 2024-03-19 |
Family
ID=90209239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410178403.9A Pending CN117726053A (en) | 2024-02-09 | 2024-02-09 | Carbon emission monitoring method and system applied to digital platform system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117726053A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258247A (en) * | 2013-05-17 | 2013-08-21 | 广东工业大学 | Carbon emission optimization method based on target cascading analytical method |
| CN108288855A (en) * | 2018-01-23 | 2018-07-17 | 国电南京自动化股份有限公司 | A kind of dynamic strategy simulation and optimization method based on micro-capacitance sensor |
| US20210064687A1 (en) * | 2019-08-30 | 2021-03-04 | Fujitsu Limited | Local search with global view for large scale combinatorial optimization |
| CN113191063A (en) * | 2021-04-25 | 2021-07-30 | 华北电力大学（保定） | Scene generation method considering wind and light prediction error |
| CN114493213A (en) * | 2022-01-18 | 2022-05-13 | 上海祺鲲信息科技有限公司 | Carbon emission data acquisition and processing method based on Internet of things |
| CN114742294A (en) * | 2022-04-09 | 2022-07-12 | 广州网文三维数字技术有限公司 | Neural network algorithm for carbon emission prediction |
| CN115049095A (en) * | 2022-03-23 | 2022-09-13 | 上海创能国瑞新能源科技股份有限公司 | Dynamic optimization system for improving energy efficiency and carbon emission of equipment and related device |
| CN115755614A (en) * | 2022-11-25 | 2023-03-07 | 南方电网数字平台科技(广东)有限公司 | Energy consumption optimization regulation and control method and device based on carbon emission monitoring |
| CN116401238A (en) * | 2023-04-10 | 2023-07-07 | 中国工商银行股份有限公司 | Deviation monitoring method, apparatus, device, storage medium and program product |
| CN117391292A (en) * | 2023-10-17 | 2024-01-12 | 华能曲阜热电有限公司 | Carbon emission energy-saving management analysis system and method |
| CN117408536A (en) * | 2023-12-15 | 2024-01-16 | 山东省物化探勘查院 | AI-based underwater mapping real-time analysis system |
-
2024
- 2024-02-09 CN CN202410178403.9A patent/CN117726053A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258247A (en) * | 2013-05-17 | 2013-08-21 | 广东工业大学 | Carbon emission optimization method based on target cascading analytical method |
| CN108288855A (en) * | 2018-01-23 | 2018-07-17 | 国电南京自动化股份有限公司 | A kind of dynamic strategy simulation and optimization method based on micro-capacitance sensor |
| US20210064687A1 (en) * | 2019-08-30 | 2021-03-04 | Fujitsu Limited | Local search with global view for large scale combinatorial optimization |
| CN113191063A (en) * | 2021-04-25 | 2021-07-30 | 华北电力大学（保定） | Scene generation method considering wind and light prediction error |
| CN114493213A (en) * | 2022-01-18 | 2022-05-13 | 上海祺鲲信息科技有限公司 | Carbon emission data acquisition and processing method based on Internet of things |
| CN115049095A (en) * | 2022-03-23 | 2022-09-13 | 上海创能国瑞新能源科技股份有限公司 | Dynamic optimization system for improving energy efficiency and carbon emission of equipment and related device |
| CN114742294A (en) * | 2022-04-09 | 2022-07-12 | 广州网文三维数字技术有限公司 | Neural network algorithm for carbon emission prediction |
| CN115755614A (en) * | 2022-11-25 | 2023-03-07 | 南方电网数字平台科技(广东)有限公司 | Energy consumption optimization regulation and control method and device based on carbon emission monitoring |
| CN116401238A (en) * | 2023-04-10 | 2023-07-07 | 中国工商银行股份有限公司 | Deviation monitoring method, apparatus, device, storage medium and program product |
| CN117391292A (en) * | 2023-10-17 | 2024-01-12 | 华能曲阜热电有限公司 | Carbon emission energy-saving management analysis system and method |
| CN117408536A (en) * | 2023-12-15 | 2024-01-16 | 山东省物化探勘查院 | AI-based underwater mapping real-time analysis system |
Non-Patent Citations (1)
| Title |
|---|
| YI ZHOU ET AL.: "Regression analysis and driving force model building of CO2 emissions in China", 《SCIENTIFIC REPORTS》, 24 March 2021 (2021-03-24), pages 1 - 14 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Capozzoli et al. | Fault detection analysis using data mining techniques for a cluster of smart office buildings | |
| KR101713985B1 (en) | Method and apparatus for prediction maintenance | |
| CN108596229A (en) | Online abnormal monitoring, diagnosing method and system | |
| CN115578015A (en) | Sewage treatment overall process supervision method and system based on Internet of things and storage medium | |
| CN103617110A (en) | Server device condition maintenance system | |
| CN114267178B (en) | Intelligent operation maintenance method and device for station | |
| González-Vidal et al. | IoT for water management: Towards intelligent anomaly detection | |
| CN106570567A (en) | Main network maintenance multi-constraint multi-target evaluation expert system and optimization method | |
| CN116862081B (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN108763966B (en) | Tail gas detection cheating supervision system and method | |
| CN116433045A (en) | Accurate supply management method and system for electric power emergency materials | |
| Shitharth et al. | A new probabilistic relevancy classification (PRC) based intrusion detection system (IDS) for SCADA network | |
| CN116720324A (en) | Traction substation key equipment fault early warning method and system based on prediction model | |
| CN117370818B (en) | Intelligent diagnosis method and intelligent environment-friendly system for water supply and drainage pipe network based on artificial intelligence | |
| Killeen et al. | Unsupervised dynamic sensor selection for iot-based predictive maintenance of a fleet of public transport buses | |
| Heydarian et al. | Automated benchmarking and monitoring of an earthmoving operation's carbon footprint using video cameras and a greenhouse gas estimation model | |
| CN117726053A (en) | Carbon emission monitoring method and system applied to digital platform system | |
| CN114397842B (en) | Intelligent inspection reinforcement method for safety of power monitoring network | |
| CN116205528A (en) | Illegal construction identification method and system based on construction site power data | |
| CN114168409A (en) | Service system running state monitoring and early warning method and system | |
| CN113269247B (en) | Training method and device for complaint early warning model, computer equipment and storage medium | |
| CN115237606B (en) | Edge calculation control system | |
| CN117786372B (en) | Distributed photovoltaic power generation data processing method and system based on machine learning | |
| CN116579677B (en) | Full life cycle management method and system for high-speed railway electric service vehicle-mounted equipment | |
| CN115270140B (en) | Software security vulnerability management method and system |
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 |