PAA-CUSUM-based station accident precursor identification method and related equipment

Abstract

The invention discloses a PAA-CUSUM-based station accident pre-mega identification method and related equipment, and belongs to the field of gas transmission station risk prevention and control and safety pre-warning. According to the method, first monitoring data of each station yard subunit in preset time are collected, then the first monitoring data are processed by adopting a PPA method, namely segmented aggregation approximation, to-be-analyzed data are processed by utilizing a CUSUM method, namely an accumulation sum method, and whether each station yard subunit has an abnormal condition or not is judged according to an accumulation sum result; second monitoring data of the station yard subunits with abnormal conditions in the preset sliding window are collected, and the accident precursor station yard subunits are identified by comparing the change degree of the second monitoring data; by adopting the method, the accident precursor can be identified, precautions are taken in the bud, the gas transmission station is guided to carry out effective risk prevention and control and safety early warning, and a most effective risk management strategy with the lowest cost is provided for station operators.

Description

Technical Field

[0001] The present invention belongs to the field of risk prevention and control and safety early warning of gas transmission stations, and specifically relates to a station accident precursor identification method based on PAA-CUSUM and related equipment. Background Art

[0002] For urban consumers along long-distance pipelines, natural gas stations are crucial for filtering and separating impurities, regulating flow and pressure, and distributing gas. As crucial urban infrastructure and a key component of natural gas transportation, accidents at these stations can have serious consequences.

[0003] Due to the complexity of gas transmission plant processes and the high flammability and explosiveness of the transported media, hazardous factors such as corrosion and misoperation can lead to equipment failures and even accidents. The severity of these accidents highlights the need for risk management within gas transmission plants. In the event of an accident, some gas transmission plants are located in urban or residential areas, making it more difficult to evacuate personnel and implement emergency measures to protect property. In existing engineering practices, the most common early warning method is combustible gas detectors. These devices rely solely on combustible gas concentration probes and flame detection probes within the plant to provide accident warnings. This approach has the following drawbacks: First, sensors can only detect abnormalities after an incident such as a gas leak or combustion has occurred. Second, if sensors fail and an incident is not promptly identified and emergency measures are not implemented, a minor incident can easily escalate into a major one. This approach can only provide early warning for indoor leaks, but its detection capabilities for outdoor leaks and non-leakage faults (such as blockages) are significantly insufficient. Furthermore, existing natural gas plant system risk analysis models are overly complex, requiring only periodic risk assessments and updated calculations based on safety system engineering methods. However, this approach struggles to identify potential risks in a timely manner. Therefore, it is necessary to further improve the station system's ability to detect potential risks in real time. Summary of the Invention

[0004] In order to overcome the shortcomings of the above-mentioned technology, the present invention provides a station accident precursor identification method and related equipment based on PAA-CUSUM, which can solve the technical problem that the existing gas transmission station early warning measures cannot detect potential risks in a timely manner.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A station accident precursor identification method based on PAA-CUSUM, comprising:

[0007] S1: Collect the first monitoring data of each station sub-unit within a preset time;

[0008] S2: performing segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed;

[0009] S3: Based on the CUSUM method, the data to be analyzed are accumulated and processed, and the accumulation and sum results are used to determine whether there are any abnormalities in each station sub-unit;

[0010] S4: collecting second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window;

[0011] S5: Identify the station sub-unit with the largest degree of change in the second monitoring data as an accident precursor station sub-unit.

[0012] Furthermore, in S1, the gas transmission station system is divided into regions according to functions to obtain multiple station sub-units.

[0013] Furthermore, in S4, the length of the preset sliding window is set to 24 hours.

[0014] Furthermore, in S3, by comparing each cumulative sum result with a preset threshold, it is determined whether there is an abnormality in each station sub-unit; if the cumulative sum result exceeds the preset threshold, it is determined that there is an abnormality and the next step is continued; otherwise, it is determined that there is no abnormality and the process returns to S1.

[0015] Furthermore, in S4, the second monitoring data of the station sub-unit with abnormal conditions within the preset sliding window are plotted into a PAA-CUSUM control chart, and the station sub-unit with the largest degree of change is determined by comparing the PAA-CUSUM control charts.

[0016] Furthermore, the first monitoring data and the second monitoring data are pressure data and / or flow data.

[0017] Furthermore, in S3, by comparing each cumulative sum result with the abnormality standard, it is determined whether there is an abnormality in each station sub-unit; wherein, the abnormality standard includes: first, the cumulative sum result has a downward or upward trend; second, the cumulative sum result exceeds the preset alarm threshold; if the cumulative sum result meets the above two conditions at the same time, it is determined that the corresponding station sub-unit has an abnormality; otherwise, it is determined that there is no abnormality.

[0018] A PAA-CUSUM-based station accident precursor identification system is used to implement the steps of the PAA-CUSUM-based station accident precursor identification method, including:

[0019] The first data acquisition module is used to collect the first monitoring data of each station sub-unit within a preset time;

[0020] A first data processing module is used to perform segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed;

[0021] The second data processing module is used to accumulate and process the data to be analyzed based on the CUSUM method, and judge whether there is any abnormality in each station sub-unit through the accumulation and results;

[0022] A second data acquisition module is used to collect second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window;

[0023] The data identification module is used to identify the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

[0024] A device comprising:

[0025] memory for storing computer programs;

[0026] The processor is configured to implement the steps of the above-mentioned method for identifying station accident precursors based on PAA-CUSUM when executing the computer program.

[0027] A computer-readable storage medium stores a computer program, which, when executed by a processor, is used to implement the steps of the above-mentioned method for identifying station accident precursors based on PAA-CUSUM.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present invention provides a station accident precursor identification method based on PAA-CUSUM. The method first collects first monitoring data of each station subunit within a preset time, then uses the PPA method, i.e., segmented aggregation approximation, to process the first monitoring data, and then uses the CUSUM method, i.e., cumulative sum method, to process the data to be analyzed. The cumulative sum result is used to determine whether each station subunit has an abnormality, and then collects second monitoring data of the station subunit with the abnormality within a preset sliding window. By comparing the degree of change in the second monitoring data, the station subunit with the accident precursor is identified. Compared with traditional gas transmission station early warning measures, the method identifies process anomalies through monitoring data, analyzes events with the highest potential risks and propagation patterns, and can better discover potential risks in real time. In the method, the PPA method is used to ensure the validity of the monitoring data, and the CUSUM method is used to identify small trend changes, thereby improving the identification accuracy and efficiency. The method can identify accident precursors, prevent accidents before they occur, and guide gas transmission stations to carry out effective risk prevention and control and safety early warning, providing the station operators with the lowest cost and most effective risk management strategy.

[0030] Preferably, in the present invention, the gas transmission station system is divided into a plurality of station sub-units according to functional areas, so as to facilitate real-time monitoring of each unit of the gas transmission station system and ensure the reliability of monitoring.

[0031] Preferably, in the present invention, the length of the preset sliding window is set to 24 hours. Setting a 24-hour sliding window reduces time complexity and improves operating speed and monitoring accuracy.

[0032] Preferably, in the present invention, by drawing a PAA-CUSUM control chart for the cumulative sum results of abnormal situations and then comparing the PAA-CUSUM control charts, the station sub-unit with the largest degree of change can be quickly and accurately identified, and the accident precursor station sub-unit is determined, which greatly improves the recognition accuracy and efficiency.

[0033] Preferably, in the present invention, an abnormal standard is provided for judging whether there is an abnormality in the cumulative sum result, which is equivalent to providing double insurance. Only when the cumulative sum result satisfies the above two conditions of a downward or upward trend and exceeds the preset alarm threshold, it is judged that there is an abnormality in the cumulative sum result. This can reduce the system's misjudgment rate and missed judgment rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a gas transmission station system divided by functional areas according to an embodiment of the present invention;

[0035] Figure 2 The PAA-CUSUM control chart corresponding to the pressure data of outlet 1# provided in the embodiment of the present invention;

[0036] Figure 3 The PAA-CUSUM control diagram of an abnormal object provided by an embodiment of the present invention, wherein (a) is a ball valve; (b) is a pressure regulating valve;

[0037] Figure 4 A flow chart of a method for identifying station accident precursors based on PAA-CUSUM provided in an embodiment of the present invention;

[0038] Figure 5 A flow chart of a station accident precursor identification method based on PAA-CUSUM provided by the present invention;

[0039] Figure 6 This is a structural diagram of a station accident precursor identification system based on PAA-CUSUM provided by the present invention.

[0040] Reference numerals:

[0041] 1-Gas transmission pipeline; 2-Emergency shut-off valve; 3-Filter separator; 4-Ball valve; 5-Drain valve; 6-Gas collecting pipe; 7-Pressure regulating valve; 8-Flow meter; 9-Drainage tank; 10-Drainage pipe. DETAILED DESCRIPTION

[0042] The present invention provides a method for identifying station accident precursors based on PAA-CUSUM, such as Figure 5 As shown, the following steps are included:

[0043] S1: Collect the first monitoring data of each station sub-unit within a preset time;

[0044] S2: performing segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed;

[0045] S3: Based on the CUSUM method, the data to be analyzed are accumulated and processed, and the accumulation and sum results are used to determine whether there are any abnormalities in each station sub-unit;

[0046] S4: collecting second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window;

[0047] S5: Identify the station sub-unit with the largest degree of change in the second monitoring data as an accident precursor station sub-unit.

[0048] Among them, more specific:

[0049] In S1, the gas transmission station system is divided into regions according to function to obtain multiple station sub-units.

[0050] In S3, by comparing each cumulative sum result with the preset threshold, it is determined whether there is an abnormality in each station sub-unit; if the cumulative sum result exceeds the preset threshold, it is determined that there is an abnormality and the next step is continued; otherwise, it is determined that there is no abnormality and the process returns to S1.

[0051] You can also use:

[0052] By comparing each cumulative sum result with the abnormality standard, it is determined whether there is any abnormality in each station sub-unit; wherein, the abnormality standard includes: first, the cumulative sum result has a downward or upward trend; second, the cumulative sum result exceeds the preset alarm threshold; if the cumulative sum result meets the above two conditions at the same time, it is determined that the corresponding station sub-unit has an abnormality; otherwise, it is determined that there is no abnormality.

[0053] In S4, the length of the preset sliding window is set to 24 hours.

[0054] The second monitoring data of the station sub-unit with abnormal conditions within the preset sliding window are plotted into a PAA-CUSUM control chart, and the station sub-unit with the largest degree of change is determined by comparing the PAA-CUSUM control charts.

[0055] The first monitoring data and the second monitoring data are pressure data and / or flow data.

[0056] like Figure 6 As shown, the present invention also provides a station accident precursor identification system based on PAA-CUSUM, including: a first data acquisition module, used to collect first monitoring data of each station sub-unit within a preset time; a first data processing module, used to perform segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; a second data processing module, used to accumulate and process the data to be analyzed based on the CUSUM method, and judge whether there is an abnormality in each station sub-unit through the accumulation and result; a second data acquisition module, used to collect second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window; a data identification module, used to identify the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

[0057] The present invention also provides a device, comprising: a memory for storing a computer program; and a processor for implementing the steps of the PAA-CUSUM-based station accident precursor identification method when executing the computer program.

[0058] When the processor executes the computer program, the above-mentioned steps of identifying station accident precursors based on PAA-CUSUM are implemented, for example: collecting first monitoring data of each station sub-unit within a preset time; performing segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; accumulating and processing the data to be analyzed based on the CUSUM method, and judging whether each station sub-unit has an abnormal situation through the accumulation and result; collecting second monitoring data of the station sub-unit with an abnormal situation within a preset sliding window; and identifying the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

[0059] Alternatively, when the processor executes the computer program, the functions of each module in the above-mentioned system are realized, for example: a first data acquisition module is used to collect the first monitoring data of each station sub-unit within a preset time; a first data processing module is used to perform segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; a second data processing module is used to accumulate and process the data to be analyzed based on the CUSUM method, and judge whether there is an abnormality in each station sub-unit through the accumulation and results; a second data acquisition module is used to collect the second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window; a data identification module is used to identify the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

[0060] Exemplarily, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of completing preset functions, and the instruction segments are used to describe the execution process of the computer program in the PAA-CUSUM-based station accident precursor identification device. For example, the computer program can be divided into a first data acquisition module, a first data processing module, a second data processing module, a second data acquisition module, and a data identification module. The first data acquisition module is used to collect first monitoring data from each station sub-unit within a preset time; the first data processing module is used to perform segmented aggregation and approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; the second data processing module is used to accumulate and process the data to be analyzed based on the CUSUM method, and determine whether each station sub-unit has an abnormality based on the accumulated and summed results; the second data acquisition module is used to collect second monitoring data from the station sub-unit with an abnormality within a preset sliding window; and the data identification module is used to identify the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

[0061] The PAA-CUSUM-based station accident precursor identification device can be a computing device such as a desktop computer, laptop, PDA, or cloud server. The PAA-CUSUM-based station accident precursor identification device can include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above examples of PAA-CUSUM-based station accident precursor identification devices do not constitute a limitation on the PAA-CUSUM-based station accident precursor identification device. The device can include more components than those described above, or a combination of certain components, or different components. For example, the PAA-CUSUM-based station accident precursor identification device can also include input and output devices, network access devices, buses, etc.

[0062] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc. The processor is the control center of the PAA-CUSUM-based station accident precursor identification system, and utilizes various interfaces and lines to connect various parts of the entire PAA-CUSUM-based station accident precursor identification system.

[0063] The memory can be used to store the computer program and / or module, and the processor implements various functions of the station accident precursor identification device based on PAA-CUSUM by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory.

[0064] The memory may mainly include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area may store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0065] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of the method for identifying station accident precursors based on PAA-CUSUM.

[0066] If the integrated modules / units of the PAA-CUSUM-based station accident precursor identification system are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0067] Based on this understanding, the present invention implements all or part of the process of the aforementioned PAA-CUSUM-based station accident precursor identification method, and can also be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the aforementioned PAA-CUSUM-based station accident precursor identification method. The computer program includes computer program code, which can be in source code form, object code form, executable file, or a pre-set intermediate form.

[0068] The computer-readable storage medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0069] It should be noted that the content contained in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable storage media do not include electrical carrier signals and telecommunication signals.

[0070] The present invention will be further described below in conjunction with the embodiments and drawings:

[0071] Example

[0072] In order to solve the problem mentioned in the background technology: In existing engineering practices, the most common early warning method is the combustible gas detector, which only carries out accident early warning based on the combustible gas concentration probe, flame identification probe, etc. in the station. This method has the following disadvantages: First, the abnormal information can only be identified by the sensor after an accident such as combustible gas leakage or combustion occurs; Second, once the sensor fails, the accident that has occurred is not identified in time and emergency measures are not taken, and a minor accident can easily evolve into a major accident. It can be seen that the above method can only provide early warning for indoor leaks, while the detection capability for outdoor leaks and non-leakage faults (such as blockages) is obviously insufficient. At the same time, the existing natural gas station system risk analysis model is too complicated, and it can only carry out risk assessment and calculate and update station risks based on the safety system engineering method on a regular basis. However, it is difficult to discover its potential risks in a timely manner using this method.

[0073] To address the above issues, this embodiment provides a method for identifying station accident precursors based on PAA-CUSUM. This method identifies process anomalies through monitoring data and analyzes events with the highest potential risks and their propagation patterns. This method helps station staff identify potential hazards and provides a basis for daily risk mitigation.

[0074] like Figure 4 As shown, this embodiment provides a method for identifying station accident precursors based on PAA-CUSUM, and the specific steps are as follows:

[0075] Step 1: Implement the sliding window method and set the sliding window width to w = 24 hours. After finding the accident precursor, the PAA-CUSUM algorithm can be reset and a new cycle can be performed.

[0076] In this step, the gas transmission station system is divided into multiple station sub-units according to functional areas in advance.

[0077] Step 2: Input the monitoring data of the SCADA system. Here, sampling is performed once every minute. The monitoring data uses pressure data. Of course, flow data can also be collected. In this embodiment, pressure data is used as an example.

[0078] Step 3: The PAA algorithm uses the mean of the data elements to represent the sequence segment. The PAA algorithm is used to preprocess the pressure data at the system outlet to obtain the data to be analyzed. This algorithm aims to divide the time series into segments of equal length and calculate the mean of each segment as a feature representation of the time series.

[0079] Suppose a time series data X={x1, x2, ..., x n}, such as pressure or temperature data. Represent it as a vector of length N The i-th element is defined as:

[0080]

[0081] The PAA method reflects the overall trend of the original data and forms a low-dimensional data series characterized by the mean. Furthermore, by reducing the time series from n-dimensional space to N-dimensional space, the efficiency of time series anomaly detection can be improved, the impact of noise can be reduced, and the efficiency of subsequent data processing can be improved.

[0082] Step 4: For the data to be analyzed after the third step, use the CUSUM algorithm to calculate the cumulative value and boundary value of the lower limit and upper limit. The specific formula is as follows:

[0083]

[0084]

[0085]

[0086]

[0087] Where i represents the i-th element after preprocessing by PAA algorithm, CL i Indicates the lower accumulated value, CU i represents the upper cumulative value, UCL and LCL are the upper and lower boundary values, respectively. T is the target value, k is the displacement size to be detected, h is the standardized decision interval, σ is the control variable, and m is the number of subgroups.

[0088] Step 4: Determine whether there are any abnormalities in the station sub-units. Here, the cumulative sum of the above-obtained station sub-units can be compared with the preset threshold. If it exceeds the threshold, it is determined that there is an abnormality; otherwise, there is no abnormality.

[0089] In this step, as another preferred real-time method of the present technical solution, in order to reduce the system's false positive rate and missed positive rate, an abnormality standard is provided for determining whether an abnormal situation exists, specifically: (1) when there is a downward or upward trend (some consecutive points increase or decrease); (2) when the alarm threshold set according to the ISO standard (ISO 7870-4) is exceeded. In order to be more in line with the actual situation, we choose m = 1, k = 0.5, h = 3.5; if the cumulative sum result obtained by processing meets both of the above conditions, the system is judged to be abnormal and the monitored station sub-unit is in an abnormal state.

[0090] Step 5: If there is an exception, go to step 6; if there is no exception, return to step 2.

[0091] Step 6: In the case of abnormal conditions, retrieve (collect) the monitoring data of each sub-object (i.e. each component to be monitored) in each sub-unit of each station within 24 hours before the alarm time. i and CU i Perform accumulation and processing, then overlay the images, and draw the corresponding PAA-CUSUM control chart based on which change trend is more obvious.

[0092] Step 7: Based on the operations in step 6, find the abnormal objects and calculate the degree of change of their PAA-CUSUM control chart according to the following formula.

[0093]

[0094]

[0095] Among them, {x i} is the dataset before the change point, {xj} is the data set after the change point, and they all follow Gaussian distribution. u It means that {x i}, the probability of the trend rising after the change. Similarly, P d is the probability of a downward trend. L represents the set {x j To calculate the upward change u and downward change d more intuitively, the formula can be converted to formulas (8)-(9).

[0096]

[0097]

[0098] Step 8: Find the object with the greatest degree of change and determine the accident precursor station sub-unit or sub-object.

[0099] Here, it needs to be explained that each station sub-unit also includes a number of sub-objects, and the sub-objects are the components to be monitored.

[0100] It should be noted that if the PAA-CUSUM control chart trend continues to decline, it indicates that there may be a leak in the gas transmission station system. In engineering practice, a handheld combustible gas concentration detector can be used for auxiliary diagnosis to improve accuracy.

[0101] The following is a set of examples to verify and analyze the performance of the recognition method provided by this embodiment, as follows:

[0102] Historical fault records from gas transmission stations were used to evaluate the performance of the proposed model. To better illustrate the method's ability to sensitively detect potential system risks, we analyzed data from 0 to 24 hours prior to the fault. Precursor fault identification at gas transmission stations can be divided into three steps.

[0103] First, the method detects outlet pressure anomalies over a 24-hour period. Then, by calculating the degree of change in indicator readings for each object area (station subunit), the object with the highest potential risk, where the degree of change is the most significant, can be identified.

[0104] Next, determine whether the abnormal indicator reading shows a reading that is lower or higher than usual. For example, if it shows a downward trend, the status of the stress node can be set to "low" and the Δp value can be calculated. If the Δp of the component failure node in the acquired object is significantly higher than that of other nodes, then this component has the highest potential risk. Based on the Δp, a ranked list of suspicious objects can also be provided, if they are less critical. Finally, the component status node is checked, where the state with the highest Δp is the possible failure state of the component.

[0105] The data has been tested in the first stage and no out-of-control has occurred, so we can capture very small shifts directly by drawing the PAA-CUSUM control chart.

[0106] like Figure 1 As shown in the figure, a gas transmission station system is used as an example. This system includes a gas transmission pipeline 1 (GTP), an emergency shut-off valve 2 (ESDV), a filter separator 3 (FS), a ball valve 4 (BV), a drain valve 5 (DV), a gas collecting pipe 6 (GCP), a pressure regulating valve 7 (PRV), a flow meter 8 (FG), a drain tank 9, and a drain pipe 10 (DP). In the figure, P represents a pressure indicator; F represents a flow indicator; T represents a temperature indicator; and L represents a liquid level indicator. In this system, impurities such as water and hydrogen sulfide in the gas cause corrosion and leakage in the pipeline between ball valve 4 and pressure regulating valve 7, a subunit of the station subunit.

[0107] like Figure 2 As shown, the PAA-CUSUM station accident precursor identification method provided by this embodiment detects that the outlet #1 pressure is continuously decreasing, indicating that a certain position in the system is abnormal. Therefore, the relevant monitoring data of each subunit is retrieved for PAA-CUSUM analysis, and the above formulas (8)-(9) are used to calculate the degree of change. Figure 3 As shown in Figures (a) and (b), respectively, the indicator readings for ball valve 4 and pressure regulating valve 7 based on the PAA-CUSUM algorithm are shown. After CUSUM analysis, the other indicator readings are 0 or fluctuate around 0, indicating that the target area is risk-free. The calculated degrees of change for objects 4 and 7 are 2.21 and 1.27, respectively. Therefore, ball valve 4 has the highest potential risk. It can be seen that this method overcomes the problem that traditional methods can only detect accident precursors through combustible gas leaks, providing a more reliable method for identifying station accident precursors. Based on the structure and process flow of the gas transmission station system, an accident precursor identification algorithm based on the PAA-CUSUM algorithm was established. The PAA algorithm performs well in data dimensionality reduction and denoising, while the CUSUM algorithm performs well in identifying subtle trend changes in data. The analysis results can provide a ranking of the potential risks of each functional system in the gas transmission station from high to low, thereby guiding the gas transmission station to carry out effective risk prevention and control and safety warnings.

[0108] In summary, the present invention provides a method for identifying station accident precursors based on PAA-CUSUM. The method first collects the first monitoring data of each station sub-unit within a preset time, then uses the PPA method, i.e., segmented aggregation approximation, to process the first monitoring data, and then uses the CUSUM method, i.e., cumulative sum method, to process the data to be analyzed, and judges whether there is an abnormality in each station sub-unit through the cumulative sum result, and then collects the second monitoring data of the station sub-unit with the abnormality within a preset sliding window, and identifies the station sub-unit with accident precursor by comparing the degree of change of the second monitoring data; compared with traditional gas transmission station early warning measures, the method identifies process anomalies through monitoring data, analyzes events with the highest potential risks and propagation patterns, and can better discover potential risks in real time; in the method, the PPA method is used to ensure the validity of the monitoring data, and the CUSUM method is used to identify small trend changes, thereby improving the recognition accuracy and efficiency; the method can identify accident precursors, prevent accidents before they happen, and guide gas transmission stations to carry out effective risk prevention and control and safety early warning, providing station operators with the lowest cost and most effective risk management strategy.

[0109] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention.

Claims

1. A station accident precursor identification method based on PAA-CUSUM, characterized in that: include: S1: Collect the first monitoring data of each station sub-unit within a preset time; S2: performing segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; S3: Based on the CUSUM method, the data to be analyzed are accumulated and processed, and the accumulation and sum results are used to determine whether there are any abnormalities in each station sub-unit; S4: collecting second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window; S5: Identify the station sub-unit with the largest degree of change in the second monitoring data as an accident precursor station sub-unit.

2. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: In S1, the gas transmission station system is divided into regions according to its functions to obtain multiple station sub-units.

3. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: In S4, the length of the preset sliding window is set to 24 hours.

4. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: In S3, by comparing each cumulative sum result with the preset threshold, it is determined whether there is an abnormality in each station sub-unit; if the cumulative sum result exceeds the preset threshold, it is determined that there is an abnormality and the next step is continued; otherwise, it is determined that there is no abnormality and the process returns to S1.

5. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: In S4, the second monitoring data of the station sub-unit with abnormal conditions within the preset sliding window are plotted into a PAA-CUSUM control chart, and the station sub-unit with the largest degree of change is determined by comparing the PAA-CUSUM control charts.

6. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: The first monitoring data and the second monitoring data are pressure data and / or flow data.

7. The method for identifying station accident precursors based on PAA-CUSUM according to claim 1, characterized in that: In S3, by comparing each cumulative sum result with the abnormality standard, it is determined whether each station sub-unit has an abnormal situation; wherein, the abnormality standard includes: first, the cumulative sum result has a downward or upward trend; second, the cumulative sum result exceeds the preset alarm threshold; if the cumulative sum result meets the above two conditions at the same time, it is determined that the corresponding station sub-unit has an abnormal situation; otherwise, it is determined that there is no abnormal situation.

8. A PAA-CUSUM-based station accident precursor identification system, used to implement the steps of the PAA-CUSUM-based station accident precursor identification method according to any one of claims 1 to 7, characterized in that: include: The first data acquisition module is used to collect the first monitoring data of each station sub-unit within a preset time; A first data processing module is used to perform segmented aggregation approximation processing on the first monitoring data based on the PPA method to obtain data to be analyzed; The second data processing module is used to accumulate and process the data to be analyzed based on the CUSUM method, and judge whether there is any abnormality in each station sub-unit through the accumulation and results; A second data acquisition module is used to collect second monitoring data of the station sub-unit with abnormal conditions within a preset sliding window; The data identification module is used to identify the station sub-unit with the largest degree of change in the second monitoring data as the accident precursor station sub-unit.

9. A device, characterized in that include: memory for storing computer programs; A processor, configured to implement the steps of the method for identifying station accident precursors based on PAA-CUSUM according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the station accident precursor identification method based on PAA-CUSUM according to any one of claims 1 to 7.

Images