Carbon emission reduction optimization system based on knowledge graph

Abstract

The invention discloses a carbon emission reduction optimization system based on a knowledge graph, and relates to the technical field of carbon emission reduction optimizing.After an emission point of a current technological process is identified, corresponding emission detection data are obtained through detection, pollution risk assessment is conducted, emission coefficients are generated according to the emission risk data of the emission point, and the emission coefficients are calculated according to the emission coefficients. Screening out risk points from the plurality of emission points according to the emission coefficients, matching corresponding treatment strategies for emissions of the risk points, and sequentially treating the emissions of the risk points; and after abnormal emission is collected and detected, if the reduction proportion of abnormal intensity is lower than expectation, an optimization degree is generated according to carbon emission data at each emission point, control parameters of the production and processing technology are optimized, and the optimized production and processing technology is executed. Through combination and selection of a plurality of optimization modes, reduction of carbon emission is realized, and feedback of a carbon emission reduction process in a current stage is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission reduction optimization, and specifically to a carbon emission reduction optimization system based on a knowledge graph. Background Art

[0002] Carbon emission reduction refers to the process of taking a series of measures, including saving energy, improving energy utilization efficiency, developing clean energy, promoting low-carbon technologies and products, and strengthening carbon sink construction (such as afforestation, etc.), to reduce the emissions of greenhouse gases such as carbon dioxide generated in human social and economic activities. These actions aim to mitigate the trend of global warming, reduce the occurrence of extreme weather events, protect the ecological environment, and achieve a win-win situation for environmental protection and economic and social development.

[0003] In a Chinese invention patent with the application publication number CN114580759A, a low-carbon emission reduction assessment system for cities is disclosed, including: a collection module: used to obtain and collect parameter indicators of a target city; wherein, the parameter indicators at least include driving force indicators, pressure indicators, response indicators, state indicators, and impact indicators; an evaluation index system module: used to select evaluation indicators through preset low-carbon emission reduction standard criteria, and establish an evaluation index system through the evaluation indicators; an evaluation module: used to evaluate the parameter indicators based on the evaluation index system and generate an evaluation result.

[0004] Combined with the above application and the content in the prior art: When industrial products are being produced and processed, a large amount of carbon emissions are generated due to the direct or indirect use of various fossil fuels, which in turn causes environmental pollution or triggers climate problems. Therefore, in order to control and optimize excessive carbon emissions, after implementing corresponding emission reduction strategies, it is necessary to evaluate the carbon emissions generated during the production process.

[0005] Existing carbon emission reduction effect evaluation methods usually collect carbon emission data before and after emission reduction optimization, and judge whether the current stage of emission reduction has achieved the expected effect through the carbon emission data. However, the effectiveness of this emission reduction evaluation is relatively single. Especially when there are still many abnormal emission behaviors at present, single-level emission reduction evaluation is difficult to achieve the expected positive feedback effect, resulting in the inability to sustain the subsequent carbon emission reduction effect and being unable to contribute to continuous carbon emission reduction.

[0006] Therefore, the present invention provides a carbon emission reduction optimization system based on a knowledge graph. Summary of the Invention

[0007] (I) Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides a carbon emission reduction optimization system based on a knowledge graph. After identifying the emission points in the current process flow, the corresponding emission detection data is detected and obtained, and then a pollution risk assessment is carried out. An emission coefficient is generated from the emission risk data of the emission points. Based on the emission coefficient, risk points are screened out among several emission points. After matching corresponding treatment strategies for the emissions of the risk points, the emissions of each risk point are processed in turn; after abnormal emissions are collected and detected, if the reduction ratio of the abnormal density is lower than expected, an optimization degree is generated from the carbon emission data at each emission point, the control parameters of the production and processing process are optimized, and the optimized production and processing process is executed. By combining and selecting various optimization methods, the reduction of carbon emissions is achieved; thus, the technical problems in the background art are solved.

[0008] (2) Technical solutions To achieve the above objectives, the present invention is realized through the following technical solutions: A carbon emission reduction optimization system based on a knowledge graph, including an abnormal analysis unit, which, after identifying abnormal emission data in the production and processing process, generates an abnormal density from the abnormal emission data If the abnormal density exceeds the preset expectation, an abnormal identification instruction is sent to the outside; among them, the abnormal density is generated from the abnormal emission data in the abnormal emission data set where

[0009] Weight coefficient: is the number of abnormal emission events, is the severity value of the abnormal emission event, is the corresponding mean value; is the time interval between the i-th and j-th abnormal emission events, is the mean value of the i-th time interval; A risk assessment unit, which, after identifying the emission points in the current process flow, detects and obtains the corresponding emission detection data and then conducts a pollution risk assessment. If the obtained risk score exceeds the preset risk threshold, an alarm instruction is sent to the outside; An emission point screening unit, which generates an emission coefficient from the emission risk data of the emission points Based on the emission coefficient risk points are screened out among several emission points. After matching corresponding treatment strategies for the emissions of the risk points, the emissions of each risk point are processed in turn; A fault detection unit, which conducts fault detection on the production and processing equipment. If there is an operation fault, a corresponding maintenance plan is given by the fault maintenance knowledge graph, and after a maintenance interval that meets the constraint conditions, the faulty equipment is maintained; Optimization unit, after collecting and detecting abnormal emissions, if the reduction ratio of abnormal density is lower than expected, the optimization degree is generated from the carbon emission data at each emission point , the control parameters of the production and processing process are optimized, and the optimized production and processing process is executed.

[0010] Furthermore, collect the carbon emission data during the production and processing process, use the carbon emission data as input, and perform abnormal analysis using the trained abnormal emission identification model to identify and obtain the corresponding abnormal emission data, including the type, severity, and time node of the abnormal emission, and summarize and generate an abnormal emission data set.

[0011] Furthermore, after receiving the abnormal identification instruction, query and obtain the current production and processing process and then perform process analysis to identify the emission points at each stage; the sensor network detects and obtains the corresponding emission detection data at the emission points, obtains the carbon emission data of each emission point, uses the detected carbon emission data as input, and uses the trained emission risk assessment model to perform pollution risk assessment to obtain the corresponding pollution risk score.

[0012] Furthermore, after receiving the alarm instruction, determine the sensitivity of the emission point to the production process through sensitivity analysis and obtain the corresponding sensitivity coefficient; use the time node when the alarm instruction is received as the alarm node, and based on the risk score of the risk point, the distribution of the alarm nodes, and the sensitivity coefficient generate an emission coefficient , if the obtained emission coefficient exceeds the preset emission threshold, the corresponding emission point is regarded as a risk point.

[0013] Furthermore, the method for generating the emission coefficient based on the risk score of the risk point, the distribution of the alarm nodes, and the sensitivity coefficient is as follows:

[0014] In the formula: is the risk score of the i-th risk point at time t, is the carbon emission monitoring period, is the risk threshold, takes the value of 2.718, is the time decay coefficient, and μ is the distance decay coefficient; is the j-th alarm node at time t, is the time interval between the i-th and j-th alarm nodes, and n is the number of alarm nodes, is the risk score on the j-th alarm node; weight coefficient, , , .

[0015] Furthermore, identify the carbon emission data and other emission data at the risk points, obtain the emission categories such as by-products and waste, pre-obtain several emission treatment strategies, and match the corresponding emission treatment strategies for the emissions by the emission treatment strategy library; Construct an electronic map covering the production and processing area and mark each risk point on the electronic map; according to the position of the risk point and the emission coefficient , plan the inspection path by the pre-trained path planning algorithm, and process the emissions at each risk point in turn according to the inspection path.

[0016] Furthermore, if the number of risk points exceeds the expectation, detect the operating status of the production and processing equipment related to the current emission process and obtain the corresponding operating status data; use the trained fault detection model for fault detection with the operating status data as the input and output the fault detection data.

[0017] Furthermore, after extracting the features from the fault detection data, obtain the fault features, pre-construct a fault maintenance knowledge graph with the maintenance of processing equipment faults as the target word; constrain the maintenance interval of the faulty equipment, and maintain the faulty equipment after the maintenance interval that meets the constraint conditions. The constraint method of the maintenance interval is as follows: where

[0018] is the number of carbon emission monitoring cycles, is the difference in abnormal density between the i-th and j-th carbon emission monitoring cycles, is the average value of the differences, and the weight coefficient: , , and , and .

[0019] Furthermore, continuously collect emission data, detect abnormal emission events, and regenerate the abnormal density from the detection data . If the reduction ratio of the abnormal density is lower than expected; Use the trained emission risk assessment model to output the corresponding pollution risk score with the carbon emission data as the input ; Align the pollution risk scores before and after optimization at each emission point to generate the optimization degree . If the obtained optimization degree exceeds the pre-set optimization threshold, send an optimization instruction to the outside.

[0020] Further, after receiving the optimization instruction, collect various control parameters of the production and processing process, take reducing the abnormal emission frequency as the optimization goal, optimize the control parameters of the production and processing process with a pre-trained multi-objective optimization algorithm, and continuously collect carbon emission data and generate a carbon emission reduction report after implementing the optimized production and processing process.

[0021] (III) Beneficial Effects The present invention provides a carbon emission reduction optimization system based on a knowledge graph, which has the following beneficial effects: 1. For the carbon emission data at each time node in real time, by identifying anomalies in the carbon emission process and identifying abnormal emission events therein, abnormal carbon emissions can be processed in a timely manner.

[0022] 2. Construct the corresponding anomaly density from the anomaly status data , and judge the comprehensive anomaly degree of abnormal carbon emissions in the current stage according to the anomaly density . This is equivalent to an extension of anomaly detection. When abnormal carbon emissions persistently exist in the current stage, they can be processed in a timely manner to reduce or alleviate the current carbon emission process.

[0023] 3. Combine the current production and processing process, screen out the carbon emission points therein. When adjusting and optimizing the carbon emissions in the production and processing process, the optimization targeting and optimization adjustment efficiency can be improved; by evaluating each emission point, it can be judged whether the carbon emissions of the current emission point exceed the standard, and real-time monitoring can be achieved when the standard is exceeded.

[0024] 4. Generate an emission coefficient by relying on the risk score of the risk point, the time node when the alarm instruction is received, and the sensitivity coefficient , and screen out the parts that need to be focused on among several emission points according to the emission coefficient . When time and cost are limited, targeted processing can be carried out.

[0025] 5. After screening out the risk points that need to be processed, by planning the corresponding inspection path, each risk point can be processed in turn, improving the overall efficiency of emissions treatment; by performing fault detection or fault prediction on it, the possible operating faults of the production and processing equipment can be discovered or predicted in a timely manner, and when there are operating faults, they can be processed in a timely manner.

[0026] 6. Through the fault maintenance knowledge graph, a corresponding maintenance plan can be given to deal with the equipment faults existing in the production and processing equipment. By maintaining the maintenance frequency of the production and processing equipment, making the maintenance frequency adapt to the production status and carbon emission status, it is possible to avoid the excessive maintenance frequency of the production and processing equipment from affecting the production process.

[0027] 7. Generate an optimization degree after processing the feedback data , and use the optimization degree to evaluate the carbon emission optimization effect in the current stage, and determine whether the current optimization effect has reached the expected effect. Further processing can be carried out when the expected effect is not reached, thereby improving the effectiveness of the optimization.

[0028] 8. Optimize and automatically control the current production and processing process to achieve a response to the optimization method. Based on the evaluation of carbon emissions, through the combination and selection of multiple optimization methods, the reduction of carbon emissions is achieved, and the feedback of the carbon emission reduction process in the current stage is realized by periodically generating a carbon emission reduction report. Description of the Drawings

[0029] Figure 1 is a schematic flow diagram of the carbon emission reduction optimization method based on a knowledge graph according to the present invention; Figure 2 is a schematic structural diagram of the carbon emission reduction optimization system based on a knowledge graph according to the present invention. Detailed Embodiments

[0030] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0031] Please refer to Figure 1 , the present invention provides a carbon emission reduction optimization method based on a knowledge graph, including Step 1. After identifying abnormal emission data in the production and processing process, generate an abnormal density from the abnormal emission data . If the abnormal density exceeds the preset expectation, send an abnormal identification instruction to the outside; The content of Step 1 is as follows: Step 101. Collect carbon emission data in the current production and processing process, such as fossil fuel consumption, electricity use, and waste data, and train a convolutional network with the labeled sample data to obtain a trained abnormal emission identification model; Using the carbon emission data as input, perform abnormal analysis using the trained abnormal emission identification model to identify and obtain the corresponding abnormal emission data, including the type, severity, and time node of the abnormal emission, and summarize and generate an abnormal emission data set; During use, in the process of production and processing, in order to reduce carbon emissions during production, the carbon emission data at each time node is identified for anomalies in the carbon emission process in real time, and abnormal emission events are identified. When there are abnormal carbon emissions, they can be processed in a timely manner; Step 102: Under dimensionless conditions, generate abnormal density from the abnormal emission data in the abnormal emission data set , where

[0032] Weight coefficient: , , ; is the number of abnormal emission events, is the severity value of the abnormal emission event, is the corresponding mean value; is the time interval between the i-th and j-th abnormal emission events, is the mean value of the i-th time interval; According to historical data and management expectations for abnormal emissions, a density threshold is set in advance; if the obtained abnormal density exceeds the preset density threshold, it indicates that there are more current abnormal emission events and a large carbon emission, and timely intervention and processing are required, and an abnormal identification instruction is sent to the outside; During use, combine the content in Steps 101 and 102: As further content, when abnormal carbon emissions continuously occur currently, construct the corresponding abnormal density from the relevant abnormal state data , and judge the comprehensive abnormal degree of the current abnormal carbon emissions based on the abnormal density , which is equivalent to an extension of abnormal detection. When abnormal carbon emissions persist currently, they can be processed in a timely manner to reduce or alleviate the current carbon emission process.

[0033] Combine the above application and the content in the prior art: When producing and processing industrial products, due to the direct or indirect use of various fossil fuels, a large amount of carbon emissions are generated, which in turn causes environmental pollution or climate problems. Therefore, in order to control and optimize excessive carbon emissions, after implementing the corresponding emission reduction strategies, it is necessary to evaluate the carbon emissions generated during production.

[0034] Existing carbon emission reduction effect evaluation methods usually collect carbon emission data before and after emission reduction optimization, and judge whether the emission reduction in the current stage has achieved the expected effect through the carbon emission data. However, the effectiveness of this emission reduction evaluation is relatively single. Especially when there are still many abnormal emission behaviors at present, the single-level emission reduction evaluation is difficult to achieve the expected positive feedback effect, resulting in the inability to sustain the subsequent carbon emission reduction effect and the inability to contribute to continuous carbon emission reduction.

[0035] Step 2: After identifying the emission points of the current process flow, detect and obtain the corresponding emission detection data and then conduct a pollution risk assessment. If the obtained risk score exceeds the preset risk threshold, send an alarm instruction to the outside. The above Step 2 includes the following contents: Step 201: After receiving the abnormal identification instruction, query and obtain the current production and processing process, conduct a process analysis of the current production process, and identify the emission points at each stage of the current process flow; deploy a sensor network at the emission points, such as air quality sensors and carbon emission detection sensors, etc. The sensor network detects and obtains the corresponding emission detection data at the emission points, obtains the carbon emission data of each emission point, including gas emissions, liquid emissions, and solid waste, such as carbon dioxide or waste organic matter, etc., and generates a carbon emission data set after summarization. During use, after decomposing and identifying the current production and processing process, combined with the current production and processing process, screen out the carbon emission points, which can improve the optimization and optimization adjustment efficiency when adjusting and optimizing the carbon emissions in the production and processing process. Step 202: Train a neural convolutional network with the labeled sample data to obtain a trained emission risk assessment model. Using the detected carbon emission data as the input, use the trained emission risk assessment model to conduct a pollution risk assessment, obtain the corresponding pollution risk score. If the risk score exceeds the preset risk threshold, it means that the current carbon emission risk is relatively high and needs to be adjusted. At this time, send an alarm instruction to the outside. During use, combine the contents in Steps 201 and 202: As a further content, after determining the reference standards for carbon emissions and pollution risks of each carbon emission point, by evaluating each emission point, judge whether the carbon emissions of the current emission point exceed the standard, and real-time monitoring can be achieved when it exceeds the standard.

[0036] Step 3: Generate an emission coefficient from the emission risk data of the emission points , based on the emission coefficient Screen out the risk points among several emission points, match the corresponding treatment strategies for the emissions of the risk points, and then process the emissions of each risk point in turn. Step 3 includes the following content: Step 301: After receiving the alarm instruction, determine the sensitivity of the emission point to the production process through sensitivity analysis, and obtain the corresponding sensitivity coefficient; take the time node when the alarm instruction is received as the alarm node. Under dimensionless conditions, generate the emission coefficient according to the risk score of the risk point, the distribution of the alarm node, and the sensitivity coefficient in the following way: as follows:

[0037] In the formula: is the risk score of the i-th risk point at time t, is the carbon emission monitoring period, is the risk threshold, takes the value of 2.718, is the time decay coefficient, with a value between 0 and 1, μ is the distance decay coefficient, with a value between 0.01 and 0.1; is the j-th alarm node at time t, is the time interval between the i-th and j-th alarm nodes, n is the number of alarm nodes, is the risk score on the j-th alarm node; weight coefficient, , , ; According to historical data and the emission management expectations at the emission point, preset the emission threshold; if the obtained emission coefficient exceeds the preset emission threshold, it indicates that the carbon emissions at the current emission point are relatively large and the pollution risk is high. Take the corresponding emission point as a risk point, and screen the emission points that need to be improved according to the risk value; When in use, based on the evaluation of each emission point, generate the emission coefficient according to the risk score of the risk point, the time node when the alarm instruction is received, and the sensitivity coefficient , and can screen out the parts that need to be focused on among several emission points according to the emission coefficient , and can handle them specifically when time and cost are limited.

[0038] Step 302: Identify the carbon emission data and other emission data at the risk point, and obtain the emission categories such as by-products and waste; through online retrieval and offline construction, pre-obtain several emission treatment strategies for reducing the pollution caused by emissions and reducing carbon emissions, such as waste recycling, landfill, or decomposition treatment, etc., and generate an emission treatment strategy library after summarization; according to the correspondence between the emission data and the treatment strategies, match the corresponding emission treatment strategies for the emissions by the emission treatment strategy library; As a further aspect, based on the emissions at the emission points, by matching targeted treatment methods, it is possible to achieve a reduction in carbon emissions.

[0039] Step 303: Construct an electronic map covering the production and processing area, mark each risk point on the electronic map; according to the position and emission coefficient of the risk point , use a pre-trained path planning algorithm to plan the inspection path, and process the emissions at each risk point in sequence according to the inspection path; When in use, combine the content in Steps 301 to 303: After screening the risk points that need to be processed, by planning the corresponding inspection path, each risk point can be processed in sequence, thereby improving the overall efficiency of emissions treatment.

[0040] Step Four: Conduct fault detection on the production and processing equipment. If there is an operating fault, the fault maintenance knowledge graph gives the corresponding maintenance plan, and after a maintenance interval that meets the constraint conditions, maintain the faulty equipment; The said Step Four includes the following content: Step 401: If the number of risk points exceeds the expectation, detect the operating status of the production and processing equipment related to the current emission process, obtain the corresponding operating status data, including equipment energy consumption, operating power, power factor, etc.; Train a machine learning algorithm with the labeled sample data to obtain a trained fault detection model; use the trained fault detection model for fault detection with the operating status data as the input, and output the fault detection data, and summarize to generate a fault detection data set; When in use, based on the collected operating production data of the production and processing equipment, by conducting fault detection or fault prediction on it, it is possible to timely detect or predict possible operating faults of the production and processing equipment. If there is an operating fault, it can be processed in time.

[0041] Step 402: After extracting the features of the fault detection data to obtain the fault features, use the processing equipment fault maintenance as the target word, and after in-depth retrieval and entity relationship construction, pre-construct a fault maintenance knowledge graph; according to the correspondence between the fault maintenance plan and the fault features, the fault maintenance knowledge graph gives the corresponding maintenance plan; And constrain the maintenance interval of the faulty equipment , and after a maintenance interval that meets the constraint conditions, maintain the faulty equipment. The constraint method of the maintenance interval is as follows:

[0042] Among them, is the number of carbon emission monitoring cycles, is the difference in abnormal density between the i-th and j-th carbon emission monitoring cycle diagrams, is the average value of the difference, weight coefficient: , , and ; During use, combine the content in steps 401 and 402: After detecting the faults of the production and processing equipment, the corresponding maintenance plan can be given through the fault maintenance knowledge graph, which can specifically handle the equipment faults existing in the production and processing equipment, specifically reduce the energy consumption data of the processing equipment, reduce the operating load, and overall reduce the carbon emissions in the production and processing process. By maintaining the maintenance frequency of the production and processing equipment, the maintenance frequency can be adapted to the production status and carbon emission status, which can avoid the excessive maintenance frequency of the production and processing equipment from affecting the production process.

[0043] Step Five: After collecting and detecting abnormal emissions, if the reduction ratio of the abnormal density is lower than expected, generate an optimization degree from the carbon emission data at each emission point, optimize the control parameters of the production and processing process, and execute the optimized production and processing process; The said Step Five includes the following content: Step 501: Continuously collect emission data within a preset observation period, detect abnormal emission events, and regenerate the abnormal density from the detection data. If the reduction ratio of the abnormal density is lower than expected, trigger the carbon emission control feedback mechanism and collect carbon emission data; Step 502: Use the carbon emission data as input, and use the trained emission risk assessment model to output the corresponding pollution risk score , align the pollution risk scores before and after optimization at each emission point, and generate the optimization degree in the following manner:

[0044] Among them, is the intermediate value of the optimization degree, is its average value, , n is the number of emission points, and are the pollution risk scores at the i-th emission point before and after optimization respectively, and are the corresponding average values; According to historical data and the expected optimization management of carbon emissions, preset the optimization threshold; If the obtained optimization degree exceeds the pre-set optimization threshold, it indicates that the carbon emissions in the current stage have achieved certain effects. On the contrary, if the obtained optimization degree does not exceed the expectation, it means that the current production and processing technology still needs to be optimized. At this time, an optimization instruction is sent to the outside; During use, after adaptively processing the carbon emissions in the production process, feedback data is collected. After processing the feedback data, an optimization degree is generated , and the optimization degree is used to evaluate the optimization effect of carbon emissions in the current stage, and to judge whether the current optimization effect has achieved the expected effect. Further processing can be carried out when the expected effect is not achieved, so as to improve the effectiveness of optimization.

[0045] Step 503: After receiving the optimization instruction, collect various control parameters of the production and processing technology, take reducing the abnormal emission frequency as the optimization goal, and optimize the control parameters of the production and processing technology with a pre-trained multi-objective optimization algorithm. After executing the optimized production and processing technology, continuously collect carbon emission data and generate a carbon emission reduction report; During use, combine the content in Steps 501 to 503: As a further content, when the existing optimization methods fail to achieve the expected effect, optimize and automatically control the current production and processing technology to achieve the response to the optimization methods. Based on the evaluation of carbon emissions, through the combination and selection of multiple optimization methods, the reduction of carbon emissions is achieved. Further, through the periodic generation of carbon emission reduction reports, the feedback on the carbon emission reduction process in the current stage is realized.

[0046] The construction of the knowledge graph for the fault maintenance of processing and production equipment is a systematic project, which involves multiple links such as information collection, collation, relationship extraction, and graph construction. The following is a detailed construction method: Before constructing the knowledge graph for the fault maintenance of processing and production equipment, it is first necessary to clarify the goals and application scenarios of the graph. For example, whether the graph is to help maintenance personnel quickly locate and solve equipment faults, or to provide equipment fault early warning and preventive maintenance suggestions, etc. Clarifying the goals and application scenarios helps with subsequent information collection and graph design.

[0047] Equipment basic information: including the model, specifications, manufacturer, production date, etc. of the equipment. Fault cases: Collect historical equipment fault cases, including fault phenomena, fault causes, solutions, and maintenance processes, etc.

[0048] Maintenance manuals: Obtain the maintenance manuals and operation guides of the equipment to understand the daily maintenance and repair requirements of the equipment.

[0049] Expert experience: Invite experts in the field of equipment maintenance for interviews to obtain their maintenance experience and suggestions.

[0050] Named entity recognition: Identify key entities from the collected information, such as equipment components, fault types, maintenance tools, etc. Relationship extraction: Determine the relationships between entities, such as the composition relationship between components, the corresponding relationship between faults and components, the corresponding relationship between maintenance methods and fault types, etc.

[0051] Data standardization: Standardize the identified entities and relationships to ensure data consistency and accuracy.

[0052] Select a knowledge graph construction tool: Select a suitable knowledge graph construction tool according to actual needs, such as Neo4j, Apache Jena, etc. Create nodes and relationships: Create entity nodes and relationship edges in the construction tool and enter the collected information into the graph. Graph optimization: Optimize the graph, such as merging duplicate nodes, adjusting the weights of relationships, etc., to improve the accuracy and usability of the graph.

[0053] Fault diagnosis: When the equipment fails, maintenance personnel can quickly locate the faulty components and possible fault causes by querying the knowledge graph and refer to the solution methods in the graph for maintenance.

[0054] Preventive maintenance: Develop a preventive maintenance plan for the equipment based on the equipment maintenance manual and expert experience in the graph to reduce the occurrence of equipment failures.

[0055] Graph update: As the equipment is continuously updated and maintenance experience accumulates, the knowledge graph needs to be updated and improved regularly to ensure the timeliness and accuracy of the graph.

[0056] Data privacy: When collecting and processing equipment fault maintenance information, attention needs to be paid to protecting the enterprise's data privacy and intellectual property rights. Data quality: Ensure that the collected information is accurate, complete, and timely to improve the practicality and reliability of the knowledge graph. Technology selection: Select appropriate knowledge graph construction tools and technical solutions according to the actual needs and resource conditions of the enterprise.

[0057] In summary, the construction of a knowledge graph for the fault maintenance of processing and production equipment requires multiple links such as clarifying the goal, collecting information, sorting out relationships, constructing the graph, and application and maintenance. By constructing the knowledge graph, the enterprise can quickly locate and solve equipment faults, improving equipment reliability and production efficiency.

[0058] Please refer to Figure 2 , the present invention provides a carbon emission reduction optimization system based on a knowledge graph, including, Anomaly analysis unit, after identifying abnormal emission data in the production and processing process, generating anomaly density from the abnormal emission data If the abnormal density exceeds the preset expectation, an abnormal recognition instruction is sent to the outside; A risk assessment unit, after identifying the emission points of the current process flow, detects and obtains the corresponding emission detection data and then conducts a pollution risk assessment. If the obtained risk score exceeds the preset risk threshold, an alarm instruction is sent to the outside; An emission point screening unit generates an emission coefficient from the emission risk data of the emission points and, based on the emission coefficient screens out the risk points among several emission points, matches corresponding treatment strategies for the emissions of the risk points, and then processes the emissions of each risk point in turn; A fault detection unit conducts fault detection on the production and processing equipment. If there is an operating fault, a corresponding maintenance plan is given by the fault maintenance knowledge graph, and after a maintenance interval that meets the constraint conditions, the faulty equipment is maintained; An optimization unit, after collecting and detecting abnormal emissions, if the reduction ratio of the abnormal density is lower than expected, generates an optimization degree from the carbon emission data at each emission point and optimizes the control parameters of the production and processing process, and executes the optimized production and processing process.

[0059] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraint conditions of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0060] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0061] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only for some logical function divisions, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0062] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0063] As mentioned above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

Claims

1. Carbon emission reduction optimization system based on knowledge graph, characterized in that: including, Anomaly analysis unit, after identifying abnormal emission data during the production and processing process, generates anomaly density from the abnormal emission data , if the anomaly density exceeds the preset expectation, sends an anomaly identification instruction to the outside; among them, the anomaly density is generated from the abnormal emission data in the abnormal emission data set : ; Weight coefficient: , , is the number of abnormal emission events, is the severity value of the abnormal emission event, is the corresponding mean value; is the i th and the time interval between the jth abnormal emission events, is the i th mean value of the time interval; a risk assessment unit, which, after identifying the emission points of the current process flow, detects and obtains the corresponding emission detection data and then conducts a pollution risk assessment. If the obtained risk score exceeds the preset risk threshold, it sends an alarm instruction to the outside; Emission point screening unit, generating an emission coefficient from the emission risk data of the emission points , based on the emission coefficient screening out risk points among a number of emission points, matching corresponding treatment strategies for the emissions of the risk points, and then treating the emissions of each risk point in turn; a fault detection unit, which conducts fault detection on the production and processing equipment. If there is an operating fault, the corresponding maintenance plan is given by the fault maintenance knowledge graph, and after a maintenance interval that meets the constraint conditions, the faulty equipment is maintained; Optimization unit. After collecting and detecting abnormal emissions, if the reduction ratio of the abnormal density is lower than expected, an optimization degree is generated from the carbon emission data at each emission point , the control parameters of the production and processing process are optimized, and the optimized production and processing process is executed.

2. The carbon emission reduction optimization system based on a knowledge graph according to claim 1, characterized in that: Collect carbon emission data during the production and processing process, use the trained abnormal emission identification model to conduct abnormal analysis with the carbon emission data as the input, identify and obtain the corresponding abnormal emission data, including the type, severity and time node of the abnormal emission, and summarize and generate an abnormal emission data set.

3. The carbon emission reduction optimization system based on a knowledge graph according to claim 1, characterized in that: After receiving the abnormal identification instruction, query and obtain the current production and processing process and then conduct process analysis to identify the emission points at each stage; the sensor network detects and obtains the corresponding emission detection data at the emission points, obtains the carbon emission data of each emission point, uses the trained emission risk assessment model to conduct pollution risk assessment with the detected carbon emission data as the input, and obtains the corresponding pollution risk score.

4. The carbon emission reduction optimization system based on a knowledge graph according to claim 3, characterized in that: After receiving the alarm instruction, determine the sensitivity of the emission point to the production process through sensitivity analysis and obtain the corresponding sensitivity coefficient; use the time node when the alarm instruction is received as the alarm node, and based on the risk score of the risk point, the distribution of the alarm node, and the sensitivity coefficient Generate an emission coefficient , if the obtained emission coefficient exceeds the preset emission threshold, regard the corresponding emission point as a risk point.

5. The carbon emission reduction optimization system based on a knowledge graph according to claim 4, characterized in that: According to the risk scores of risk points, the distribution of alarm nodes and the sensitivity coefficients Generate emission factors The method is as follows: ; Wherein: is the risk score of the i-th risk point at time t, is the carbon emission monitoring period, is the risk threshold, takes the value of 2.718, is the time decay coefficient, and μ is the distance decay coefficient; is the j-th alarm node at time t, is the time interval between the i-th and j-th alarm nodes, and n is the number of alarm nodes, is the risk score and weight coefficient on the j-th alarm node, ， , .

6. The carbon emission reduction optimization system based on a knowledge graph according to claim 5, characterized in that: Identify the carbon emission data and other emission data at the risk points, obtain the types of emissions such as by-products and waste, pre-obtain several emission treatment strategies, and the emission treatment strategy library matches the corresponding emission treatment strategy for the emissions; Build an electronic map covering the production and processing areas, and mark each risk point on the electronic map; according to the location of the risk point and the emission coefficient , use the pre-trained path planning algorithm to plan the inspection path, and process the emissions of each risk point in turn according to the inspection path.

7. The carbon emission reduction optimization system based on a knowledge graph according to claim 6, characterized in that: If the number of risk points exceeds the expectation, conduct an operating status detection on the production and processing equipment related to the current emission process, and obtain the corresponding operating status data; Use the trained fault detection model to conduct fault detection with the operating status data as the input, and output the fault detection data.

8. The carbon emission reduction optimization system based on a knowledge graph according to claim 7, characterized in that: After feature extraction of the fault detection data, obtain the fault features, and pre-construct a fault maintenance knowledge graph with the processing equipment fault maintenance as the target word; Maintenance interval for faulty equipment is restricted. After the maintenance interval that meets the restricted conditions, the faulty equipment is maintained. The restricted way of the maintenance interval is as follows: ; among them, is the number of carbon emission monitoring cycles, is the i th difference in abnormal density between the j th and the th carbon emission monitoring cycle diagrams, ， is the average value of the differences, and the weight coefficient: .

9. The carbon emission reduction optimization system based on a knowledge graph according to claim 8, characterized in that: Continuously collect emission data, detect abnormal emission events, and regenerate abnormal density from the detected data , if the reduction ratio of the abnormal density is lower than expected; Using carbon emission data as input, the trained emission risk assessment model outputs corresponding pollution risk scores ; Align the pollution risk scores before and after optimization for each emission point to generate an optimization degree . If the obtained optimization degree exceeds the pre-set optimization threshold, an optimization instruction is sent to the outside 10. The carbon emission reduction optimization system based on a knowledge graph according to claim 9, characterized in that: After receiving the optimization instruction, collect the control parameters of the production and processing process, take reducing the abnormal emission frequency as the optimization goal, use the pre-trained multi-objective optimization algorithm to optimize the control parameters of the production and processing process, and after executing the optimized production and processing process, continuously collect the carbon emission data and generate a carbon emission reduction report.

Images