A method for automatically writing high-consequence area-by-area case reports based on a large model

Abstract

The present invention relates to the field of pipeline safety technology, and provides a method for automatically compiling a high consequence area one area one case report based on a large model, including: dividing the monitoring area into functional areas, setting different monitoring mechanisms for the monitoring items of each functional area, and converting the data stream generated by the monitoring mechanism into monitoring text; obtaining the historical high consequence area report of each type of functional area, correspondingly extracting the monitoring text when the historical high consequence area report was issued, obtaining the evaluation weight according to the keyword information in the evaluation results issued by the historical high consequence area report, and constructing a mapping relationship to generate a basic evaluation model for the functional area, and whenever the interval unit time and the preset conditions occur, inputting the real-time monitoring data into the basic evaluation model of the functional area to generate the E area Z case high consequence area report. The present invention sets personalized monitoring mechanisms and evaluation models for different functional areas, and provides more targeted risk assessment and reporting.

Description

Technical Field

[0001] The present invention relates to the field of pipeline safety technology, and in particular to a method for automatically compiling a high-consequence area-by-area case report based on a large model. Background Art

[0002] High Consequence Areas (HCAs) refer to areas where pipeline leaks may cause significant adverse impacts on public health, safety, the environment and society.

[0003] The definition of high-consequence areas takes into account specific types of areas, such as densely populated areas, important facility areas or environmentally sensitive areas, and is divided into different levels according to the severity of the accident consequences. The specific geographical location and scope of high-consequence areas may change with factors such as population growth, resource and environmental changes, and urban development. These areas are the focus of government supervision and social attention, and are also key areas for pipeline and enterprise safety management.

[0004] In order to effectively manage and ensure the safety of high consequence areas (HCAs), companies and regulatory agencies usually adopt a variety of monitoring methods and regular inspection measures and write detailed report documents.

[0005] In the prior art, CN 110070242 B discloses a gas pipeline high consequence area identification and evaluation system and method, which analyzes and processes information in a pre-established database module to extract pipeline basic data and surrounding geographic information data; from the pipeline basic data and surrounding geographic information data, according to a pre-set consequence element selection rule, the consequence element information data related to the pipeline high consequence area is extracted to form a consequence element set; the data in the consequence element set is used to classify the area along the pipeline according to pre-set regional grade assessment parameters to obtain regional grade data; the consequence element set data and the regional grade data are identified, and the identification is performed according to a pre-set gas pipeline high consequence area identification criterion to obtain the high consequence area; and a graded evaluation is performed according to the severity. However, in the high consequence area review process, the information is collected in a manner involving the fusion of multiple types of information streams, and the text processing and fusion of the assessment results of multiple data sources are not disclosed, that is, the technical solution cannot solve this technical problem.

[0006] The writing of high consequence area reports involves human inspections, sensor networks, data collectors and communication technologies. In addition, there are multiple monitoring levels and risk levels in high consequence areas. Therefore, it is difficult to efficiently integrate various data streams related to the review in high consequence areas with existing technologies.

[0007] Therefore, it is necessary to design an automatic writing method for high-consequence area one-area one-case report based on a large model to solve the technical problem of difficulty in efficiently integrating various data streams related to the review in the high-consequence area. Summary of the invention

[0008] In view of this, the present invention proposes an automatic writing method for high consequence area one area one case report based on a large model, aiming to solve the technical problem that it is difficult to efficiently integrate various data streams related to the review in the high consequence area.

[0009] The present invention proposes a method for automatically writing a high-consequence area-by-area-by-case report based on a large model, comprising:

[0010] Divide the monitoring area into functional areas, set different monitoring mechanisms for monitoring items in each functional area, and convert the data stream generated by the monitoring mechanism into monitoring text;

[0011] Obtaining a historical high consequence area report for each type of functional area, correspondingly extracting the monitoring text when the historical high consequence area report was issued, and analyzing regional characteristics according to the monitoring mechanism;

[0012] Compare each of the described regional characteristics with historical high consequence area reports to filter keyword information;

[0013] Acquire assessment weights based on keyword information in the historical high consequence area report, and construct a mapping relationship based on the assessment weights to generate a basic assessment model for the functional area;

[0014] Whenever the unit time interval and the preset conditions occur, the real-time monitoring data is input into the basic assessment model of the functional area to generate the E area Z case high consequence area report;

[0015] The E Zone Z Case High Consequence Area Report and its writing logic are stored in the server.

[0016] Preferably, the monitoring area is divided into functional areas, and different monitoring mechanisms are set for the monitoring items in each functional area, including:

[0017] Determine the geographical location of the oil and gas pipeline and the area it passes through, set the first pipeline density as the first-level monitoring area A, set the second pipeline density as the second-level monitoring area B, and set the third pipeline density as the third-level monitoring area C, wherein the first pipeline density is greater than the second pipeline density, which is greater than the third pipeline density, and the higher the monitoring level, the smaller the interval between the upper threshold and the lower threshold for triggering notification in the monitoring of the monitoring project, wherein the first-level monitoring area A is monitored through sensor technology and manual inspections to generate real-time data streams, for the second-level monitoring area B, the population density within its risk range is evaluated to set the monitoring method, and for the third-level monitoring area C, the environmental sensitivity within its risk range is evaluated to set the monitoring method;

[0018] The functional area division is used to divide the monitoring area into a number of different functional areas according to the different characteristics, uses and risk levels of each area;

[0019] The setting of different monitoring mechanisms for the monitoring items of each functional area also includes setting a corresponding management unit for each functional area and a corresponding supervision strategy for the category f transported by the pipeline in each management unit.

[0020] Preferably, for the secondary monitoring area B, the population density within the risk range is assessed and the monitoring method is set, including:

[0021] Obtain the outer diameter d of each pipeline involved in the secondary monitoring area B, where the unit of the outer diameter is millimeter, and the maximum allowable operating pressure P of the pipeline, where the unit of the operating pressure is MPa, and calculate the radius R of the affected area of ​​each pipeline leakage, wherein the radius R of the affected area is calculated by the following formula:

[0022] ;

[0023] According to the radius R of the affected area, the path H of each relevant pipeline in the secondary monitoring area B is obtained, and the risk range is set to each point on the path H as the center point and each area covered by the radius R of the affected area;

[0024] Obtaining a population density M within the risk range, comparing it with a preset first population density M1 and a second population density M2, where M1<M2, and setting a monitoring method according to the comparison result;

[0025] When M<M1, remote sensing technology is used for monitoring to generate periodic data streams;

[0026] When M1≤M<M2, sensor technology is used for monitoring to generate real-time data streams;

[0027] When M2≤M, sensor technology plus manual inspections are used to generate real-time data streams.

[0028] Preferably, the monitoring method for assessing the environmental sensitivity within the risk range of the third-level monitoring area C includes:

[0029] The environmental sensitivity influencing factors involved in the third-level monitoring area C are obtained, and the environmental sensitivity Q is calculated by the following formula:

[0030]

[0031] Where n is the number of environmental sensitive factors, is the weight of the xth indicator, is the score of the xth indicator;

[0032] Obtain the environmental sensitivity Q, compare it with a preset first environmental sensitivity Q1 and a second environmental sensitivity Q2, wherein Q1<Q2, and set a monitoring method according to the comparison result;

[0033] When Q<Q1, remote sensing technology is used to monitor the third-level monitoring area in the first cycle to generate a periodic data stream;

[0034] When Q1≤Q<Q2, remote sensing technology is used to monitor the third-level monitoring area in the second period to generate a periodic data stream;

[0035] When Q2≤Q, remote sensing technology is used to monitor the third-level monitoring area in the third period to generate a periodic data stream;

[0036] The monitoring frequency of the first cycle is smaller than the monitoring frequency of the second cycle, which is smaller than the monitoring frequency of the third cycle.

[0037] Preferably, converting the data stream generated by the monitoring mechanism into monitoring text comprises:

[0038] The timestamp j and location stamp k of each piece of information are automatically captured in the monitoring text to associate events from different data sources and evaluate related events across regions or across time;

[0039] The monitoring text also presets a regional monitoring text Xg and a unit monitoring text Xgi, where i is the number of related facilities in the area, g represents the type of monitoring data, X=A, B, C;

[0040] The unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, wherein the monitoring and screening is multi-dimensional monitoring, and the data of each dimension is screened, the multi-dimensional monitoring is set according to the pipeline transportation category f, and a significant threshold is set for each dimension, and non-significant data points that are less than the significant threshold within a unit time are eliminated, and the retained data are recorded as significant data points, and the significant data points are output as information, and the multi-dimensional monitoring results in the relevant area are refined to generate the regional monitoring text Xg;

[0041] The unit monitoring text Xgi adopts a preset structured time-series text format and is configured with a distributed retrieval index, which is used to actively search other surrounding areas in the report through the distributed retrieval index after the high-consequence report is issued when a facility detection data point is not within the automatic generation range.

[0042] Preferably, a historical high consequence area report is obtained for each type of functional area, and the monitoring text when the historical high consequence area report is issued is correspondingly extracted, and the regional characteristics are analyzed according to the monitoring mechanism, and each of the regional characteristics is compared with the historical high consequence area report to filter keyword information, including:

[0043] The topic information in the historical high consequence area report is analyzed by using the topic model LDA, and the functional area and facility detection data points to which it belongs are matched. The words that overlap with the monitoring data type g of the corresponding facility point measurement point in the content text of the historical high consequence area report are extracted as keywords, and part-of-speech tagging is used for classification and integration to generate keyword information;

[0044] The evaluation weight is obtained based on the keyword information in the historical high consequence area report, including:

[0045] Wherein, a basic weight is set for the category to which the part-of-speech tag of each keyword information in the classification integration belongs, and a first modified weight and a second modified weight are set to adjust the basic weight to generate the evaluation weight;

[0046] Extract the chart information in the historical high consequence area report, obtain the field position and numerical text or photo number in the chart information, set different first correction weights according to the field position, set the second correction weight according to the size of the numerical text, and mark the photo number.

[0047] Preferably, the step of constructing a mapping relationship according to the evaluation weights to generate the basic evaluation model of the functional area includes:

[0048] Obtain a historical high consequence area report for each of the functional areas, perform text preprocessing on the content text in the historical high consequence area report, perform sentence segmentation on the preprocessed text, wherein each sentence only includes one causal relationship, use dependency syntax to expand the sentence structure, and extract causal relationship words therein, construct features for the relationship words, use a machine learning algorithm to train the mapping relationship to determine the functional area to which it belongs, and generate a verification set based on the facility detection data points in the content text determined by the photo number in the historical high consequence area report, verify the mapping relationship, generate a basic assessment model for each functional area, and compile the time stamp j and location stamp k of the information source into the basic assessment model.

[0049] Preferably, whenever a unit time interval and a preset condition occur, the real-time monitoring data is input into the basic assessment model of the functional area to generate an E-area Z-case high consequence area report, including:

[0050] The preset conditions include that when the unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, the monitoring and screening is multi-dimensional monitoring, and the data of each dimension is screened, the multi-dimensional monitoring is set according to the pipeline transportation category f, and an alarm threshold is set for each dimension, wherein the alarm threshold is greater than the significant threshold, and normal data points that are less than the alarm threshold within a unit time are eliminated, and the retained data are recorded as dangerous data points, and the dangerous data points are output as information to generate the corresponding E zone Z case high consequence area report.

[0051] Preferably, the E-Zone Z Case High Consequence Area Report and its writing logic are stored in a server, and the coding method of the E-Zone Z Case High Consequence Area Report is that E is a matrix of numbers composed of Xgi-f, and Z is a single pipeline numbered sequentially along the oil and / or air flow direction;

[0052] The writing logic includes obtaining the timestamp j and location stamp k of each data acquisition and the monitoring method, and explaining the principles of the significant threshold and alarm threshold of each facility detection data point, and using a trend chart or a line chart to represent the trend of the significant threshold and alarm threshold in unit time.

[0053] Preferably, the E-Zone Z-Case High Consequence Area Report and its writing logic are stored in a server, further comprising:

[0054] Directly sending the high consequence area report to review server nodes, each of which hosts review expert resources;

[0055] Obtain the matrix of item E in the code of the high consequence area report, determine the technical field of the server address according to the transported category f in Xgi-f, determine the professional field of the server address according to the monitored data type g, obtain the expert with the closest distance to the location stamp k, and associate the years of work experience of the expert resource according to the level of the X monitoring area, where:

[0056] When X=A, only expert resources with more than 10 years of experience are associated;

[0057] When X=B, the associated experience is more than 6 years of expert resources;

[0058] When X=C, and the number of relevant facilities in the area i≥3, the expert resources with more than 3 years of experience are associated;

[0059] When X=C, and when the number of relevant facilities in the region i<3, expert resources with less than 3 years of experience are associated.

[0060] Compared with the prior art, the beneficial effects of the present invention are that, through the functional area division and monitoring mechanism setting, efficient integration of various monitoring data in high consequence areas is achieved, and the technical difficulties of data flow integration are solved; based on historical high consequence area reports and monitoring texts, in-depth analysis of regional characteristics is carried out to ensure the accuracy and reliability of the assessment model; by constructing a basic assessment model and inputting monitoring data in real time, dynamic risk assessment of high consequence areas is achieved to ensure timely discovery and response to risks; using large models and data-driven methods, high consequence area reports are automatically generated to improve the efficiency and accuracy of report writing; the generated reports and their writing logic are stored on the server to ensure systematic management and convenient retrieval of reports; through the above-mentioned technical means, the safety management level of high consequence areas is improved to ensure that major adverse effects on public health, safety, environment and society are minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0062] Figure 1 A flowchart of a method for automatically writing a high-consequence area-by-area-case report based on a large model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0063] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0064] See also Figure 1 As shown, this embodiment proposes a method for automatically writing a high-consequence area-by-area-case report based on a large model, including:

[0065] Step S1: dividing the monitoring area into functional areas, setting different monitoring mechanisms for monitoring items in each functional area, and converting the data stream generated by the monitoring mechanism into monitoring text;

[0066] Step S2: obtaining a historical high consequence area report for each type of functional area, correspondingly extracting the monitoring text when the historical high consequence area report was issued, and analyzing regional characteristics according to the monitoring mechanism;

[0067] Step S3: Compare each of the regional characteristics with historical high consequence area reports to filter keyword information;

[0068] Step S4: obtaining evaluation weights according to the keyword information in the historical high consequence area report, and constructing a mapping relationship according to the evaluation weights to generate a basic evaluation model for the functional area;

[0069] Step S5: whenever a unit time interval and a preset condition occur, input the real-time monitoring data into the basic assessment model of the functional area to generate a high consequence area report for Zone E and Case Z;

[0070] Step S6: storing the E-Zone Z Case High Consequence Area Report and its writing logic to the server.

[0071] Specifically, according to step S1 of this embodiment: the purpose is to divide the functional areas and set up the monitoring mechanism. The construction idea is to divide the areas according to different functions and needs in the monitoring area. Each functional area will have different monitoring items, such as environmental parameters, equipment operating status, etc. Different monitoring mechanisms are set for each monitoring item. These mechanisms will continuously collect data and convert it into monitoring text; the main purpose of this step is to establish a detailed and organized monitoring system to ensure that the data in each area can be accurately captured and recorded.

[0072] According to step S2 of this embodiment, the purpose is to obtain historical data and analyze regional characteristics. The construction idea is to collect historical high-consequence area reports for each type of functional area and extract the monitoring texts when these reports were issued. By analyzing these monitoring texts, the characteristic performance of each area in different situations can be understood, for example, the monitoring data characteristics when equipment fails or the environment is abnormal. This step is to accumulate empirical data and understand the performance of different areas in various situations for subsequent modeling.

[0073] According to step S3 of this embodiment, the purpose is to filter keywords, and the idea is to extract keyword information (such as fault, danger, alarm, etc.) in the assessment results from historical high consequence area reports.

[0074] According to step S4 of this embodiment: the purpose is to construct a basic evaluation model, assign certain weights to these keywords, establish a mapping relationship between keywords and evaluation results, and based on this mapping relationship, combine historical monitoring texts to construct a basic evaluation model for each functional area. This step is the core step, and the model is used to simulate and predict possible high-consequence events in the future.

[0075] According to step S5 of this embodiment: the purpose is real-time data input and report generation. The construction idea is to set fixed time intervals and specific conditions. When the time interval is reached or the preset conditions occur, the current real-time monitoring data is input into the basic assessment model to generate a real-time high consequence area report. This process is automated and relies on the model established in the previous steps and the real-time data flow, thus achieving efficient report writing.

[0076] According to step S6 of this embodiment: the purpose is to store the report and writing logic, and the construction idea is to store the generated high consequence area report and its writing process and logic on the server. The purpose of this is to ensure the traceability of the report and the security of the data, and to provide data support for subsequent analysis and improvement.

[0077] It can be understood that this embodiment achieves efficient integration of multi-source monitoring data and improves data processing efficiency; among them, an assessment model is constructed based on historical data and real-time data to achieve accurate risk assessment and early warning, and large models and automation technology are used to generate high-consequence area reports, which improves the efficiency and accuracy of report writing, and the reports and writing logic are stored on the server to ensure the digital management and traceability of the data.

[0078] Through the above implementation scheme and technical principles, the present invention can significantly improve the safety management level of high consequence areas and ensure that major adverse impacts on public health, safety, environment and society are minimized.

[0079] In some embodiments of the present application, the monitoring area is divided into functional areas, and different monitoring mechanisms are set for monitoring items in each functional area, including:

[0080] Determine the geographical location of the oil and gas pipeline and the area it passes through, set the first pipeline density as the first-level monitoring area A, set the second pipeline density as the second-level monitoring area B, and set the third pipeline density as the third-level monitoring area C, wherein the first pipeline density is greater than the second pipeline density, which is greater than the third pipeline density, and the higher the monitoring level, the smaller the interval between the upper threshold and the lower threshold for triggering notification in the monitoring of the monitoring project, wherein the first-level monitoring area A is monitored through sensor technology and manual inspections to generate real-time data streams, for the second-level monitoring area B, the population density within its risk range is evaluated to set the monitoring method, and for the third-level monitoring area C, the environmental sensitivity within its risk range is evaluated to set the monitoring method;

[0081] Functional area division is used to divide the monitoring area into several different functional areas according to the different characteristics, uses and risk levels of each area;

[0082] Setting different monitoring mechanisms for monitoring items in each functional area also includes setting a corresponding management unit for each functional area and a corresponding supervision strategy for category f transported through the pipeline in each management unit.

[0083] Specifically, this embodiment refines the functional area division and monitoring mechanism settings, including clarifying the specific location of the oil and gas pipeline and analyzing the different areas it passes through. This step ensures a comprehensive understanding of the pipeline, thereby providing a data basis for subsequent area division and monitoring mechanism formulation.

[0084] Specifically in the division of functional areas:

[0085] According to the density of pipelines, the monitoring area is divided into different levels of monitoring areas:

[0086] Level 1 monitoring area A: The pipeline density is the highest.

[0087] Secondary monitoring area B: medium pipeline density.

[0088] Level 3 monitoring area C: The pipeline density is the lowest.

[0089] The higher the monitoring level, the smaller the interval between the upper threshold and the lower threshold for triggering notification of the monitoring item, which means that the monitoring is more strict and sensitive.

[0090] Level 1 monitoring area A: Monitored by sensor technology and manual inspections, generating real-time data streams. Suitable for high-risk, high-density areas to ensure timely detection and response to potential problems.

[0091] Secondary monitoring area B: Set monitoring methods according to the population density within the risk range, pay attention to densely populated areas that may be more affected, and conduct corresponding monitoring and early warning.

[0092] Level 3 monitoring area C: The monitoring method is set according to the environmental sensitivity within the risk range, focusing on monitoring areas that may have an impact on the environment to protect environmental safety.

[0093] Among them, management unit refers to dividing the pipeline into different management units according to certain standards, which is conducive to the organization and standardization of pipeline management. It can be understood that each management unit is responsible for managing pipelines within a certain range, which can clarify responsibilities and facilitate management and supervision; and, according to the characteristics and needs of the pipeline, rationally allocate management resources to improve resource utilization efficiency; in addition, according to the different categories of pipeline transportation, formulate corresponding regulatory standards and measures; and, for each management unit, conduct risk assessment according to its environment, pipeline type and other factors, determine the regulatory focus and key control objects; and, for pipelines of different categories and different risk levels, take different regulatory measures, such as strengthening the installation of safety equipment, increasing the frequency of inspections, and strengthening data monitoring.

[0094] The supervision strategy refers to the formulation of corresponding emergency plans and disposal measures for possible emergencies to ensure timely response and handling when accidents occur, including regular inspection and evaluation of the supervision measures of each management unit, and timely adjustment and improvement of the supervision strategy to ensure the safe and stable operation of the pipeline.

[0095] It is understandable that accurate monitoring of oil and gas pipelines is achieved by dividing functional areas and implementing differentiated monitoring based on pipeline density and regional characteristics, and more stringent monitoring mechanisms are adopted in high-risk areas to ensure safety; different monitoring strategies are formulated according to regional risk levels and characteristics, which optimizes the allocation and use of resources, with more resources invested in high-risk areas and appropriate reductions in low-risk areas, thereby improving overall monitoring efficiency; based on the accumulation of real-time and historical data, the monitoring mechanism can be dynamically adjusted to flexibly respond to changes and emergencies in different regions, thereby improving emergency response capabilities; and by establishing management units and formulating detailed supervision strategies, full coverage of oil and gas pipeline transportation is achieved, ensuring that every link is under effective supervision.

[0096] In some embodiments of the present application, a monitoring method is set for evaluating the population density within the risk range of the secondary monitoring area B, including:

[0097] Obtain the outer diameter d of each pipeline involved in the secondary monitoring area B, where the unit of the outer diameter is millimeter, and the maximum allowable operating pressure P of the pipeline, where the unit of the operating pressure is MPa, and calculate the radius R of the affected area of ​​each pipeline leakage, where the radius R of the affected area is calculated by the following formula:

[0098] ;

[0099] According to the radius R of the affected area, the path H of each relevant pipeline in the secondary monitoring area B is obtained, and the risk range is set as each area covered by the radius R of the affected area with each point on the path H as the center point;

[0100] Obtain the population density M within the risk range, compare it with the preset first population density M1 and second population density M2, where M1 < M2, and set the monitoring method according to the comparison result;

[0101] When M<M1, remote sensing technology is used for monitoring to generate periodic data streams;

[0102] When M1≤M<M2, sensor technology is used for monitoring to generate real-time data streams;

[0103] When M2≤M, sensor technology plus manual inspections are used to generate real-time data streams.

[0104] It is understandable that the monitoring methods are graded according to the radius of the affected area and the population density, so as to achieve refined monitoring of areas with different risk levels; a combination of remote sensing technology, sensor technology and manual inspections is adopted according to different situations to ensure the real-time and accuracy of monitoring data; monitoring resources are flexibly deployed according to different population densities to improve resource utilization efficiency; monitoring methods are flexibly adjusted according to actual conditions to adapt to changes in different risk levels and improve the ability to respond to emergencies; by evaluating and monitoring the population density within the risk range, potential risks can be discovered in a timely manner and early warnings can be issued, which helps reduce the possibility of accidents.

[0105] In some embodiments of the present application, a monitoring method is set for evaluating the environmental sensitivity within the risk range of the third-level monitoring area C, including:

[0106] Obtain the environmental sensitivity factors involved in the third-level monitoring area C, and calculate the environmental sensitivity Q using the following formula:

[0107]

[0108] Where n is the number of environmental sensitive factors, is the weight of the xth indicator, is the score of the xth indicator;

[0109] Obtaining an environmental sensitivity Q, comparing it with a preset first environmental sensitivity Q1 and a second environmental sensitivity Q2, wherein Q1<Q2, and setting a monitoring method according to the comparison result;

[0110] When Q<Q1, remote sensing technology is used to monitor the third-level monitoring area in the first cycle to generate a periodic data stream;

[0111] When Q1≤Q<Q2, remote sensing technology is used to monitor the third-level monitoring area in the second period to generate a periodic data stream;

[0112] When Q2≤Q, remote sensing technology is used to monitor the third-level monitoring area in the third period to generate a periodic data stream;

[0113] The monitoring frequency of the first cycle is smaller than the monitoring frequency of the second cycle, which is smaller than the monitoring frequency of the third cycle.

[0114] Specifically, the calculation of environmental sensitivity Q involves multiple environmental sensitivity influencing factors, including but not limited to the population density of the water source downstream area in the third-level monitoring area and the ecological sensitivity in the third-level monitoring area. Among them, the ecological sensitivity assessment in the environmental sensitivity influencing factors is not limited to being obtained by the following methods:

[0115] Ecopath with Ecosim (ecological and environmental modeling software) and Habitat Suitability Index (HSI) models (habitat suitability index models), where Ecopath with Ecosim is an ecological modeling software that helps understand the complex relationships in ecology, including the flow of energy and nutrients; and Habitat Suitability Index (HSI) models assess the quality of habitat for specific species. These models consider various factors, such as food supply, refuge and other environmental conditions to determine the suitability of an area for a specific species.

[0116] CARLATOR (Professional Environmental Assessment Software) and Eco-informa (Ecological Information), where CARLATOR is a professional software tool for environmental impact assessment. It helps to assess the potential environmental impact of various activities or projects, while Eco-informa provides detailed information on environmental issues, including information and data on environmental issues, which helps to analyze environmental conditions and trends.

[0117] Each factor has a certain impact on the sensitivity of the environment, so these factors need to be considered comprehensively; by assigning weights to each factor, the different contributions of different factors to environmental sensitivity can be reflected, and the size of the weight can be determined based on expert opinions, historical data, scientific research and other factors; the score of each factor represents the performance or status of the factor in actual conditions, and the score can be determined based on specific data or evaluation results.

[0118] Specifically, regarding the comparison of environmental sensitivity Q, the calculated environmental sensitivity Q is compared with the preset first environmental sensitivity Q1 and second environmental sensitivity Q2 to determine the strictness of the monitoring method. The comparison result determines the frequency and period of monitoring, that is, different Q values ​​correspond to different monitoring strategies. This comparison can help determine the monitoring needs of environmentally sensitive areas, so as to take corresponding monitoring measures in a targeted manner.

[0119] Specifically, remote sensing technology is a technology for obtaining information about the earth's surface, which can collect data from a long distance (such as satellites or aircraft). This technology is widely used in environmental monitoring and can provide large-scale, multi-phase, and multi-resolution data, which helps to monitor environmental changes; under different environmental sensitivities, using remote sensing technology to monitor at different periods can reasonably allocate monitoring resources and improve monitoring efficiency based on the evaluation results of environmental sensitivity.

[0120] It is understandable that the setting of the monitoring cycle is based on the assessment of environmental sensitivity and the recognition of potential risks. Different environmental sensitivity areas may have different risk levels, so it is necessary to set the corresponding monitoring cycle according to the actual situation. The decreasing monitoring frequency (first cycle < second cycle < third cycle) means that the frequency of monitoring areas with low environmental sensitivity can be reduced, thereby optimizing resource allocation and improving monitoring efficiency. Through this method based on multi-factor comprehensive evaluation, environmental sensitivity can be evaluated more comprehensively and objectively, providing a scientific basis for environmental monitoring and risk management. At the same time, combined with remote sensing technology and reasonable monitoring cycle settings, the efficiency of environmental monitoring can be improved, and refined management and protection of environmentally sensitive areas can be achieved.

[0121] In some embodiments of the present application, converting the data stream generated by the monitoring mechanism into monitoring text includes:

[0122] The timestamp j and location stamp k of each piece of information are automatically captured in the monitoring text to associate events from different data sources and evaluate related events across regions or time.

[0123] The monitoring text also presets the regional monitoring text Xg and the unit monitoring text Xgi, where i is the number of related facilities in the area, g represents the type of monitoring data, X=A, B, C;

[0124] The unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, wherein the monitoring and screening is multi-dimensional monitoring, and the data of each dimension is screened. The multi-dimensional monitoring is set according to the pipeline transportation category f, and a significant threshold is set for each dimension, and non-significant data points that are less than the significant threshold within a unit time are eliminated, and the retained data are recorded as significant data points. The significant data points are output, and the multi-dimensional monitoring results in the relevant area are refined to generate the regional monitoring text Xg;

[0125] The unit monitoring text Xgi adopts a preset structured time-series text format and is configured with a distributed retrieval index. When a facility detection data point is not within the automatic generation range, after a high-consequence report is issued, the distributed retrieval index can still be used to actively search for other surrounding areas in the report.

[0126] Specifically, this embodiment provides an example of a specific implementation code, including:

[0127] class MonitoringData:

[0128] def __init__(self, timestamp, location, data):

[0129] self.timestamp = timestamp # timestamp

[0130] self.location = location# location stamp

[0131] self.data = data# Monitoring data

[0132] class UnitMonitoringText:

[0133] def __init__(self, region, facility_data):

[0134] self.region = region # Region number

[0135] self.facility_data = facility_data # Related facility detection data

[0136] def generate_unit_monitoring_text(self):

[0137] significant_data_points = self.filter_significant_data_points()# Filter significant data points

[0138] monitoring_text = f"Region: {self.region}\n"

[0139] for data_point in significant_data_points:

[0140] monitoring_text += f"Timestamp: {data_point.timestamp}, Location:{data_point.location}, Data: {data_point.data}\n"

[0141] return monitoring_text

[0142] def filter_significant_data_points(self):

[0143] significant_data_points = []

[0144] for data_point in self.facility_data:

[0145] # Perform multi-dimensional monitoring and screening

[0146] if self.is_significant_data(data_point):

[0147] significant_data_points.append(data_point)

[0148] return significant_data_points

[0149] def is_significant_data(self, data_point):

[0150] # Implement multi-dimensional monitoring and screening here

[0151] # For example, judging based on the pipeline transport category and the set significant threshold

[0152] # Return True if it is a significant data point, False if it is not a significant data point

[0153] pass

[0154] # Sample monitoring data

[0155] monitoring_data1 = MonitoringData("2024-05-31 08:00:00", "LocationA", {"Pressure": 10, "Temperature": 25})

[0156] monitoring_data2 = MonitoringData("2024-05-31 08:30:00", "LocationB", {"Pressure": 15, "Temperature": 30})

[0157] monitoring_data3 = MonitoringData("2024-05-31 09:00:00", "LocationC", {"Pressure": 8, "Temperature": 20})

[0158] # Example unit monitoring text generation

[0159] region = "Region A"

[0160] facility_data = [monitoring_data1, monitoring_data2, monitoring_data3]

[0161] unit_monitoring_text = UnitMonitoringText(region, facility_data)

[0162] print(unit_monitoring_text.generate_unit_monitoring_text())

[0163] In some embodiments of the present application, a historical high consequence area report is obtained for each type of functional area, and corresponding monitoring text when the historical high consequence area report is issued is extracted, and regional characteristics are analyzed according to the monitoring mechanism, and each regional feature is compared with the historical high consequence area report to filter keyword information, including:

[0164] The topic information in the historical high consequence area report is analyzed by using the topic model LDA, and the functional area and facility detection data points to which it belongs are matched. The words that overlap with the monitoring data type g of the corresponding facility point measurement point in the content text of the historical high consequence area report are extracted as keywords, and part-of-speech tagging is used for classification and integration to generate keyword information;

[0165] The evaluation weights are obtained based on the keyword information in the historical high-consequence area report, including:

[0166] Wherein, a basic weight is set for the category to which the part-of-speech tag of each keyword information in the classification integration belongs, and a first modified weight and a second modified weight are set to adjust the basic weight to generate an evaluation weight;

[0167] Extract chart information from historical high consequence area reports, obtain the field position and numerical text or photo number in the chart information, set different first correction weights according to the field position, set the second correction weight according to the size of the numerical text, and mark the photo number.

[0168] It can be understood that the historical high consequence area reports are analyzed through the topic model LDA, the topic information therein is extracted, and matched with the functional area and facility detection data points, so as to identify keywords that overlap with the monitoring data type; the part-of-speech tagging is used for classification and integration to generate keyword information, which can better understand the context and meaning of the keywords in the text; according to the keyword information in the historical high consequence area reports, combined with the category to which the part-of-speech tagging belongs, the basic weight is set, and the basic weight is adjusted through the first correction weight and the second correction weight to obtain the evaluation weight; such a weight acquisition method can adjust the weight according to the keyword information in the historical report, combined with professional knowledge and experience, and improve the accuracy and reliability of the weight; the chart information in the historical high consequence area report is analyzed, the field position and the numerical text or photo number are identified, and different first correction weights and second correction weights are set according to different field positions and numerical text sizes to evaluate the importance of key information; processing chart information in this way can more effectively utilize the digital information and picture information in the report, and provide a more comprehensive and accurate reference for the evaluation.

[0169] In some embodiments of the present application, constructing a mapping relationship based on the evaluation weights to generate a basic evaluation model for the functional area includes:

[0170] Obtain the historical high consequence area report for each functional area, perform text preprocessing on the content text in the historical high consequence area report, perform sentence segmentation on the preprocessed text, each sentence includes only one causal relationship, use dependency syntax to expand the sentence structure, extract the causal relationship words, construct features for the relationship words, use machine learning algorithms to train the mapping relationship to determine the functional area, and generate a verification set based on the facility detection data points in the content text determined by the photo numbers in the historical high consequence area report, verify the mapping relationship, generate a basic assessment model for each functional area, and compile the timestamp j and location stamp k of the information source into the basic assessment model.

[0171] It is understandable that through machine learning algorithm training, causal terms in historical high consequence area reports can be automatically identified and attributed to corresponding functional areas. This can realize the automatic classification of information sources and improve work efficiency. Using technical means such as dependency syntax analysis and feature extraction, causal terms can be accurately identified, and a basic assessment model with high accuracy can be constructed to accurately classify and locate information sources; generate a verification set to verify the basic assessment model and evaluate its classification accuracy in different functional areas, thereby improving the credibility and stability of the model; the basic assessment model compiles the time stamp and location stamp of the information source, which can be used for traceability analysis, tracking the source and flow of the information source, and providing a reliable basis for subsequent analysis and decision-making; through automated functional area classification and information source location, relevant information can be quickly and accurately processed and analyzed, providing timely and accurate reference for decision makers, improving decision efficiency and quality.

[0172] In some embodiments of the present application, whenever a unit time interval and a preset condition occur, the real-time monitoring data is input into the basic assessment model of the functional area to generate an E-area Z-case high consequence area report, including:

[0173] The preset conditions include that when the unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, the monitoring screening is multi-dimensional monitoring, and the data of each dimension is screened. The multi-dimensional monitoring is set according to the pipeline transportation category f, and an alarm threshold is set for each dimension, wherein the alarm threshold is greater than the significant threshold, and normal data points that are less than the alarm threshold within a unit time are eliminated, and the retained data are recorded as dangerous data points, and the dangerous data points are output to generate the corresponding E zone Z case high consequence area report.

[0174] It can be understood that the present embodiment monitors and screens the facility detection data points in each functional area according to preset conditions, so that the real-time monitoring data can be timely input into the basic assessment model to generate the corresponding E zone Z case high consequence area report; and, for the facility detection data points in each functional area, the system performs multi-dimensional monitoring and screening, sets them according to the pipeline transportation category f, and sets an alarm threshold for each dimension, so that the facility status can be monitored more comprehensively and dangerous data points can be accurately identified; in addition, the screened dangerous data points are output as information, and the corresponding E zone Z case high consequence area report is generated, so that potential safety risks can be discovered in time and corresponding measures can be taken to deal with them.

[0175] In some embodiments of the present application, the E-Z case high consequence area report and its writing logic are stored in the server, and the coding method of the E-Z case high consequence area report is that E is a matrix of numbers composed of Xgi-f, and Z is a single pipeline numbered sequentially along the oil and / or air flow direction;

[0176] The logic writing includes obtaining the time stamp j and location stamp k of each data acquisition and the monitoring method, and explaining the principles of the significant threshold and alarm threshold of each facility detection data point, and using a trend chart or line chart to represent the trend of the significant threshold and alarm threshold in unit time.

[0177] It is understandable that large-scale language models are often viewed as “black boxes” that lack explainability and transparency of generated content, affecting the credibility of reports and application effects. This may lead to distrust of model output results in high-risk areas (such as high-consequence area assessments). By retaining the writing logic, reporting black boxes can be avoided, and the internal decision-making process can be prevented from being difficult to explain, affecting the review.

[0178] In some embodiments of the present application, the E-Zone Z Case High Consequence Area Report and its writing logic are stored in the server, further comprising:

[0179] Directly send high consequence area reports to review server nodes, each of which hosts review expert resources;

[0180] Get the matrix of item E in the code of the high consequence area report. In Xgi-f, determine the technical field of the server address according to the transported category f, determine the professional field of the server address according to the monitored data type g, and obtain the nearest expert assigned to the location stamp k. Associate the years of work experience of the expert resource according to the level of the X monitoring area, where:

[0181] When X=A, only expert resources with more than 10 years of experience are associated;

[0182] When X=B, the expert resources with more than 6 years of experience are associated;

[0183] When X=C, and the number of relevant facilities in the region i≥3, the expert resources with more than 3 years of experience are associated;

[0184] When X=C, and when the number of relevant facilities in the region i<3, the expert resources with less than 3 years of experience are associated.

[0185] It is understandable that through intelligent allocation and optimal use of review expert resources, the system can improve the efficiency and quality of the review, ensure that each report can receive professional review and guidance, and provide timely and effective decision-making support for relevant departments and personnel.

[0186] Compared with the prior art, the beneficial effects of this embodiment are that through the division of functional areas and the setting of monitoring mechanisms, efficient integration of various monitoring data in high-consequence areas is achieved, and the technical difficulties of data flow integration are solved; based on historical high-consequence area reports and monitoring texts, in-depth analysis of regional characteristics is carried out to ensure the accuracy and reliability of the assessment model; by constructing a basic assessment model and inputting monitoring data in real time, dynamic risk assessment of high-consequence areas is achieved to ensure timely discovery and response to risks; using large models and data-driven methods, high-consequence area reports are automatically generated to improve the efficiency and accuracy of report writing; the generated reports and their writing logic are stored on the server to ensure the digital management and convenient retrieval of reports; through the above-mentioned technical means, the safety management level of high-consequence areas is improved to ensure that major adverse impacts on public health, safety, environment and society are minimized.

[0187] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0188] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0189] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0190] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0191] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for automatically writing high-consequence area-by-area-by-case reports based on a large model, characterized in that: include: Divide the monitoring area into functional areas, set different monitoring mechanisms for monitoring items in each functional area, and convert the data stream generated by the monitoring mechanism into monitoring text; Obtaining a historical high consequence area report for each type of functional area, correspondingly extracting the monitoring text when the historical high consequence area report was issued, and analyzing regional characteristics according to the monitoring mechanism; Compare each of the described regional characteristics with historical high consequence area reports to filter keyword information; Acquire assessment weights based on keyword information in the historical high consequence area report, and construct a mapping relationship based on the assessment weights to generate a basic assessment model for the functional area; Whenever the interval unit time and the preset conditions occur, the real-time monitoring data is input into the basic assessment model of the functional area to generate the E-area Z-case high consequence area report, where E is a matrix of Xgi-f, where f is the pipeline transportation category, i is the number of related facilities in the area, g represents the type of monitoring data, X=A, B, C, where A is the first-level monitoring area, B is the second-level monitoring area, and C is the third-level monitoring area; Z is a single pipeline numbered sequentially along the oil and / or air flow direction; The E-Zone Z high consequence area report and its writing logic are stored in the server; Obtain a historical high consequence area report for each type of functional area, extract the monitoring text when the historical high consequence area report was issued, analyze the regional characteristics according to the monitoring mechanism, and compare each of the regional characteristics with the historical high consequence area report to filter keyword information, including: The topic information in the historical high consequence area report is analyzed by using the topic model LDA, and the functional area and facility detection data points to which it belongs are matched. The words that overlap with the content text in the historical high consequence area report and the monitoring data type g of the corresponding facility point measurement point are extracted as keywords, and part-of-speech tagging is used for classification and integration to generate keyword information; The evaluation weight is obtained based on the keyword information in the historical high consequence area report, including: Wherein, a basic weight is set for the category to which the part-of-speech tag of each keyword information in the classification integration belongs, and a first modified weight and a second modified weight are set to adjust the basic weight to generate the evaluation weight; Extracting chart information from the historical high consequence area report, obtaining the field position and numerical text or photo number in the chart information, setting different first correction weights according to the field position, setting the second correction weight according to the size of the numerical text, and marking the photo number; Establishing a mapping relationship between the keyword information and the evaluation result through the evaluation weight of the keyword information; The step of constructing a mapping relationship according to the evaluation weights to generate a basic evaluation model for the functional area includes: Obtain a historical high consequence area report for each of the functional areas, perform text preprocessing on the content text in the historical high consequence area report, perform sentence segmentation on the preprocessed text, wherein each sentence only includes one causal relationship, use dependency syntax to expand the sentence structure, and extract causal relationship words therein, construct features for the relationship words, use a machine learning algorithm to train the mapping relationship to determine the functional area to which it belongs, and generate a verification set based on the facility detection data points in the content text determined by the photo number in the historical high consequence area report, verify the mapping relationship, generate a basic assessment model for each functional area, and compile the time stamp j and location stamp k of the information source into the basic assessment model.

2. The method for automatically writing a high-consequence area-by-area-by-case report based on a large model according to claim 1, characterized in that: The monitoring area is divided into functional areas, and different monitoring mechanisms are set for the monitoring items in each functional area, including: Determine the geographical location of the oil and gas pipeline and the area it passes through, set the first pipeline density as the first-level monitoring area A, set the second pipeline density as the second-level monitoring area B, and set the third pipeline density as the third-level monitoring area C, wherein the first pipeline density is greater than the second pipeline density, which is greater than the third pipeline density, and the higher the monitoring level, the smaller the interval between the upper threshold and the lower threshold for triggering notification in the monitoring of the monitoring project, wherein the first-level monitoring area A is monitored through sensor technology and manual inspections to generate real-time data streams, for the second-level monitoring area B, the population density within its risk range is evaluated to set the monitoring method, and for the third-level monitoring area C, the environmental sensitivity within its risk range is evaluated to set the monitoring method; The functional area division is used to divide the monitoring area into a number of different functional areas according to the different characteristics, uses and risk levels of each area; The setting of different monitoring mechanisms for the monitoring items of each functional area also includes setting a corresponding management unit for each functional area and a corresponding supervision strategy for the category f transported by the pipeline in each management unit.

3. The method for automatically writing a high-consequence area-by-area-case report based on a large model according to claim 2 is characterized in that: For the secondary monitoring area B, the population density within its risk range is assessed and monitoring methods are set, including: Obtain the outer diameter d of each pipeline involved in the secondary monitoring area B, where the unit of the outer diameter is millimeter, and the maximum allowable operating pressure P of the pipeline, where the unit of the operating pressure is MPa, and calculate the radius R of the affected area of ​​each pipeline leakage, wherein the radius R of the affected area is calculated by the following formula: ； According to the radius R of the affected area, the path H of each relevant pipeline in the secondary monitoring area B is obtained, and the risk range is set to each point on the path H as the center point and each area covered by the radius R of the affected area; Obtaining a population density M within the risk range, comparing it with a preset first population density M1 and a second population density M2, where M1<M2, and setting a monitoring method according to the comparison result; When M<M1, remote sensing technology is used for monitoring to generate periodic data streams; When M1≤M<M2, sensor technology is used for monitoring to generate real-time data streams; When M2≤M, sensor technology plus manual inspections are used to generate real-time data streams.

4. The method for automatically writing a high-consequence area-by-area-by-case report based on a large model according to claim 3 is characterized in that: For Level 3 monitoring area C, the environmental sensitivity within its risk range is assessed and monitoring methods are set, including: The environmental sensitivity influencing factors involved in the third-level monitoring area C are obtained, and the environmental sensitivity Q is calculated by the following formula: ； Where n is the number of environmental sensitive factors, is the weight of the xth indicator, is the score of the xth indicator, where the environmental sensitivity influencing factors include the population density in the downstream area of ​​the water source in the third-level monitoring area C; Obtain the environmental sensitivity Q, compare it with a preset first environmental sensitivity Q1 and a second environmental sensitivity Q2, wherein Q1<Q2, and set a monitoring method according to the comparison result; When Q<Q1, remote sensing technology is used to monitor the third-level monitoring area in the first cycle to generate a periodic data stream; When Q1≤Q<Q2, remote sensing technology is used to monitor the third-level monitoring area in the second period to generate a periodic data stream; When Q2≤Q, remote sensing technology is used to monitor the third-level monitoring area in the third period to generate a periodic data stream; The monitoring frequency of the first cycle is smaller than the monitoring frequency of the second cycle, which is smaller than the monitoring frequency of the third cycle.

5. The method for automatically writing a high-consequence area-by-area-case report based on a large model according to claim 4 is characterized in that: Converting the data stream generated by the monitoring mechanism into monitoring text includes: The timestamp j and location stamp k of each piece of information are automatically captured in the monitoring text to associate events from different data sources and evaluate related events across regions or across time; The monitoring text also presets a regional monitoring text Xg and a unit monitoring text Xgi, where i is the number of related facilities in the area, g represents the type of monitoring data, X=A, B, C; The unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, wherein the monitoring and screening is multi-dimensional monitoring, and the data of each dimension is screened, the multi-dimensional monitoring is set according to the pipeline transportation category f, and a significant threshold is set for each dimension, and non-significant data points that are less than the significant threshold within a unit time are eliminated, and the retained data are recorded as significant data points, and the significant data points are output as information, and the multi-dimensional monitoring results in the relevant area are refined to generate the regional monitoring text Xg; The unit monitoring text Xgi adopts a preset structured time-series text format and is configured with a distributed retrieval index, which is used to actively search other surrounding areas in the report through the distributed retrieval index after the high-consequence report is issued when a facility detection data point is not within the automatic generation range.

6. The method for automatically writing a high-consequence area-by-area-case report based on a large model according to claim 5, characterized in that: Whenever the unit time interval and the preset conditions occur, the real-time monitoring data is input into the basic assessment model of the functional area to generate the E-Zone Z Case High Consequence Area Report, including: The preset conditions include that when the unit monitoring text Xgi monitors and screens the facility detection data points in each functional area, the monitoring and screening is multi-dimensional monitoring, and the data of each dimension is screened, the multi-dimensional monitoring is set according to the pipeline transportation category f, and an alarm threshold is set for each dimension, wherein the alarm threshold is greater than the significant threshold, and normal data points that are less than the alarm threshold within a unit time are eliminated, and the retained data are recorded as dangerous data points, and the dangerous data points are output as information to generate the corresponding E zone Z case high consequence area report.

7. The method for automatically writing a high-consequence area-by-area-by-case report based on a large model according to claim 6, characterized in that: The E-Zone Z-Case High Consequence Area Report and its writing logic are stored in a server, wherein the coding method of the E-Zone Z-Case High Consequence Area Report is that E is a matrix of numbers composed of Xgi-f, and Z is a single pipeline numbered sequentially along the oil and / or air flow direction; The writing logic includes obtaining the timestamp j and location stamp k of each data acquisition and the monitoring method, and explaining the principles of the significant threshold and alarm threshold of each facility detection data point, and using a trend chart or a line chart to represent the trend of the significant threshold and alarm threshold in unit time.

8. The method for automatically writing a high-consequence area-by-area-by-case report based on a large model according to claim 7, characterized in that: The E-Zone Z-Case High Consequence Area Report and its writing logic are stored in the server, and further comprising: Directly sending the high consequence area report to review server nodes, each of which hosts review expert resources; Obtain the matrix of item E in the code of the high consequence area report, determine the technical field of the server address according to the transported category f in Xgi-f, determine the professional field of the server address according to the monitored data type g, obtain the expert with the closest distance to the location stamp k, and associate the years of work experience of the expert resource according to the level of the X monitoring area, where: When X=A, only expert resources with more than 10 years of experience are associated; When X=B, the expert resources with more than 6 years of experience are associated; When X=C, and the number of relevant facilities in the area i≥3, the expert resources with more than 3 years of experience are associated; When X=C, and when the number of relevant facilities in the region i<3, expert resources with less than 3 years of experience are associated.

Images