Fire blast risk early warning system for petrochemical industry and flame arrester

Abstract

The invention provides a blast risk early warning system and a flame arrester for petrochemical engineering, and relates to the technical field of risk early warning, the system comprises an in-pipe state prediction module used for generating a preset in-pipe distributed state network; the external state prediction module is used for generating a preset external distributed state network; the real-time monitoring module is used for generating an in-pipe real-time state network and an external real-time state network; the fire blast risk point location identification module is used for generating a first risk point location; the fire-retardant simulation module is used for generating a first fire-retardant failure index and a second fire-retardant failure index; and the failure early warning module is used for generating early warning information. The technical problems that in the prior art, due to the poor monitoring effect and lack of real-time performance evaluation of the flame arrester, the failure accident of the flame arrester cannot be predicted in time, and the fire blast risk early warning accuracy is poor can be solved, and the technical effects that the early warning accuracy is improved, and fire blast accidents are prevented from being expanded are achieved.

Description

Technical Field

[0001] The present application relates to the field of risk warning technology, and in particular to an explosion risk warning system and a flame arrester for petrochemical industry. Background Art

[0002] Petrochemical production involves high temperature, high pressure, flammable, explosive and other dangerous factors. Once an accident occurs, it will not only cause huge economic losses, but also may cause serious harm to personnel and the environment. Therefore, how to effectively warn and prevent the risk of explosion in petrochemical production has become an urgent problem to be solved in the industry. In the petrochemical production process, traditional explosion risk warning systems mostly rely on manual inspections and fixed sensor monitoring, or rely on flame arresters to block flame propagation and reduce losses. The monitoring effect is poor, and there is a lack of real-time evaluation of the performance of flame arresters, which makes it impossible to timely predict flame arrester failure accidents, further leading to the expansion of accidents.

[0003] In summary, the prior art has technical problems such as poor monitoring effect and lack of real-time evaluation of the performance of the flame arrester, which leads to the inability to timely predict flame arrester failure accidents and poor accuracy of explosion risk warning. Summary of the invention

[0004] The purpose of this application is to provide an explosion risk warning system and flame arrester for petrochemical industry, so as to solve the technical problems in the prior art that the flame arrester failure accident cannot be predicted in time due to poor monitoring effect and lack of real-time performance evaluation of the flame arrester, and the explosion risk warning accuracy is poor.

[0005] In view of the above problems, the present application provides an explosion risk warning system and a flame arrester for petrochemical industry.

[0006] In a first aspect, the present application provides an explosion risk warning system for petrochemical industry, the system comprising: an in-pipe state prediction module, used to obtain oil and gas pipelines in a preset petrochemical area, predict the internal distributed state of the pipeline based on the oil and gas transmission demand information, and generate a preset in-pipe distributed state network, wherein the preset in-pipe distributed state network includes multiple in-pipe nodes, and the multiple in-pipe nodes have in-pipe node identification information including preset temperature, preset flow rate and preset flow rate; an external state prediction module, used to predict the external distributed state of the pipeline based on the preset in-pipe distributed state network according to a preset monitoring radius, and generate a preset external distributed state network, wherein the preset external distributed state network includes multiple external nodes, and the multiple external nodes have external node identification information including external temperature, pressure and sound waves; a real-time monitoring module, used to monitor the internal and external state of the pipeline through the internal monitoring module and the external monitoring module of the pipeline. Perform real-time status monitoring on the multiple in-pipe nodes and the multiple external nodes to generate an in-pipe real-time status network and an external real-time status network; a combustion and explosion risk point identification module is used to compare the preset in-pipe distributed status network with the in-pipe real-time status network, the preset external distributed status network with the external real-time status network to identify the combustion and explosion risk points and generate a first risk point; a fire-blocking simulation module is used to locate the neighborhood flame arrester in the oil and gas pipeline based on the first risk point, generate a first neighborhood flame arrester and a second neighborhood flame arrester, and perform combustion and explosion fire-blocking simulation on the first neighborhood flame arrester and the second neighborhood flame arrester in combination with the in-pipe real-time status network to generate a first fire-blocking failure index and a second fire-blocking failure index; a failure warning module is used to generate a warning message when the first fire-blocking failure index and / or the second fire-blocking failure index are greater than a preset index, wherein the warning message carries the corresponding flame arrester.

[0007] In the second aspect, the present application also provides a flame arrester for use in petrochemical industry for warning of combustion and explosion risks, comprising: a flame arrester body and a communication port, wherein the communication port is used to receive the warning information output by the combustion and explosion risk warning system for petrochemical industry described in the first aspect, and the flame arrester body is used to execute a warning response according to the warning information.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The in-pipe state prediction module is used to obtain the oil and gas pipelines in the preset petrochemical area, predict the internal distributed state of the pipeline based on the oil and gas transmission demand information, and generate a preset in-pipe distributed state network, wherein the preset in-pipe distributed state network includes multiple in-pipe nodes, and the multiple in-pipe nodes have in-pipe node identification information including preset temperature, preset flow rate and preset flow rate; the external state prediction module is used to predict the external distributed state of the pipeline based on the preset in-pipe distributed state network according to the preset monitoring radius, and generate a preset external distributed state network, wherein the preset external distributed state network includes multiple external nodes, and the multiple external nodes have external node identification information including external temperature, pressure and sound wave; the real-time monitoring module is used to monitor the multiple in-pipe nodes and the multiple external nodes through the pipeline internal monitoring module and the pipeline external monitoring module. A real-time status monitoring is performed at each point to generate a real-time status network inside the pipe and an external real-time status network; a combustion and explosion risk point identification module is used to compare the preset distributed status network inside the pipe with the real-time status network inside the pipe, and the preset external distributed status network with the external real-time status network to identify the combustion and explosion risk points and generate a first risk point; a fire-blocking simulation module is used to locate the neighborhood flame arrester in the oil and gas pipeline based on the first risk point, generate a first neighborhood flame arrester and a second neighborhood flame arrester, and perform combustion and explosion fire-blocking simulation on the first neighborhood flame arrester and the second neighborhood flame arrester in combination with the real-time status network inside the pipe to generate a first fire-blocking failure index and a second fire-blocking failure index; a failure warning module is used to generate a warning message when the first fire-blocking failure index and / or the second fire-blocking failure index are greater than a preset index, wherein the warning message carries the corresponding flame arrester. By monitoring the internal and external environment of oil and gas pipelines, identifying explosion risk points, analyzing the fire failure indicators of neighboring flame arresters based on the risk points, and generating early warning information based on the fire failure indicators, we can achieve early warning of explosion risks for oil and gas pipelines and timely flame arrester failure warnings, achieving the technical effect of improving early warning accuracy and preventing the expansion of explosion accidents.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 This is a schematic diagram of the structure of the explosion risk warning system used in petrochemical industry in this application;

[0013] Figure 2 This is a flow chart of generating a first fire barrier failure index and a second fire barrier failure index in a petrochemical explosion risk warning system used in this application.

[0014] Description of reference numerals:

[0015] In-pipe state prediction module 11, external state prediction module 12, real-time monitoring module 13, explosion risk point identification module 14, fire retardant simulation module 15, failure warning module 16. DETAILED DESCRIPTION

[0016] This application solves the technical problems in the prior art that the flame arrester failure accidents cannot be predicted in time and the accuracy of the explosion risk warning is poor due to the poor monitoring effect and the lack of real-time evaluation of the flame arrester performance. By monitoring the internal and external environment of the oil and gas pipeline, the explosion risk points are identified, and the fire arrester failure index of the neighboring flame arrester is analyzed based on the risk point. Early warning information is generated according to the fire arrester failure index, and the explosion risk warning of the oil and gas pipeline is realized, and the flame arrester failure warning is carried out in time, so as to achieve the technical effect of improving the accuracy of the warning and preventing the expansion of explosion accidents.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] Embodiment 1

[0019] Please refer to the attached Figure 1 The present application provides an explosion risk warning system for petrochemical industry, the system comprising:

[0020] The in-pipe state prediction module 11 is used to obtain the oil and gas pipelines in the preset petrochemical area, predict the distributed state inside the pipeline based on the oil and gas transmission demand information, and generate a preset in-pipe distributed state network, wherein the preset in-pipe distributed state network includes multiple in-pipe nodes, and the multiple in-pipe nodes have in-pipe node identification information including preset temperature, preset flow rate and preset flow rate.

[0021] Specifically, the preset petrochemical area refers to the petrochemical production or processing area to be warned of the risk of explosion, which contains a large number of oil and gas pipelines for transporting oil. Based on the oil and gas transmission demand information, the internal distributed state of the pipeline is predicted, that is, according to the demand for oil and gas transmission, the oil and gas flow state at different positions inside the pipeline is predicted under normal conditions, that is, without leakage and other abnormal conditions, including temperature, flow rate and flow. According to the prediction results, different positions inside the pipeline are used as multiple in-pipe nodes, and the in-pipe node identification information is established based on the temperature, flow rate and flow of each node. The multiple in-pipe nodes are connected to generate a preset in-pipe distributed state network. The multiple in-pipe nodes and in-pipe node identification information in the preset in-pipe distributed state network are the ideal conditions under which oil and gas do not leak, that is, the ideal state under the condition of low explosion risk, which provides a basis for subsequent explosion risk judgment.

[0022] The external state prediction module 12 is used to predict the external distributed state of the pipeline based on the preset in-pipe distributed state network and according to the preset monitoring radius, and generate a preset external distributed state network, wherein the preset external distributed state network includes multiple external nodes, and the multiple external nodes have external node identification information including external temperature, pressure and sound waves.

[0023] Specifically, the cause of explosion accidents in petrochemical scenarios is mostly caused by leakage at a certain location of the oil and gas pipeline. Pipeline leakage will cause changes in the flow rate and flow of crude oil in the pipeline. At the same time, leakage will also cause changes in parameters such as temperature, pressure and sound waves around the pipeline. Therefore, it is necessary not only to monitor the flow rate, flow velocity and temperature in the pipe, but also to monitor parameters such as temperature, pressure and sound waves around the pipeline. The preset monitoring radius refers to the monitoring radius set with the oil and gas pipeline as the center, that is, a monitoring range is set around the pipeline, which is specifically set by professional and technical personnel in this field.

[0024] Within the preset monitoring radius, the external distributed state of the pipeline is predicted, that is, the temperature, pressure, and sound waves of the external environment of the pipeline are predicted under normal circumstances. According to the prediction results, the external distributed state prediction of the pipeline is generated. The external distributed state prediction of the pipeline includes multiple external nodes, each of which represents a specific location outside the pipeline. Each external node has external node identification information including external temperature, pressure, and sound waves. It provides support for subsequent explosion risk warning.

[0025] The real-time monitoring module 13 is used to perform real-time status monitoring on the multiple in-pipe nodes and the multiple external nodes through the pipeline internal monitoring module and the pipeline external monitoring module to generate an in-pipe real-time status network and an external real-time status network.

[0026] Specifically, the pipeline internal monitoring module refers to a group of sensors and detection equipment installed inside the pipeline, which is used to monitor the temperature, flow rate, flow velocity and other parameters inside the pipeline in real time. Similarly, the pipeline external monitoring module refers to a group of sensors and detection equipment installed outside the pipeline, which is used to monitor the environmental parameters such as temperature, pressure and sound waves outside the pipeline in real time. These external monitoring modules can be installed in the soil around the pipeline, or installed on the ground facilities near the pipeline, and can collect the status data of the external environment of the pipeline in real time. The real-time temperature, flow rate and flow velocity of the liquid at multiple nodes in the pipeline are collected in real time through the pipeline internal monitoring module to form a real-time state network inside the pipeline. The external temperature, pressure, sound waves and other parameters of multiple external nodes are collected in real time through the pipeline external monitoring module to form an external real-time state network. The real-time state network inside the pipeline shows the real-time state of each node inside the pipeline, including parameters such as temperature, flow velocity, flow rate, etc., while the external real-time state network shows the real-time state of each node outside the pipeline, including parameters such as external temperature, pressure, sound waves, etc. As a result, real-time information of the internal and external environment of the pipeline is obtained, which is convenient for timely detection of abnormal conditions in pipeline operation, such as leakage, and provides data support for subsequent explosion risk warning.

[0027] The explosion risk point identification module 14 is used to identify the explosion risk points by comparing the preset in-pipe distributed state network with the in-pipe real-time state network, the preset external distributed state network with the external real-time state network, and generate a first risk point.

[0028] Specifically, first, the data in the real-time status network in the pipe is compared with the preset distributed status network in the pipe, and the pipe nodes whose deviations from the preset temperature, preset flow rate and preset flow rate of the pipe nodes in the preset distributed status network in the pipe are obtained, and they are added to the first risk point, where the deviation threshold is set by professional and technical personnel in this field. The data in the external real-time status network is compared with the preset external distributed status network, and the external nodes whose deviations from the external temperature, pressure and sound waves of the external nodes in the preset external distributed status network exceed the deviation threshold are obtained, and the pipe nodes closest to the external nodes are identified and added to the first risk point. That is to say, changes in parameters such as temperature, pressure, flow rate, and flow rate inside and outside the pipeline can reflect the risk of explosion. For example, a sudden increase in temperature and an abnormal increase in pressure may indicate that there is a risk of explosion at this point.

[0029] The fire-blocking simulation module 15 is used to locate the neighborhood flame arrester in the oil and gas pipeline based on the first risk point, generate the first neighborhood flame arrester and the second neighborhood flame arrester, and perform explosion and fire-blocking simulation on the first neighborhood flame arrester and the second neighborhood flame arrester in combination with the real-time status network in the pipe to generate the first fire-blocking failure index and the second fire-blocking failure index.

[0030] Specifically, flame arresters are usually installed at different locations inside the oil and gas pipeline to prevent the spread of flames after an explosion accident occurs. They can be automatically activated when the flame spreads, and the spread of flames can be suppressed by physical or chemical means. Thus, two flame arresters located before and after the first risk point in the oil and gas pipeline are obtained. Since the flame may spread forward or backward at the first risk point, the two flame arresters are recorded as the first neighborhood flame arrester and the second neighborhood flame arrester respectively. The first neighborhood flame arrester and the second neighborhood flame arrester are simulated for explosion and fire prevention in combination with the real-time status network in the pipe, that is, the flame prevention effect of the flame arrester under actual working conditions is simulated using real-time monitoring data. Through simulation, the first fire prevention failure index and the second fire prevention failure index can be generated. The first fire prevention failure index and the second fire prevention failure index reflect the failure probability or failure risk of the first neighborhood flame arrester and the second neighborhood flame arrester under specific working conditions. If the index shows that the risk of flame arrester failure is high, it means that when an explosion accident occurs at the first risk point, the first neighborhood flame arrester and the second neighborhood flame arrester cannot effectively prevent the spread of flames, and there is a greater risk of accidents.

[0031] The failure warning module 16 is configured to generate a warning message when the first flame arrester failure index and / or the second flame arrester failure index is greater than a preset index, wherein the warning message carries the corresponding flame arrester.

[0032] Specifically, when the first fire-blocking failure index and / or the second fire-blocking failure index is greater than the preset index, it indicates that the corresponding flame arrester may not be able to effectively prevent the spread of flames under the current working conditions, and there is a risk of failure. At this time, an early warning message is generated so that corresponding measures can be taken in time. Among them, the preset index is set by professional and technical personnel in this field based on practical experience, and it is used as a criterion for judging whether the flame arrester has failed. Once it is detected that the first fire-blocking failure index and / or the second fire-blocking failure index exceeds the preset index, an early warning message is immediately generated. The early warning information should include the location of the flame arrester that exceeds the preset index, and the early warning information needs to be promptly transmitted to pipeline managers and relevant safety departments. This can be achieved through an automatic alarm system, text messages, emails or other means of communication. Thereby, an early warning of the risk of explosion is achieved, the safety performance of the pipeline system is effectively improved, and the occurrence of explosion accidents is reduced.

[0033] Furthermore, the in-pipe state prediction module 12 is also used for:

[0034] The pipeline connection structure of the oil and gas pipeline is obtained, and nodes with changed oil and gas transportation characteristics are identified to generate a plurality of changed node partitions; the plurality of in-pipe nodes are generated by the plurality of changed node partitions, and standardized predictions of the partition temperature, flow rate and flow rate are performed according to the oil and gas transmission demand information to generate preset temperature, preset flow rate and preset flow rate, and obtain the in-pipe node identification information; the preset in-pipe distributed state network is constructed by the plurality of in-pipe nodes and the in-pipe node identification information.

[0035] Furthermore, the in-pipe state prediction module 12 is also used for:

[0036] Positive samples are collected for the oil and gas pipeline to generate multiple groups of pipeline service record data, wherein any group of pipeline service record data includes oil and gas transmission demand samples and standard temperature samples, standard flow velocity samples and standard flow samples of the multiple in-pipe nodes; a multi-node standardized prediction model is trained and constructed based on the multiple groups of pipeline service record data; standardized prediction of the zoned temperature, flow velocity and flow rate of the oil and gas transmission demand information is performed through the multi-node standardized prediction model to generate the in-pipe node identification information.

[0037] Specifically, the pipeline connection structure of the oil and gas pipeline is obtained. The pipeline connection structure includes geometric features such as branching, intersection, and bending of the pipeline, as well as the type of pipeline connectors, valve positions, etc. The nodes that change the characteristics of oil and gas transportation are identified, that is, which nodes in the pipeline may cause changes in temperature, flow rate, and flow rate during oil and gas transmission. These change nodes may be caused by changes in pipeline diameter, the presence of valves, the influence of pump stations, or other factors. The identified change nodes are divided into different areas according to their impact range. The temperature, flow rate, and flow rate in each partition are relatively consistent, so that the risks in each partition can be predicted more accurately. Multiple in-pipe nodes are generated by partitioning with multiple change nodes, that is, representative locations are selected in each partition, such as the center location as in-pipe nodes. These nodes will serve as key points for state prediction and monitoring.

[0038] Furthermore, standardized predictions of the temperature, velocity and flow of each partition are made based on the oil and gas transmission demand information, that is, the temperature, velocity and flow of each partition under normal working conditions are predicted by using historical data and transmission demand, combined with the physical characteristics of the pipeline and the principles of fluid dynamics. These prediction results will be used as preset temperatures, preset velocity and preset flow. The preset temperature, preset velocity and preset flow corresponding to each in-pipe node are combined as in-pipe node identification information. Finally, a preset in-pipe distributed state network is constructed with multiple in-pipe nodes and in-pipe node identification information, that is, all in-pipe nodes and their identification information are integrated into a network to form a distributed state network inside the pipeline, which is used as the preset in-pipe distributed state network for real-time monitoring of the internal state of the pipeline, predicting risk points, and improving the accuracy of explosion risk warnings.

[0039] The steps of performing standardized prediction of zone temperature, flow rate and flow rate according to the oil and gas transmission demand information and generating preset temperature, preset flow rate and preset flow rate are as follows:

[0040] Specifically, positive samples are collected for oil and gas pipelines, that is, data of pipelines under normal service conditions are collected. These data reflect the state of the pipelines during normal operation, including oil and gas transmission requirements and parameters such as temperature, flow rate and flow rate at multiple nodes inside the pipeline, so as to generate multiple sets of pipeline service record data. Each set of pipeline service record data includes oil and gas transmission demand samples, as well as standard temperature samples, standard flow rate samples and standard flow rate samples at multiple nodes inside the pipeline. These standard samples are normal state parameters of the pipeline under specific conditions and can be used as training data to establish a prediction model. A multi-node standardized prediction model is constructed based on training of multiple sets of pipeline service record data, that is, machine learning algorithms such as regression analysis and neural networks are used to train the collected data. The goal of this model is to learn the behavior patterns of pipelines under different working conditions, so as to predict the states of different nodes inside the pipeline under different transmission requirements. The training of machine learning models is a common technical means used by technicians in this field and will not be expanded here.

[0041] The multi-node standardized prediction model is used to perform standardized predictions on the partitioned temperature, flow rate and flow of the oil and gas transmission demand information. That is, the current oil and gas transmission demand information is input, and the model will output each changed node partition, that is, the predicted temperature, flow rate and flow of each node in the pipe, which are used as the preset temperature, preset flow rate and preset flow, thereby generating the node identification information in the pipe, which is convenient for building a preset distributed status network in the pipe, and providing a basis for real-time monitoring and management of the pipeline.

[0042] Furthermore, the external state prediction module 13 is also used for:

[0043] For the multiple in-pipe nodes in the preset in-pipe distributed state network, multiple external feature mapping models corresponding to the multiple external nodes are constructed according to the preset monitoring radius; external feature prediction is performed on the in-pipe node identification information of the multiple in-pipe nodes based on the multiple external feature mapping models, and the external predicted temperatures, external predicted pressures and external predicted sound waves of the multiple external nodes are generated, and the external node identification information is established; the preset external distributed state network is constructed with the multiple external nodes and the external node identification information.

[0044] Specifically, for multiple in-pipe nodes in a preset in-pipe distributed state network, multiple external feature mapping models corresponding to multiple external nodes are constructed according to a preset monitoring radius. That is to say, within the range of the preset monitoring radius, a model is established for each external node to predict the impact of multiple in-pipe nodes on the external environment of the pipeline, and to predict the external temperature, pressure and sound waves of the external nodes within the preset monitoring radius. The external feature mapping model can be a data-driven machine learning model. Specifically, it can be obtained by collecting in-pipe node identification information samples under normal operating conditions of the oil and gas pipeline, including temperature samples, flow samples and flow velocity samples of the in-pipe nodes, and corresponding external node identification samples of each external node, including external temperature samples, pressure samples and sound wave samples. Using the in-pipe node identification information samples as training input and the external node identification samples as training output, multiple external feature mapping models are constructed based on the existing machine learning model training. The training of machine learning models is a common technical means used by those skilled in the art and will not be expanded here.

[0045] Based on multiple external feature mapping models, the internal node identification information of multiple internal nodes is predicted for external features, that is, the external feature mapping model is used to predict the temperature, pressure, sound wave and other parameters of the external environment of the pipeline, and the external predicted temperature, external predicted pressure and external predicted sound wave of multiple external nodes are generated to form the external node identification information. A preset external distributed state network is constructed with multiple external nodes and external node identification information, that is, all external nodes and their identification information are integrated into one network to form a preset external distributed state network. Provide support for early warning of oil and gas explosion risks.

[0046] Further, as attached Figure 2 As shown, the fire-blocking simulation module 15 is also used for:

[0047] Obtain first design parameters of the first neighborhood flame arrester, wherein the first design parameters include information on the fire-blocking structure and the fire-blocking material; perform a fire-blocking test based on the first design parameters to obtain a first upper limit condition for an explosive flame, wherein the first upper limit condition for an explosive flame includes a flame temperature and a pressure wave; use the first risk point as a simulated ignition point, and simulate the propagation state of an explosive flame in combination with the real-time state network in the tube to generate a first simulated flame state that propagates to the first neighborhood flame arrester; compare the first simulated flame state with the first upper limit condition for an explosive flame to generate the first fire-blocking failure index; continue to perform fire-blocking tests and explosion flame propagation state simulations on the second neighborhood flame arrester to generate the second fire-blocking failure index.

[0048] Furthermore, the fire-blocking simulation module 15 is also used for:

[0049] Acquire multiple groups of first combustion and explosion flame samples; construct a first fire-blocking digital model of the first neighborhood flame arrester based on the first design parameters; input the multiple groups of first combustion and explosion flame samples into the first fire-blocking digital model for simulation in sequence, and obtain multiple groups of first fire-blocking simulation results; determine whether the multiple groups of first fire-blocking simulation results include fire-blocking failure results, and if so, obtain a first continuous valid-failure sample combination; perform sample distance shrinkage variation based on the first continuous valid-failure sample combination to generate multiple groups of second combustion and explosion flame samples; test the multiple groups of second combustion and explosion flame samples through the first fire-blocking digital model, and obtain a second continuous valid-failure sample combination; determine whether the sample deviation of the second continuous valid-failure sample combination is less than a preset deviation, and if so, use the fire-blocking valid sample in the second continuous valid-failure sample combination as the upper limit condition of the first combustion and explosion flame, and if not, continue to perform sample distance shrinkage variation.

[0050] Furthermore, the fire-blocking simulation module 15 is also used for:

[0051] Determine whether the multiple groups of first fire-retardant simulation results include fire-retardant failure results. If not, perform sample expansion and mutation after serialization adjustment on the multiple groups of first combustion and explosion flame samples to generate multiple groups of third combustion and explosion flame samples; test the multiple groups of third combustion and explosion flame samples through the first fire-retardant digital model to obtain the upper limit condition of the first combustion and explosion flame.

[0052] Specifically, the first design parameter includes the fire-blocking structure and fire-blocking material information of the first neighborhood flame arrester, which can evaluate the performance of the flame arrester and determine the conditions under which the flame arrester can effectively suppress the spread of flames. It needs to be determined in combination with actual conditions. The fire-blocking test based on the first design parameter is to conduct a fire-blocking effect test to obtain the first upper limit condition of the explosion flame. The first upper limit condition of the explosion flame refers to the maximum flame temperature and pressure wave that the first neighborhood flame arrester can safely suppress.

[0053] The first risk point is used as the simulated ignition point, and the explosion flame propagation state is simulated in combination with the real-time state network in the pipe, that is, the real-time monitoring data is used to simulate the process of flame propagation from the first risk point. Specifically, the pipeline characteristics from the first risk point to the first neighborhood flame arrester can be collected, including pipeline diameter, length, slope and other characteristics, and the flame propagation data under the same pipeline characteristics can be collected. The flame propagation data includes the temperature and pressure of the flame after propagation, and a simulation platform is built based on the existing digital twin platform. Through the simulation platform, the flame propagation from the first risk point to the first neighborhood flame arrester is simulated, and the first simulated flame state propagating to the first neighborhood flame arrester can be generated, that is, the state of the simulated flame when it reaches the first neighborhood flame arrester during the propagation process, including parameters such as the temperature and pressure of the flame.

[0054] The first simulated flame state is compared with the first upper limit condition of the explosion flame, that is, the flame state parameters obtained by simulation are compared with the upper limit condition of the flame that can be safely suppressed by the flame arrester. If the first simulated flame state exceeds the first upper limit condition of the explosion flame, it indicates that the flame arrester may not be able to effectively suppress the flame, and the first flame arrest failure index is generated according to the degree of deviation between the two. If the simulated flame state does not exceed the first upper limit condition of the explosion flame, the first flame arrest failure index is initialized to 0.

[0055] The same method is used to continue to perform fire barrier tests and explosion flame propagation state simulation on the second neighborhood flame arrester, that is, repeat the above process, test and simulate the second neighborhood flame arrester, and generate a second fire barrier failure index. The process of generating the second fire barrier failure index is the same as the process of generating the first fire barrier failure index, which will not be repeated here. The first fire barrier failure index and the second fire barrier failure index are crucial to evaluating the actual working effect and safety performance of the flame arrester. They can evaluate whether the flame arrester can effectively prevent the spread of flames under actual working conditions, thereby providing an early warning of the risk of explosion and ensuring the safe operation of the pipeline.

[0056] Among them, the process of performing a fire-blocking test based on the first design parameter to obtain the upper limit condition of the first explosion flame is as follows: Specifically, multiple groups of first explosion flame samples are obtained, that is, data containing parameters such as temperature and pressure of the explosion flame under different conditions are collected as multiple groups of first explosion flame samples. Further, using the fire-blocking structure and fire-blocking material information of the first neighborhood flame arrester, a mathematical or computational model is established as the first fire-blocking digital model, which is used to simulate the fire-blocking effect of the flame arrester under actual working conditions. Specifically, it can be constructed based on existing digital twin or three-dimensional simulation technology, which is a common technical means for those skilled in the art and will not be expanded here.

[0057] Multiple groups of first explosion flame samples are sequentially input into the first fire-blocking digital model for simulation, that is, multiple groups of first explosion flame samples are used as input, the first fire-blocking digital model is run, and it is observed whether the flame arrester can effectively suppress the flame, and multiple groups of first fire-blocking simulation results are obtained, that is, the fire-blocking effect data corresponding to each group of samples, such as whether the flame is blocked from the first neighborhood flame arrester. Then, it is determined whether the multiple groups of first fire-blocking simulation results contain fire-blocking failure results, that is, it is checked whether there is a situation in the simulation results where the flame arrester fails to effectively suppress the flame. If there is a fire-blocking failure result, further based on the multiple groups of first fire-blocking simulation results, continuous valid and failed samples are selected, that is, multiple groups of first explosion flame samples are serialized and adjusted from small to large, and combined with the corresponding first fire-blocking simulation results, two explosion flame samples whose continuous simulation results change from effectively suppressing the flame to failing to effectively suppress the flame are obtained, forming the first continuous valid-failure sample combination, so as to further determine the critical conditions for the failure of the first neighborhood flame arrester.

[0058] Based on the first continuous valid-failure sample combination, the sample distance shrinkage variation is performed, that is, the data between the continuous valid and failed samples are adjusted, that is, according to the preset distance threshold, multiple new explosion flame samples less than the preset distance threshold are generated between two samples in the first continuous valid-failure sample combination as multiple groups of second explosion flame samples for further testing and evaluating the performance of the first neighborhood flame arrester. Multiple groups of second explosion flame samples are tested by the first flame arresting digital model, and the second continuous valid-failure sample combination is obtained.

[0059] Determine whether the sample deviation of the second consecutive valid-failed sample combination is less than the preset deviation. The sample deviation refers to the degree of difference between valid and failed samples. If the sample deviation is less than the preset deviation, it indicates that the critical state is basically reached, and the flame arrester valid sample in the second consecutive valid-failed sample combination is used as the first upper limit condition for the explosion flame. If the sample deviation is still large, it is necessary to continue to shrink the sample distance until the preset deviation is reached. Through simulation and testing, the performance of the flame arrester under different conditions is accurately evaluated, and the failure mechanism of the flame arrester is evaluated by continuously adjusting and optimizing the samples, thereby improving the accuracy of the failure index evaluation, and then improving the accuracy of the explosion risk warning.

[0060] Wherein, when judging whether the plurality of groups of first fire-blocking simulation results include fire-blocking failure results, if the plurality of groups of first fire-blocking simulation results do not include fire-blocking failure results, it means that there is a high possibility that the flame upper limit condition of the first neighborhood flame arrester is not detected. At this time, the plurality of groups of first combustion and explosion flame samples are serialized and adjusted, and then sample expansion and mutation are performed to generate a plurality of groups of third combustion and explosion flame samples;

[0061] Specifically, judging whether multiple groups of first flame arrester simulation results contain flame arrester failure results means checking whether the flame arrester fails to effectively suppress the flame under certain conditions during the simulation. If the simulation results show that the flame arrester can work normally under all conditions, that is, there is no flame arrester failure, then it is necessary to perform sample expansion and variation on the first combustion and explosion flame sample, that is, modify the original sample data to generate more diverse combustion and explosion flame samples. For example, adjust the temperature, pressure, propagation speed and other parameters of the flame, output the adjusted and mutated sample data, and generate multiple groups of third combustion and explosion flame samples, which are more diverse.

[0062] Through the first fire-blocking digital model, multiple groups of third explosion flame samples are tested, that is, the fire-blocking digital model is run again using these new sample data to observe the performance of the first neighborhood flame arrester. If the fire-blocking simulation result still does not include the fire-blocking failure result, the sample expansion and variation is continued until the fire-blocking simulation result includes the fire-blocking failure result, and then the continuous valid-failure sample combination is continued to be obtained to generate the first explosion flame upper limit condition, that is, the maximum flame temperature and pressure parameters that the flame arrester can safely suppress. Through the expansion and variation of samples, more diverse test data is generated, so as to more comprehensively evaluate the performance of the flame arrester, ensure the accuracy of the first explosion flame upper limit condition, and then improve the accuracy of the explosion risk warning.

[0063] In summary, the explosion risk warning system for petrochemical industry provided by this application has the following technical effects:

[0064] The in-pipe state prediction module is used to obtain the oil and gas pipelines in the preset petrochemical area, predict the internal distributed state of the pipeline based on the oil and gas transmission demand information, and generate a preset in-pipe distributed state network, wherein the preset in-pipe distributed state network includes multiple in-pipe nodes, and the multiple in-pipe nodes have in-pipe node identification information including preset temperature, preset flow rate and preset flow rate; the external state prediction module is used to predict the external distributed state of the pipeline based on the preset in-pipe distributed state network according to the preset monitoring radius, and generate a preset external distributed state network, wherein the preset external distributed state network includes multiple external nodes, and the multiple external nodes have external node identification information including external temperature, pressure and sound wave; the real-time monitoring module is used to monitor the multiple in-pipe nodes and the multiple external nodes through the pipeline internal monitoring module and the pipeline external monitoring module. A real-time status monitoring is performed at each point to generate a real-time status network inside the pipe and an external real-time status network; a combustion and explosion risk point identification module is used to compare the preset distributed status network inside the pipe with the real-time status network inside the pipe, and the preset external distributed status network with the external real-time status network to identify the combustion and explosion risk points and generate a first risk point; a fire-blocking simulation module is used to locate the neighborhood flame arrester in the oil and gas pipeline based on the first risk point, generate a first neighborhood flame arrester and a second neighborhood flame arrester, and perform combustion and explosion fire-blocking simulation on the first neighborhood flame arrester and the second neighborhood flame arrester in combination with the real-time status network inside the pipe to generate a first fire-blocking failure index and a second fire-blocking failure index; a failure warning module is used to generate a warning message when the first fire-blocking failure index and / or the second fire-blocking failure index are greater than a preset index, wherein the warning message carries the corresponding flame arrester. By monitoring the internal and external environment of oil and gas pipelines, identifying explosion risk points, analyzing the fire failure indicators of neighboring flame arresters based on the risk points, and generating early warning information based on the fire failure indicators, we can achieve early warning of explosion risks for oil and gas pipelines and timely flame arrester failure warnings, achieving the technical effect of improving early warning accuracy and preventing the expansion of explosion accidents.

[0065] Embodiment 2

[0066] Based on the same inventive concept as the explosion risk warning system for petrochemical industry in the aforementioned embodiment, the present application also provides an explosion risk warning flame arrester for petrochemical industry, including a flame arrester body and a communication port, wherein the communication port is used to receive the warning information output by the explosion risk warning system for petrochemical industry described in Example 1, and the flame arrester body is used to perform a warning response according to the warning information, that is, a warning response device is provided on the flame arrester body, which is used to respond to the warning information and generate a warning signal, and the warning signal can be information such as emails and messages sent to management personnel through the communication port.

[0067] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0068] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. An explosion risk warning system for petrochemical industry, characterized in that: The system comprises: The in-pipe state prediction module is used to obtain the oil and gas pipelines in the preset petrochemical area, predict the internal distributed state of the pipeline based on the oil and gas transmission demand information, and generate a preset in-pipe distributed state network, wherein the preset in-pipe distributed state network includes a plurality of in-pipe nodes, and the plurality of in-pipe nodes have in-pipe node identification information including a preset temperature, a preset flow rate, and a preset flow rate; An external state prediction module is used to predict the external distributed state of the pipeline based on the preset in-pipe distributed state network and according to the preset monitoring radius, and generate a preset external distributed state network, wherein the preset external distributed state network includes a plurality of external nodes, and the plurality of external nodes have external node identification information including external temperature, pressure and sound wave; A real-time monitoring module, used to perform real-time status monitoring on the multiple in-pipe nodes and the multiple external nodes through the pipeline internal monitoring module and the pipeline external monitoring module, and generate an in-pipe real-time status network and an external real-time status network; An explosion risk point identification module is used to identify explosion risk points by comparing the preset in-pipe distributed state network with the in-pipe real-time state network, the preset external distributed state network with the external real-time state network, and generate a first risk point; a fire arrester simulation module, for locating a neighborhood flame arrester in the oil and gas pipeline based on the first risk point, generating a first neighborhood flame arrester and a second neighborhood flame arrester, and performing explosion and fire arrester simulation on the first neighborhood flame arrester and the second neighborhood flame arrester in combination with the real-time status network in the pipeline, and generating a first fire arrester failure index and a second fire arrester failure index; The failure warning module is used to generate warning information when the first fire arresting failure index and / or the second fire arresting failure index is greater than a preset index, wherein the warning information carries the corresponding flame arrester.

2. The explosion risk warning system for petrochemical industry according to claim 1, characterized in that: The in-pipe state prediction module is also used for: Acquire the pipeline connection structure of the oil and gas pipeline, identify nodes with changed oil and gas transportation characteristics, and generate multiple changed node partitions; The plurality of in-pipe nodes are generated by partitioning the plurality of changed nodes, and standardized prediction of the partition temperature, flow velocity and flow rate is performed according to the oil and gas transmission demand information to generate a preset temperature, a preset flow velocity and a preset flow rate, and obtain the in-pipe node identification information; The preset in-pipe distributed state network is constructed using the multiple in-pipe nodes and the in-pipe node identification information.

3. The explosion risk warning system for petrochemical industry according to claim 2, characterized in that: The in-pipe state prediction module is also used for: Collecting positive samples of the oil and gas pipeline to generate multiple sets of pipeline service record data, wherein any set of pipeline service record data includes an oil and gas transmission demand sample and standard temperature samples, standard flow rate samples and standard flow samples of the multiple nodes in the pipeline; Building a multi-node standardized prediction model based on the multiple sets of pipeline service record data training; The multi-node standardized prediction model is used to perform standardized prediction of the zone temperature, flow velocity and flow rate of the oil and gas transmission demand information to generate the in-pipe node identification information.

4. The explosion risk warning system for petrochemical industry according to claim 1, characterized in that: The external state prediction module is also used for: For multiple in-pipe nodes in the preset in-pipe distributed state network, construct multiple external feature mapping models corresponding to multiple external nodes according to the preset monitoring radius; Based on the multiple external feature mapping models, external feature prediction is performed on the in-tube node identification information of the multiple in-tube nodes, external predicted temperatures, external predicted pressures and external predicted sound waves of the multiple external nodes are generated, and the external node identification information is established; The preset external distributed state network is constructed using the multiple external nodes and the external node identification information.

5. The explosion risk warning system for petrochemical industry according to claim 1, characterized in that: The fire blocking simulation module is also used for: Acquire first design parameters of the first neighborhood flame arrester, wherein the first design parameters include fire-blocking structure and fire-blocking material information; Perform a fire-blocking test based on the first design parameter to obtain a first upper limit condition for an explosion flame, wherein the first upper limit condition for an explosion flame includes a flame temperature and a pressure wave; Taking the first risk point as a simulated fire point, combining the real-time state network in the pipe to simulate the explosion flame propagation state, and generating a first simulated flame state that propagates to the first neighborhood flame arrester; Comparing the first simulated flame state with the first upper limit condition of the explosion flame to generate the first flame arresting failure index; Continue to perform fire arresting test and explosion flame propagation state simulation on the second neighborhood flame arrester to generate the second fire arresting failure index.

6. The explosion risk warning system for petrochemical industry according to claim 5, characterized in that: The fire blocking simulation module is also used for: Acquire multiple groups of first explosion flame samples; Constructing a first fire-blocking digital model of the first neighborhood flame arrester based on the first design parameters; Inputting the plurality of groups of first explosion flame samples into the first fire-blocking digital model in sequence for simulation, and obtaining a plurality of groups of first fire-blocking simulation results; Determine whether the plurality of groups of first fire-blocking simulation results include fire-blocking failure results, and if so, obtain a first continuous valid-failure sample combination; Perform sample distance shrinkage variation based on the first continuous valid-failed sample combination to generate multiple groups of second explosion flame samples; Testing the plurality of groups of second explosion flame samples by using the first fire-blocking digital model, and obtaining a second continuous valid-failure sample combination; Determine whether the sample deviation of the second continuous valid-failed sample combination is less than the preset deviation. If so, use the fire-retardant valid sample in the second continuous valid-failed sample combination as the first explosion flame upper limit condition. If not, continue to perform sample distance shrinkage variation.

7. The explosion risk warning system for petrochemical industry according to claim 6, characterized in that: The fire blocking simulation module is also used for: Determine whether the plurality of groups of first fire-blocking simulation results include fire-blocking failure results. If not, perform sample expansion and variation after serialization adjustment on the plurality of groups of first combustion and explosion flame samples to generate a plurality of groups of third combustion and explosion flame samples. The plurality of groups of third explosion flame samples are tested by using the first fire-blocking digital model to obtain the first explosion flame upper limit condition.

8. The explosion risk warning flame arrester used in petrochemical industry is characterized by: The explosion risk warning flame arrester for petrochemical industry includes a flame arrester body and a communication port, wherein the communication port is used to receive the warning information output by the explosion risk warning system for petrochemical industry described in claims 1 to 7, and the flame arrester body is used to execute a warning response according to the warning information.

Images