Fuel gas monitoring method based on edge cloud collaboration

Abstract

The invention belongs to the technical field of fuel gas monitoring, and discloses a fuel gas monitoring method based on edge cloud collaboration. The method comprises the following steps: S1, monitoring a gas state in real time based on equipment scene data; s2, based on the real-time pipe network pressure, the gas hidden danger degree is determined; s3, predicting a gas hidden danger state in a next time period; s4, monitoring the actual gas state, and feeding back and adjusting the prediction mode; s5, generating a gas safety report according to the prediction result; s6, outputting a gas safety report; s7, the gas control mode is adjusted according to the gas safety report; in general, the method has the remarkable advantages of being high in fault prediction accuracy, high in auxiliary decision-making capability and good in information transmission effect.

Description

Technical Field

[0001] The present invention relates to the technical field of gas monitoring, and more specifically, to a gas monitoring method based on edge-cloud collaboration. Background Art

[0002] Thanks to the rapid economic development in China in recent years, as a mainstream energy supply method in the world today, gas is widely used in thousands of households in China. However, due to the vast territory of China, in order to supply people as much as possible, the laying length and scope of gas pipelines are extremely wide. And because gas itself has the characteristics of being flammable, explosive and causing asphyxiation and poisoning, the safety related to gas transportation has become a major problem that the current gas transportation industry needs to face for a long time.

[0003] The patent with the application publication number CN112947290B discloses a device status monitoring method, monitoring system and storage medium based on edge-cloud collaboration. Through the device status monitoring method based on edge-cloud collaboration and the device performance monitoring system based on edge-cloud collaboration, it realizes the application that key information in the collected information in the device monitoring requirements is not missed and does not occupy too many cloud resources, has a high effective information density, and thus can quickly locate and view anomalies in a large amount of data. It can not only provide an overview of the overall health of the device, but also delve into details to view the original on-site waveform at the anomaly site, and can further analyze based on this to comprehensively understand and study the current working condition of the monitored device or predict future working conditions, realizing the health status and working condition assessment of the entire device. Based on the above architecture, edge-cloud collaboration is realized for comprehensive status monitoring. The edge side uses the threshold optimized by the cloud for new anomaly judgment and raw data capture, improving the accuracy of device status assessment and predicting device failures.

[0004] However, for the above-mentioned device status monitoring method, monitoring system and storage medium based on edge-cloud collaboration, although to a certain extent, through the monitoring system, it realizes high data capture, utilization and evaluation, and further realizes the prediction of device failures. However, in the use of gas equipment, gas equipment is often extremely vulnerable to various factors such as environmental factors, human factors and gas pressure factors, which may lead to damage to gas equipment and further expand into major accidents of gas leakage. Moreover, during the gas monitoring process, most of the time, it is the staff who monitor the changes of basic data in real time. Not only is it easy to cause poor monitoring efficiency due to the boring monitoring process, but also when abnormal data is detected, the staff needs to make judgments through a lot of basic data and then make decisions. And this time cost will further expand the losses of the enterprise along with gas leakage and equipment damage. Therefore, how to ensure the fault prediction, auxiliary decision-making and information transmission efficiency during gas monitoring is particularly important.

[0005] In view of this, the present invention proposes a gas monitoring method based on edge-cloud collaboration to solve the above problems. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solutions. The method includes: S1: Based on device scenario data, monitor the gas state in real time. This step specifically includes the edge node accessing the device scenario data in real time, enabling the system to calculate the device scenario data, quantify the primary data related to the device, provide accurate data for subsequent system calculations, overall improve the accuracy of system prediction calculations, and through the quantification of multiple data, the system can further deepen the calculation accuracy. The device scenario data includes pipeline material data, pipeline burial depth data, pipeline service life data, soil humidity data, and soil corrosion degree data, and the primary data includes corrosion coefficient and aging coefficient; S2: Based on the real-time pipeline network pressure, determine the degree of gas hazards. This step specifically includes the pressure sensor transmitting the gas pressure data to the edge node in real time to ensure the timeliness of the system obtaining the gas pressure data, calculating the pressure anomaly coefficient based on the gas pressure data, and combining the historical gas pressure data to obtain the pressure anomaly level, so as to ensure that the system can immediately identify abnormal fluctuations, judge the immediate risk level, and provide immediate label data for the system; S3: Predict the gas hazard state in the next time period. This step specifically includes training an LSTM time series model based on the cloud, inputting historical pressure data, corrosion coefficient, and aging coefficient, and obtaining the hazard prediction value in the next time period, enabling the system to have the linkage calculation ability from the edge node to the cloud processing, providing the core data basis for the subsequent system to generate a gas safety report. At the same time, the existence of the hazard prediction value also enables the system to have the ability to guide resource pre-allocation, achieving the purpose of the system predicting future hazards, actively preventing and controlling, and avoiding passive response; S4: Monitor the actual gas state and feedback to adjust the prediction method. This step specifically includes calculating the prediction error based on the gas pressure data and the hazard prediction value collected by the edge node in real time, enabling the system to have the ability to quantify the model deviation. At the same time, through the numerical comparison mechanism, trigger the feedback mechanism of cloud incremental learning, greatly improving the model dynamic optimization ability of the system, enabling the system to ensure that the LSTM time series model still guarantees high-precision data prediction with environmental changes; S5: Generate a gas safety report according to the prediction results. Specifically, this step includes summarizing the hidden danger prediction values and pressure anomaly levels based on the cloud, matching the hidden danger prediction values with the hidden danger level template, and automatically generating a structured report, enabling the system to convert data and analysis results into actionable and selectable decision information reports. Through standardized output, not only is the automatic parsing burden of the system effectively reduced, but also the intuitiveness of data presentation is enhanced, facilitating manual reading and understanding, ensuring that the system has the ability to establish links between data analysis, data output, and equipment control, and overall enhancing the timeliness and convenience of instruction transmission. The structured report includes prediction results, risk locations, and recommended measures. S6: Output the gas safety report. Specifically, this step includes identifying the gas safety report and outputting it based on multi-modal, and efficiently transmitting the gas safety report to the staff terminal and the gas equipment control unit, enabling the system to accurately push the gas safety report content to different receiving terminals. This not only effectively enhances the pertinence of data transmission, but also greatly reduces the impact of inefficient reports caused by high-frequency pushing on personnel decision-making, ensuring the classification ability and data validity of data transmission. S7: Adjust the gas control method according to the gas safety report. Specifically, this step includes identifying the gas safety report based on the cloud and issuing instructions to the programmable logic controller at the edge node, enabling the system to convert the gas safety report into regulatory measures in the physical world, and greatly reducing the safety accidents caused by emergency safety hazards. Furthermore, step S1 includes: S1.1: Real-time access the device scenario data through the edge node, and calculate the corrosion coefficient related to the gas equipment according to the scenario data. The specific calculation formula for the corrosion coefficient is: ; Obtain the corrosion coefficient , where is the data of the th type of pipeline material, is the pipeline burial depth data, is the pipeline service life data, is the soil humidity data, is the soil corrosion degree data, and are the corresponding weight factors respectively, and satisfy ; S1.2: Calculate the aging coefficient related to the gas equipment according to the scenario data. The specific calculation formula for the aging coefficient is: ; Obtain the aging coefficient ; Further, step S2 includes: S2.1, transmitting the gas pressure data to the edge node in real time through a pressure sensor, and calculating the pressure anomaly coefficient by combining the historical gas pressure data; The specific calculation formula for the pressure anomaly coefficient is: ; Obtain the pressure anomaly coefficient , where is the gas pressure data at the time period, is the historical mean value, is the standard deviation; S2.2, judging the pressure anomaly level according to the pressure anomaly coefficient; The specific calculation formula group for the pressure anomaly level is: ; Obtain the pressure anomaly level ; Further, step S3 includes: Training the LSTM time series model through the cloud, and inputting the historical pressure data, corrosion coefficient and aging coefficient to calculate the hidden danger prediction value in the next time period; The specific calculation formula for the hidden danger prediction value is: ; where is the pre-trained LSTM time series model, , and are the corrosion coefficient, aging coefficient and historical pressure data input into the LSTM time series model respectively; Further, step S4 includes: S4.1, by obtaining the gas pressure data and the hidden danger prediction value , calculating the prediction deviation data , and verifying the accuracy of the hidden danger prediction value ; The specific calculation formula for the prediction deviation data is: ; S4.2, according to the result of comparing the prediction deviation data with the standard deviation , select to continue outputting or feedback adjusting the prediction method. When the prediction deviation data is greater than or equal to the standard deviation , trigger the re-training of the LSTM time series model. When the prediction deviation data is less than or equal to the standard deviation , transmit the hidden danger prediction value to step S5; Further, step S5 includes: Summarize the hidden danger prediction value and the pressure anomaly level through the cloud, and match the hidden danger prediction value with the hidden danger level template to obtain the hidden danger status data , match the hidden danger status data and the pressure anomaly level with the recommended measure template respectively to generate a gas safety report; The specific calculation formula group for matching the hidden danger prediction value with the hidden danger level template is: ; Further, step S6 includes: By identifying the gas safety report, select the sending frequency and sending terminal according to the data transmission unit; Identifying the gas safety report includes identifying the measure reports of low risk, medium risk and high risk in the gas safety report; When the gas safety report is a low-risk measure report, send the gas safety report to the staff email receiver; When the gas safety report is a medium-risk measure report, send the gas safety report to the staff email receiver and send a text message reminder every half hour through the mobile phone receiver; When the gas safety report is a high-risk measure report, send the gas safety report to the staff email receiver and make an uninterrupted phone reminder through the mobile phone receiver; Further, step S7 includes: By identifying the gas safety report, transmit the control instruction issued by the cloud to the programmable logic controller of the edge node; The specific calculation formula of the control instruction is: ; Obtain the control instruction , where is the adjustment coefficient; Further, the gas safety report in step S5 includes a hidden danger recommended measure report and a pressure recommended measure report, where: The hidden danger recommended measure report includes a hidden danger low-risk measure report, a hidden danger medium-risk measure report and a hidden danger high-risk measure report; The hidden danger low-risk measure report includes instructions that the gas equipment is operating smoothly, asking the staff to maintain the current valve opening and conduct pipeline inspections according to the regular plan; The hidden danger medium-risk measure report includes instructions that the gas equipment may experience pressure fluctuations, the risk location coordinates, asking the staff to prepare emergency tools, and a group of signals representing the adjustment instruction sent to the edge node; The hidden danger high-risk measure report includes instructions on the possible leakage risk of gas equipment, the risk location coordinates, asking the staff to arrive at the risk location for emergency repair as soon as possible, and a group of representatives sending a signal to the edge node to issue an adjustment instruction; The pressure recommended measure report includes the pressure low-risk measure report, the pressure medium-risk measure report, and the pressure high-risk measure report; The pressure low-risk measure report includes instructions on the stable operation of gas equipment, asking the staff to conduct pipeline inspections according to conventional calculations; The pressure medium-risk measure report includes instructions on the short-term overpressure risk of gas equipment, the risk location coordinates, asking the staff to go to the risk location to check the valve seal line, and a group of representatives sending a signal to the edge node to issue an adjustment instruction; The pressure high-risk measure report includes instructions on the suspected leakage of gas equipment, the risk location coordinates, asking the staff to arrive at the risk location for emergency repair as soon as possible, and a group of representatives sending a signal to the edge node to issue an adjustment instruction; Furthermore, the determination method of the adjustment coefficient in step S7 includes: The specific calculation formula of the adjustment coefficient is: ; Obtain the adjustment coefficient , where is the safety specification data, is the flow pressure coefficient.

[0007] The technical effects and advantages of the gas monitoring method based on edge-cloud collaboration of the present invention: The present invention monitors the gas state in real time based on device scenario data, determines the degree of gas hazards based on real-time pipeline network pressure, predicts the gas hazard state in the next time period, monitors the actual gas state, and feeds back to adjust the prediction method. According to the prediction results, a gas safety report is generated, the gas safety report is output, and the gas control method is adjusted based on the gas safety report, so as to ensure the accuracy of the calculation and prediction results of the system under the influence of the dynamic fluctuation data of gas pressure data. In addition, the present invention also optimizes the LSTM time series model to ensure that the model can follow the update after the device scenario data changes, avoiding data calculation rigidity caused by the change of device scenario data, so as to achieve the purpose of improving the system prediction accuracy. Moreover, the gas safety report generated by the prediction results can effectively assist the staff in quickly completing the decision-making of fault response measures, greatly reducing the time cost of fault judgment and implementation decision-making of the fault staff. Combined with the programmable logic controller with edge-cloud collaboration, the gas safety report can be identified to assist the staff in stopping the operation of gas equipment first, thereby effectively reducing the losses of the enterprise. At the same time, through the multi-modal output of the gas safety report, the disadvantages of the staff in facing the information transmission with the same frequency and the same receiving end can be effectively solved, greatly improving the effectiveness and humanization of the system information transmission. Generally speaking, the present invention has the remarkable advantages of high fault prediction accuracy, strong auxiliary decision-making ability and good information transmission effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the gas monitoring method based on edge-cloud collaboration of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0009] 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.

[0010] Embodiment 1: Please refer to Figure 1 As shown, the gas monitoring method based on edge-cloud collaboration in this embodiment includes: S1: Based on the device scenario data, monitor the gas state in real time; specifically, this step includes the edge nodes accessing the device scenario data in real time, enabling the system to calculate the device scenario data, quantify the primary data related to the device, provide accurate data for subsequent system calculations, overall improve the accuracy of system prediction calculations, and through the quantification of multiple data, the system can further deepen the calculation accuracy. The device scenario data includes pipeline material data, pipeline burial depth data, pipeline service life data, soil humidity data, and soil corrosion degree data, and the primary data includes corrosion coefficient and aging coefficient; S2: Based on the real-time pipeline network pressure, determine the degree of gas hidden danger; specifically, this step includes the pressure sensor transmitting the gas pressure data to the edge nodes in real time to ensure the timeliness of the system obtaining the gas pressure data, calculating the pressure anomaly coefficient based on the gas pressure data, and combining the historical gas pressure data to obtain the pressure anomaly level, so as to ensure that the system can immediately identify abnormal fluctuations, judge the immediate risk level, and provide immediate label data for the system; S3: Predict the gas hidden danger state in the next time period; specifically, this step includes training the LSTM time series model based on the cloud, inputting the historical pressure data, corrosion coefficient, and aging coefficient, and obtaining the hidden danger prediction value in the next time period, enabling the system to have the linkage calculation ability from the edge node to the cloud processing, providing the core data basis for the subsequent system to generate a gas safety report. At the same time, the existence of the hidden danger prediction value also enables the system to have the ability to guide resource pre-allocation, achieving the purpose of the system predicting future hidden dangers, taking proactive prevention and control, and avoiding passive response; S4: Monitor the actual gas state and feedback to adjust the prediction method; specifically, this step includes calculating the prediction error based on the gas pressure data and hidden danger prediction value collected by the edge nodes in real time, enabling the system to have the ability to quantify the model deviation. At the same time, through the numerical comparison mechanism, trigger the feedback mechanism of cloud incremental learning, greatly improve the model dynamic optimization ability of the system, and enable the system to ensure that the LSTM time series model still guarantees high-precision data prediction with environmental changes; S5: Generate a gas safety report according to the prediction result; specifically, this step includes summarizing the hidden danger prediction value and pressure anomaly level based on the cloud, matching the hidden danger prediction value with the hidden danger level template, and automatically generating a structured report, enabling the system to convert data and analysis results into actionable and selectable decision information reports. Through standardized output, it not only effectively reduces the automatic parsing burden of the system, but also enhances the intuitiveness of data presentation, facilitates manual reading and understanding, ensures that the system has the ability to establish the link between data analysis, data output, and device control, and overall enhances the timeliness and convenience of instruction transmission. The structured report includes prediction results, risk locations, and recommended measures; S6: Output a gas safety report; specifically, this step includes identifying the gas safety report and outputting it based on multimodality, and efficiently transmitting the gas safety report to the staff terminal and the gas equipment control unit, enabling the system to accurately push the gas safety report content to different receiving terminals according to the content of the gas safety report. This not only effectively enhances the pertinence of data transmission, but also greatly reduces the impact of inefficient reports caused by high-frequency pushing on personnel decision-making, ensuring the classification ability and data validity of data transmission; S7: Adjust the gas control method according to the gas safety report; specifically, this step includes identifying the gas safety report based on the cloud and issuing instructions to the programmable logic controller at the edge node, enabling the system to convert the gas safety report into regulatory measures in the physical world, and greatly reducing the safety accidents caused by emergency safety hazards; The core of the present invention lies in quantifying data based on device scenario data, and performing real-time dynamic prediction on the quantified primary data and real-time gas pressure data according to the LSTM time series model trained on the cloud, achieving the purpose of the system predicting the gas hazard state in the next time period. By monitoring the gas state with the device scenario data in step S1 and calculating the gas hazard degree with the real-time gas pressure data in step S2, the system can achieve comprehensive real-time safety monitoring of gas equipment. The hazard prediction value provided by step S3 enables step S5 to generate a gas safety report for assisting staff in decision-making, thereby enhancing the decision-making assistance ability of the system. Step S4 provides a model dynamic optimization function for step S3, enabling the system to achieve high-accuracy prediction ability under data dynamic fluctuations. By transmitting the gas safety report in step S6, it is ensured that the gas safety report can be targeted to different terminals, guaranteeing the distinctiveness and convenience of data transmission. By identifying the content of the gas safety report in step S7, the system has the ability to convert data into control measures in the physical world, greatly increasing the timeliness and effectiveness of the system's handling in the face of emergencies; The prediction of the gas hazard state through the primary data and real-time gas pressure data enables the system to predict the future safety hazards of gas equipment under the dynamic changes of real-time gas pressure data, and effectively enhance the adaptability and response of the system to dynamic data according to the feedback mechanism, thereby effectively enhancing the accuracy of prediction data calculation. By classifying and matching the hazard prediction values, it is ensured that the system can unify the calculated data with the required decision-making assistance, greatly enhancing the decision-making assistance ability of the system. Through multimodal output, cloud identification, and control instructions based on the identification results, the system has the ability to analyze data and conduct targeted distribution and communication, and can be linked with the physical world, effectively reducing the rescue time between report transmission and implementation measures. This method can effectively predict the hazard degree of gas equipment, assist staff in making decisions, and optimize information communication, achieving the purpose of ensuring the safety of gas equipment; Step S1 includes: S1.1. Real - time access device scenario data through edge nodes, and calculate the corrosion coefficient related to gas equipment according to the scenario data; The specific calculation formula of the corrosion coefficient is: ; Obtain the corrosion coefficient , where is the data of the th pipeline material, is the pipeline burial depth data, is the pipeline service duration data, is the soil humidity data, is the soil corrosion degree data, and are the corresponding weight factors respectively, and satisfy ; S1.2. Calculate the aging coefficient related to gas equipment according to the scenario data; The specific calculation formula of the aging coefficient is: ; Obtain the aging coefficient ; The core of this implementation is that by calculating the characteristic coefficients of the device scenario data, the corrosion and aging degrees related to gas equipment can be effectively calculated, so as to achieve the purpose of real - time monitoring of the gas state. This method can comprehensively consider the state of the pipeline buried underground, obtain a value that can more accurately reflect the state of the gas equipment. By calculating the device scenario data to monitor the gas equipment state, the state of the gas equipment can be presented more intuitively, thereby improving the economy and convenience of gas equipment state monitoring. Through the obtained corrosion coefficient, the corrosion degree of the pipeline can be quantified, and then the purpose of directly evaluating the pipeline strength can be achieved. Through the obtained aging coefficient, the cumulative effect of pipeline degradation over time can be intuitively reflected, and combined with the corrosion coefficient, the risk of sudden leakage can be prevented, so that the system can not only provide accurate data for the next prediction calculation through the corrosion coefficient and aging coefficient, but also the corrosion coefficient and aging coefficient themselves can effectively reflect the current state of the gas equipment; Step S2 includes: S2.1. Real - time transmit gas pressure data to the edge node through a pressure sensor, and calculate the pressure anomaly coefficient in combination with historical gas pressure data; The specific calculation formula of the pressure anomaly coefficient is: ; Obtain the pressure anomaly coefficient , where is the gas pressure data at the th time period, is the historical mean, is the standard deviation; S2.2. Determine the pressure anomaly level based on the pressure anomaly coefficient; The specific calculation formula group for the pressure anomaly level is: ; Obtain the pressure anomaly level ; The core of this implementation lies in quantifying the degree to which the gas pressure deviates from the normal gas pressure range by performing Z-Score standardization on the gas pressure data to obtain the pressure anomaly coefficient, and then adopting a grading strategy for the pressure anomaly coefficient to achieve the purpose of simplifying the risk response logic. Specifically, this method preprocesses the gas pressure data to obtain the anomaly coefficient of the gas pressure, and then classifies the pressure anomaly through a grading strategy, enabling the gas pressure state within the current gas equipment to be obtained, thereby reflecting the degree of gas hidden danger, and thus achieving the purpose of quantifying and transforming the gas hidden danger degree through the gas pressure data by the system. This calculation and grading method enables the system to monitor the gas pressure state within the gas equipment in real time and classify it to obtain the pressure anomaly level, achieving the purpose of real-time feedback of the gas hidden danger degree under the dynamic fluctuation of the gas pressure data. This implementation realizes the feedback of the hidden danger degree of the system under the dynamic change of the gas pressure data through the calculation and grading of the gas pressure data, enabling the system to effectively reduce the time and labor costs required for the staff to face basic calculations; Step S3 includes: Train the LSTM time series model through the cloud, input the historical pressure data, corrosion coefficient, and aging coefficient, and calculate the hidden danger prediction value for the next time period; The specific calculation formula for the hidden danger prediction value is: ; Among them, is the pre-trained LSTM time series model, , and are the corrosion coefficient, aging coefficient, and historical pressure data input into the LSTM time series model respectively; It should be noted that the historical pressure data is the gas pressure data sequence within the past 24 hours; The core of this embodiment lies in calculating the hidden danger prediction value by using the LSTM time series model for the input historical pressure data, corrosion coefficient, and aging coefficient, so that the system can predict future hidden dangers, achieve active prevention and control, and avoid passive response. Specifically, this embodiment uses the LSTM time series model to fuse the historical pressure data, corrosion coefficient, and aging coefficient to capture the complex correlation of multi-source data. For example, the acceleration of the corrosion coefficient may lead to abnormal gas pressure, and then output a hidden danger prediction value reflecting the pressure deviation or leakage probability in the next time period. This embodiment makes the system calculate the hidden danger prediction value reflecting the future gas hidden danger state more accurately, effectively improving the reliability of system prediction; Step S4 includes: S4.1, calculate the predicted deviation data by obtaining the gas pressure data and the hidden danger prediction value , and verify the accuracy of the hidden danger prediction value ; The specific calculation formula of the predicted deviation data is: ; S4.2, according to the result of comparing the predicted deviation data with the standard deviation , select to continue outputting or feedback-adjusting the prediction method. When the predicted deviation data is greater than or equal to the standard deviation , trigger the retraining of the LSTM time series model. When the predicted deviation data is less than or equal to the standard deviation , transmit the hidden danger prediction value to step S5; The core of this embodiment lies in verifying the hidden danger prediction value through the standard deviation , so as to realize the dynamic optimization ability of the system model. Specifically, this embodiment quantifies the deviation between the hidden danger prediction value output by the model and the actual data, and then selects to continue transmitting the hidden danger prediction value or return to step S3 for model retraining according to the situation of the deviation data, so that the system can avoid the data calculation rigidity caused by the change of equipment scenario data. This embodiment enables the system to have the ability of continuous optimization, ensuring that the LSTM time series model in step S3 can maintain high prediction accuracy following the update and change of equipment scenario data, effectively improving the accuracy of system prediction; Step S5 includes: Summarize the hidden danger prediction value and the pressure anomaly level through the cloud, match the hidden danger prediction value with the hidden danger level template to obtain the hidden danger state data , and match the hidden danger state data and the pressure anomaly level with the recommended measure template respectively to generate a gas safety report; ​The specific calculation formula group for matching the hidden danger prediction value with the hidden danger level template is as follows: ; The core of this implementation method lies in quantifying the hidden danger prediction value into hidden danger status data, and combining the pressure anomaly level to match the recommended measure template, and outputting a gas safety report. Specifically, this implementation method uniformly matches the prediction result and the pressure anomaly level obtained in step S2 to the recommended measure template, enabling the system to convert data into actionable decision support information, thereby avoiding the timeliness of control instructions caused by the absence or delay of decision-making information, and effectively improving the timeliness and safety of the system's control of gas equipment; Step S6 includes: By identifying the gas safety report, select the sending frequency and sending terminal according to the data transmission unit; Identifying the gas safety report includes identifying the measure reports of low risk, medium risk and high risk in the gas safety report; When the gas safety report is a low-risk measure report, send the gas safety report to the staff email receiving end; When the gas safety report is a medium-risk measure report, send the gas safety report to the staff email receiving end and send a text message reminder every half hour through the mobile phone receiving end; When the gas safety report is a high-risk measure report, send the gas safety report to the staff email receiving end and make an uninterrupted phone reminder through the mobile phone receiving end; The core of this implementation method lies in efficiently transmitting the gas safety report to the staff receiving end. Specifically, this implementation method selects different receiving ends and reminder frequencies for transmission according to the different risk levels in the gas safety report, thereby effectively reducing the high-risk response delay caused by low-frequency push and the interference to staff caused by high-frequency push due to the same frequency and the same receiving end transmission, and effectively improving the effectiveness and humanization of the system information transmission; Step S7 includes: By identifying the gas safety report, transmit the control instruction issued by the cloud to the programmable logic controller of the edge node; The specific calculation formula of the control instruction is: ; Obtain the control instruction , where is the adjustment coefficient; The core of this implementation method lies in converting the gas safety report into control measures to complete the closed loop of the system. This implementation method adjusts the gas pressure through the adjustment coefficient to avoid secondary risks caused by sudden changes in gas pressure, and effectively improves the safety and reliability of the system control terminal; In step S5, the gas safety report includes a hidden danger suggestion measure report and a pressure suggestion measure report, where: The hidden danger suggestion measure report includes a hidden danger low-risk measure report, a hidden danger medium-risk measure report, and a hidden danger high-risk measure report; The hidden danger low-risk measure report includes stating that the gas equipment is operating smoothly, asking the staff to maintain the current valve opening and conduct pipeline inspections according to the regular plan; The hidden danger medium-risk measure report includes stating that the gas equipment may have pressure fluctuations, the risk location coordinates, asking the staff to prepare emergency tools, and a group of representatives sending a signal for adjusting instructions to the edge node; The hidden danger high-risk measure report includes stating that the gas equipment may have a leakage risk, the risk location coordinates, asking the staff to rush to the risk location for repair as soon as possible, and a group of representatives sending a signal for adjusting instructions to the edge node; The pressure suggestion measure report includes a pressure low-risk measure report, a pressure medium-risk measure report, and a pressure high-risk measure report; The pressure low-risk measure report includes stating that the gas equipment is operating smoothly, asking the staff to conduct pipeline inspections according to the regular calculation; The pressure medium-risk measure report includes stating that the gas equipment has a short-term overpressure risk, the risk location coordinates, asking the staff to go to the risk location to check the valve sealing line, and a group of representatives sending a signal for adjusting instructions to the edge node; The pressure high-risk measure report includes stating that the gas equipment has a suspected leakage, the risk location coordinates, asking the staff to rush to the risk location for repair as soon as possible, and a group of representatives sending a signal for adjusting instructions to the edge node; In step S7, the determination method of the adjustment coefficient includes: The specific calculation formula of the adjustment coefficient is: ; Obtain the adjustment coefficient , where, is the safety specification data, is the flow-pressure coefficient; It should be explained that the safety specification data refers to, for example, reducing the gas pressure by 10% in case of high risk; the flow-pressure coefficient is calibrated through the flow-pressure characteristic curve provided by the valve manufacturer; In this embodiment, the beneficial effects are as follows: real-time monitoring of the gas state based on device scenario data; determination of the degree of gas hazards based on real-time pipeline network pressure; prediction of the gas hazard state in the next time period; monitoring of the actual gas state and feedback adjustment of the prediction method; generation of a gas safety report according to the prediction result; output of the gas safety report; adjustment of the gas control method based on the gas safety report, so as to ensure the accuracy of the calculation and prediction results of the system under the influence of dynamic fluctuating data such as gas pressure data. In addition, the present invention optimizes the LSTM time series model, enabling the model to follow and update after changes in device scenario data, avoiding data calculation rigidity caused by changes in device scenario data, thereby achieving the purpose of improving the prediction accuracy of the system. Moreover, the gas safety report generated based on the prediction result can effectively assist the staff in quickly making decisions on fault response measures, greatly reducing the time cost of fault judgment and implementation decision-making for the fault staff. Combined with the programmable logic controller with edge-cloud collaboration, the gas safety report can be identified, thereby assisting the staff to stop the operation of gas equipment first, and further achieving the purpose of effectively reducing the losses of the enterprise. At the same time, through the multi-modal output of the gas safety report, the disadvantages of the staff facing information transmission with the same frequency and receiving end can be effectively solved, greatly improving the effectiveness and humanization of system information transmission. Generally speaking, the present invention has the remarkable advantages of high fault prediction accuracy, strong auxiliary decision-making ability, and good information transmission effect.

[0011] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0012] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs attached to the claims should not be regarded as limiting the claimed rights.

[0013] In addition, it is obvious that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices stated in the system claims can also be implemented by one unit or device through software or hardware. The words such as first and second are used to indicate names and do not indicate any specific order.

[0014] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A gas monitoring method based on edge-cloud collaboration, characterized in that: The method comprises: S1: Real-time monitoring of gas status based on equipment scenario data; S2: Determine the degree of gas hazard based on real-time pipeline network pressure; S3: predict the gas hidden danger status in the next time period; S4: Monitor the actual gas status and provide feedback to adjust the prediction method; S5: Generate a gas safety report based on the prediction results; S6: Output gas safety report; S7: Adjust gas control methods based on gas safety reports.

2. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S1 includes: S1.1, access the equipment scene data in real time through the edge node, and calculate the corrosion coefficient related to the gas equipment based on the scene data; The specific calculation formula of the corrosion coefficient is: ； Get the corrosion coefficient ,in, For the Pipe material data, Pipeline burial depth data, The service time data of the pipeline, is the soil moisture data, is the soil corrosivity data, and are the corresponding weight factors, and satisfy ; S1.2, calculate the aging coefficient related to the gas equipment based on the scenario data; The specific calculation formula of the aging coefficient is: ； Get the aging coefficient .

3. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S2 includes: S2.1, transmit gas pressure data to the edge node in real time through the pressure sensor, and calculate the pressure anomaly coefficient based on the historical gas pressure data; The specific calculation formula of the pressure anomaly coefficient is: ； Get the pressure anomaly coefficient ,in, For the Gas pressure data during the time period, is the historical average, is the standard deviation; S2.2, determine the pressure abnormality level based on the pressure abnormality coefficient; The specific calculation formula group of pressure abnormality level is: ； Get the abnormal pressure level .

4. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S3 includes: The LSTM time series model is trained on the cloud, and historical pressure data, corrosion coefficient, and aging coefficient are input to calculate the hidden danger prediction value in the next time period; The specific calculation formula of hidden danger prediction value is: ； in, is a pre-trained LSTM time series model, , and They are the corrosion coefficient, aging coefficient and historical pressure data input into the LSTM time series model respectively.

5. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S4 includes: S4.1, by obtaining gas pressure data and hidden danger prediction value , calculate the prediction deviation data , verify the hidden danger prediction value accuracy; The specific calculation formula for the forecast deviation data is: ； S4.2, based on the forecast deviation data and standard deviation Compare the results and choose to continue output or feedback to adjust the prediction method. When the prediction deviation data is greater than or equal to the standard deviation When the prediction deviation data is less than or equal to the standard deviation, the LSTM time series model is triggered to retrain. When the hidden danger prediction value is transmitted to step S5.

6. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S5 includes: The hidden danger prediction value and pressure abnormality level are summarized in the cloud. Match with the hidden danger level template to obtain hidden danger status data , matching the hidden danger status data and pressure abnormality level with the recommended measures template respectively, and generating a gas safety report; The specific calculation formula group that matches the hidden danger prediction value with the hidden danger level template is: 。 7. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S6 includes: By identifying the gas safety report, the transmission frequency and the transmission terminal are selected according to the data transmission unit; Identify gas safety reports including measures to identify low, medium and high risks in gas safety reports; When the gas safety report is a low-risk measure report, the gas safety report is sent to the staff mailbox receiving terminal; When the gas safety report is a medium-risk measure report, the gas safety report will be sent to the staff's mailbox receiving terminal and SMS reminders will be sent every half hour through the mobile phone receiving terminal; When the gas safety report is a high-risk measure report, the gas safety report will be sent to the staff's email receiving terminal and uninterrupted telephone reminders will be made through the mobile phone receiving terminal.

8. The gas monitoring method based on edge-cloud collaboration according to claim 1 is characterized in that: The S7 includes: By identifying the gas safety report, the control instructions issued by the cloud are transmitted to the programmable logic controller of the edge node; The specific calculation formula of the control instruction is: ； Get control instructions ,in, is the adjustment factor.

9. The gas monitoring method based on edge-cloud collaboration according to claim 5 is characterized in that: The gas safety report in step S5 includes a hidden danger recommended measures report and a pressure recommended measures report, wherein: The hidden danger recommended measures report includes hidden danger low risk measures report, hidden danger medium risk measures report and hidden danger high risk measures report; The low-risk measures report includes explaining that the gas equipment is running smoothly, asking staff to maintain the current valve opening and conduct pipeline inspections according to the regular schedule; The risk measures report for hidden dangers includes a description of possible pressure fluctuations in gas equipment, the coordinates of the risk location, staff members are asked to prepare emergency tools, and a group of representatives send adjustment instructions to the edge node; The high-risk measures report includes a description of the possible leakage risk of the gas equipment, the coordinates of the risk location, and a request for staff to arrive at the risk location as soon as possible for repairs. A group of representatives send adjustment instructions to the edge node; The stress recommended action reports include stress low risk action reports, stress medium risk action reports and stress high risk action reports; The pressure low risk measures report includes a statement that the gas equipment is running smoothly and asks the staff to conduct pipeline inspections according to routine calculations; The risk measures report for pressure medium includes a description of the short-term overpressure risk of the gas equipment, the coordinates of the risk location, asking staff to go to the risk location to check the valve sealing line, and a set of signals representing the adjustment instructions sent to the edge node; The pressure high risk measures report includes a description of suspected gas equipment leakage, the coordinates of the risk location, staff are requested to arrive at the risk location as soon as possible for emergency repairs, and a group of representatives send adjustment instructions to the edge node.

10. The gas monitoring method based on edge-cloud collaboration according to claim 8 is characterized in that: The method for determining the adjustment coefficient in step S7 includes: The specific calculation formula of the adjustment coefficient is: ； Get the adjustment factor ,in, To ensure safety, is the flow pressure coefficient.

Images