Pipeline risk monitoring system and method

Abstract

The application relates to the technical field of pipeline monitoring, and discloses a pipeline risk monitoring system and method. The system comprises a collection module and a processing module; the collection module determines a collection operation mode of a monitoring point based on a temperature influence coefficient; the processing module is configured to establish a time-parameter function based on parameters collected by a plurality of operation parameter sensors at the same time according to the collection operation mode, determine a risk level of the monitoring point according to a slope of the time-parameter function and a parameter change length, and determine a risk prediction value of the monitoring point based on a parameter set of each operation parameter sensor in the plurality of operation parameter sensors and a risk evaluation value of the monitoring point according to the risk prediction value and a risk adjustment coefficient; and the processing module is further configured to determine a pipeline risk monitoring result of the monitoring point according to the risk level and the risk evaluation value. The application can obtain a plurality of parameters according to dynamic adjustment of a collection operation mode, and has good monitoring effect.

Description

Technical Field

[0001] This invention relates to the field of pipeline monitoring technology, and more specifically, to a pipeline risk monitoring system and method. Background Technology

[0002] With the development of the oil and natural gas energy industry, requirements for pipeline safety have been raised. During pipeline transportation, the ambient temperature indirectly affects various parameters such as humidity, pressure, and flow rate. Existing monitoring systems use sensors that accumulate errors under different temperature conditions, making it impossible to flexibly adjust monitoring strategies according to temperature changes. This results in inaccurate parameters and affects pipeline monitoring. Secondly, traditional monitoring systems typically only monitor single physical parameters and cannot perform comprehensive analysis of multiple parameters. Due to the lack of comprehensive evaluation of multiple parameters, existing systems struggle to make accurate monitoring in complex environments. Summary of the Invention

[0003] In view of this, the present invention proposes a pipeline risk monitoring system and method, which aims to solve the problems of the inability to flexibly adjust the monitoring strategy according to temperature changes and the lack of comprehensive evaluation of multiple parameters.

[0004] In a first aspect, the present invention proposes a pipeline risk monitoring system, comprising: a data acquisition module and a processing module;

[0005] The acquisition module is configured to acquire temperature parameters from multiple temperature sensors evenly distributed at monitoring points, determine a temperature influence coefficient based on the temperature parameters of the multiple temperature sensors, and determine the acquisition operation mode of the monitoring points based on the temperature influence coefficient.

[0006] The processing module is configured to establish a time-parameter function based on the parameters collected by multiple operating parameter sensors at the same time while operating in the acquisition mode, and determine the risk level of the monitoring point based on the slope of the time-parameter function and the length of the parameter change.

[0007] The processing module is also configured to determine the risk prediction value of the monitoring point based on the parameter set of each of the multiple operating parameter sensors, and to determine the risk adjustment coefficient based on the statistical results of the risk assessment indicators in the parameter set of each operating parameter sensor. Based on the risk prediction value and the risk adjustment coefficient, the risk assessment value of the monitoring point is determined. The parameter set includes the parameters collected by the operating parameter sensors at different times. The risk assessment value and the risk adjustment coefficient are positively correlated.

[0008] The processing module is also configured to monitor pipeline risk results based on risk level and risk assessment value as monitoring points.

[0009] Alternatively, the temperature influence coefficient can be derived from the following formula:

[0010]

[0011] Where T represents the temperature influence coefficient, t i This represents the temperature parameter of any single temperature sensor, and n represents the number of temperature sensors.

[0012] The acquisition module is specifically configured as follows: when T is less than 25, the acquisition operation mode of the monitoring point is determined to be the first-level mode; when T is greater than or equal to 25 and less than 50, the acquisition operation mode of the monitoring point is determined to be the second-level mode; when T is greater than or equal to 50, the acquisition operation mode of the monitoring point is determined to be the third-level mode; the working sensitivity and acquisition frequency of the first-level mode, the second-level mode and the third-level mode decrease in sequence.

[0013] Optionally, the system further includes: a verification module; the verification module is configured to determine a historical operation verification set with the same acquisition operation mode as the monitoring point based on historical acquisition operation mode data, extract historical operation features from the historical operation verification set, the historical operation features including historical working sensitivity and historical acquisition frequency, and determine the operation similarity based on the historical working sensitivity and historical acquisition frequency, the operation similarity being derived by the following formula:

[0014]

[0015] Where f0 represents the acquisition frequency corresponding to the acquisition operation mode of the monitoring point, f1 represents the historical acquisition frequency, q0 represents the working sensitivity corresponding to the acquisition operation mode of the monitoring point, and q1 represents the historical working sensitivity.

[0016] When A is greater than or equal to the operation similarity threshold, the data collection and operation mode of the monitoring point is judged to be accurate; when A is less than the operation similarity threshold, the data collection and operation mode of the monitoring point is judged to be inaccurate.

[0017] Optionally, the verification module is also configured to, when the judgment of the monitoring point's collection operation mode is inaccurate, obtain representative data for each collection operation mode based on historical collection operation mode data, combine the representative data and the operation characteristics of the monitoring point to establish an aggregated dataset, obtain the feature vector of each data in the aggregated dataset, determine the expected number of clusters k as 3, initialize the parameters of the Gaussian distribution, calculate the probability that each data in the aggregated dataset belongs to each Gaussian distribution, obtain the responsibility value, and select the cluster with the largest responsibility value as the collection operation mode of the monitoring point.

[0018] Optionally, the operating parameter sensors are wireless passive sensors, and the multiple operating parameter sensors include a humidity sensor, a flow sensor, and a pressure sensor;

[0019] The processing module is specifically configured to: set the acquisition operation mode of the monitoring point to the working mode of humidity sensor, flow sensor and pressure sensor, acquire the humidity parameter of humidity sensor, the flow parameter of flow sensor and the pressure parameter of pressure sensor at the same time; establish the y-axis of time-parameter function with the unit quantity of humidity parameter, the unit quantity of flow parameter and pressure parameter and the unit quantity of pressure parameter, establish the x-axis of time-parameter function with time sequence, and substitute all humidity parameter, flow parameter and pressure parameter at the same time to obtain time-parameter function.

[0020] Optionally, the processing module is specifically configured to: connect all adjacent humidity parameters with straight lines, connect all adjacent flow parameters with straight lines, and connect all adjacent pressure parameters with straight lines; count the number of all straight lines and the number of lines with a slope greater than 0; obtain the sum of the lengths of the changes in humidity parameters with a slope greater than 0, the sum of the lengths of the changes in flow parameters with a slope greater than 0, and the sum of the lengths of the changes in pressure parameters with a slope greater than 0; obtain the sum of the lengths of all changes; and determine the ratio coefficient, which is derived from the following formula:

[0021]

[0022] Where B represents the ratio coefficient, Q a W represents the number of lines with a slope greater than 0, Q represents the number of lines connecting all of them, and W represents the number of lines connecting all of them. a W represents the total length of the change in humidity parameter. b W represents the total length of the changes in flow parameters. c The length of the sum of the changes in pressure parameters is represented by W, where W represents the sum of the lengths of all changes.

[0023] When B is greater than or equal to 1.5, the risk level of the monitoring point is determined to be Level 1 risk; when B is less than 1.5 but greater than or equal to 1, the risk level of the monitoring point is determined to be Level 2 risk; when B is less than 1, the risk level of the monitoring point is determined to be Level 3 risk; the risk level of Level 1, Level 2 and Level 3 risks decreases in that order.

[0024] Optionally, the feature value of the parameter set is the mean of all parameters in the parameter set; the processing module is specifically configured to: obtain the historical parameter set and use the historical parameter set as the model training set; use cross-validation combined with grid search to find the parameters for building the random forest model, obtain the random forest model; use the model training set to fit the random forest model; and process the mean of the parameter set according to the random forest model to obtain the risk prediction value.

[0025] Optionally, the statistical results of risk assessment indicators include the number of smoothing indicators and the number of containment indicators; the processing module is specifically configured to: compare each parameter in the parameter set with the standard parameter data to determine the size relationship between each parameter and the standard parameter data; determine the parameters that are less than or equal to the standard parameter data as smoothing indicators; determine the parameters that are greater than the standard parameter data as containment indicators; and count the number of smoothing indicators and the number of containment indicators.

[0026] Optionally, the processing module is specifically configured as follows:

[0027] Determine the ratio M between the number of smoothing indicators and the number of containment indicators; when M < 1, the first preset adjustment coefficient is used as the risk adjustment coefficient of the risk prediction value; when 1 ≤ M ≤ 2, the second preset adjustment coefficient is used as the risk adjustment coefficient of the risk prediction value; when 2 < M, the third preset adjustment coefficient is used as the risk adjustment coefficient of the risk prediction value.

[0028] Secondly, a pipeline risk monitoring method is provided, comprising: collecting temperature parameters from multiple temperature sensors evenly distributed at monitoring points, determining a temperature influence coefficient based on the temperature parameters from the multiple temperature sensors, and determining the acquisition operation mode of the monitoring points based on the temperature influence coefficient; establishing a time-parameter function based on the parameters collected by multiple operating parameter sensors operating in the acquisition operation mode at the same time, and determining the risk level of the monitoring point based on the slope of the time-parameter function and the length of parameter change; determining the risk prediction value of the monitoring point based on the parameter set of each operating parameter sensor among the multiple operating parameter sensors, and determining a risk adjustment coefficient based on the statistical results of risk assessment indicators in the parameter set of each operating parameter sensor, and determining the risk assessment value of the monitoring point based on the risk prediction value and the risk adjustment coefficient; the parameter set includes the parameters collected by the operating parameter sensors at different times; the risk assessment value and the risk adjustment coefficient are positively correlated; and the risk level and the risk assessment value are used to determine the pipeline risk monitoring result of the monitoring point.

[0029] Thirdly, a pipeline risk monitoring device is provided, including a memory and a processor; the memory is used to store computer-executed instructions, and the processor is connected to the memory via a bus; when the pipeline risk monitoring device is running, the processor executes the computer-executed instructions stored in the memory, so that the pipeline risk monitoring device performs the pipeline risk monitoring method of the second aspect.

[0030] Fourthly, a computer-readable storage medium is provided, comprising computer-executable instructions that, when executed on a computer, cause the computer to perform the pipeline risk monitoring method of the second aspect.

[0031] Fifthly, a computer program product is also provided, which includes computer instructions that, when executed on a pipeline risk monitoring device, cause the pipeline risk monitoring device to perform the pipeline risk monitoring method as described in the second aspect above.

[0032] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the pipeline risk monitoring device, or it may be packaged separately from the processor of the pipeline risk monitoring device; this application does not limit this.

[0033] The descriptions of the second, third, fourth, and fifth aspects of this application can be referenced to the detailed description of the first aspect.

[0034] In the embodiments of this application, the names of the aforementioned pipeline risk monitoring devices do not limit the devices or functional modules themselves. In actual implementation, these devices or functional modules may appear under other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalents.

[0035] The beneficial effects of this invention are as follows: By acquiring temperature parameters from multiple temperature sensors and dynamically adjusting the acquisition and operation mode of the monitoring points based on the temperature influence coefficient, the acquisition and operation mode can be adaptively adjusted according to the temperature influence coefficient when the ambient temperature changes, reducing parameter errors caused by temperature fluctuations and thus improving the accuracy and flexibility of monitoring. By using a time-parameter function and a random forest model to comprehensively process the parameters from multiple sensors, the risk level of the monitoring points can be assessed, helping to promptly identify potential risks and improving pipeline safety management. Based on the risk level and the adjusted risk assessment value, data support is provided for pipeline safety management. Attached Figure Description

[0036] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0037] Figure 1 This application provides a schematic diagram of the structure of a pipeline risk monitoring system.

[0038] Figure 2 A schematic diagram of another pipeline risk monitoring system provided in this application embodiment;

[0039] Figure 3A schematic diagram of another pipeline risk monitoring system provided in this application embodiment;

[0040] Figure 4 A schematic diagram of another pipeline risk monitoring system provided in this application embodiment;

[0041] Figure 5 A schematic flowchart of a pipeline risk monitoring method provided in an embodiment of this application;

[0042] Figure 6 This is a schematic diagram of a pipeline risk monitoring device provided in an embodiment of this application. Detailed Implementation

[0043] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0044] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0045] To improve the automation level and stability of pipeline risk monitoring, this application provides a pipeline risk monitoring system. For example... Figure 1 The diagram shown is a structural schematic of a pipeline risk monitoring system provided in an embodiment of this application. Figure 1 The pipeline risk monitoring system shown includes: a data acquisition module 101 and a processing module 102.

[0046] Considering that temperature can affect the operating status of monitored pipelines, such as the accuracy of humidity measurements, and thus the precision of humidity measurement, a change in the acquisition mode of the monitoring points is necessary if there are significant temperature differences. Therefore, the acquisition module 101 can be configured to acquire temperature parameters from multiple temperature sensors evenly distributed at the monitoring points, determine a temperature influence coefficient based on the temperature parameters from these sensors, and then determine the acquisition mode of the monitoring points based on this temperature influence coefficient.

[0047] In this way, the acquisition module 101 can collect the temperature parameters of the monitoring points in real time through multiple temperature sensors installed at the monitoring points. Furthermore, the temperature sensors are evenly distributed at the monitoring points to ensure that the collected temperature parameters are representative, allowing the temperature influence coefficient to reflect the degree of influence of the current temperature on the monitoring points. Dynamically adjusting the acquisition operation mode of the monitoring points based on the temperature influence coefficient improves the system's adaptability and monitoring accuracy under different environmental conditions, ensuring the stability of subsequent monitoring.

[0048] The processing module 102 is the core module of the pipeline risk monitoring system. It is used to set the determined data acquisition and operation mode to the working mode of multiple operating parameter sensors deployed at the monitoring point, and then perform data analysis based on this. Specifically, the processing module 102 can be configured to establish a time-parameter function based on the parameters collected by multiple operating parameter sensors operating in the data acquisition and operation mode at the same time. The risk level of the monitoring point is determined based on the slope and the length of parameter change of the time-parameter function. For example, the processing module 102 can set the data acquisition and operation mode of the monitoring point to the working mode of the operating parameter sensors deployed at the monitoring point, statistically analyze the parameters collected by each operating parameter sensor at the same time under this working mode, establish a time-parameter function in chronological order, obtain the slope and the length of parameter change of the time-parameter function, and determine the risk level of the monitoring point based on the slope and the length of parameter change.

[0049] Thus, the processing module 102 can statistically analyze the parameters of all operating parameter sensors at various times, and establish a time-parameter function in chronological order. This function describes the regularity of changes in different types of operating parameters over time, thereby reflecting the dynamic changes of the monitoring points. Furthermore, the slope of the time-parameter function represents the rate of change of the operating parameters over time, and the length of the parameter change in the time-parameter function reflects the magnitude of the changes in the operating parameters. Combining these two aspects allows for accurate determination of the risk level of the monitoring points, providing data support for subsequent analysis.

[0050] The processing module 102 is also configured to determine the predicted risk value of a monitoring point based on the parameter sets of each of the multiple operating parameter sensors, and to determine the risk adjustment coefficient based on the statistical results of the risk assessment indicators in the parameter sets of each operating parameter sensor. Finally, based on the predicted risk value and the risk adjustment coefficient, the risk assessment value of the monitoring point is determined. The parameter set includes parameters collected by the operating parameter sensors at different times. The risk assessment value and the risk adjustment coefficient are positively correlated.

[0051] For example, processing module 102 can use a random forest model to make predictions and determine the risk prediction value. The random forest model is an ensemble learning algorithm that trains multiple decision trees and combines their results to make predictions. Specifically, processing module 102 can collect all parameter data from various operating parameter sensors at different times and establish a parameter set, extract feature values ​​from the parameter set, process these feature values ​​based on the random forest model to obtain the risk prediction value, determine corresponding risk assessment indicators for the parameter set and count the number of risk assessment indicators, determine a risk adjustment coefficient based on the number of risk assessment indicators, and adjust the risk prediction value according to the risk adjustment coefficient to obtain the risk assessment value.

[0052] Based on this, the risk adjustment coefficient is determined by the number of risk assessment indicators, and the risk assessment value is adjusted according to the risk adjustment coefficient. This avoids the calculation error caused by simply relying on the model, ensures the accuracy of the system's comprehensive analysis of multiple parameters, and makes the risk assessment value able to accurately reflect the pipeline condition at the monitoring point.

[0053] The processing module 102 is also configured to use the pipeline risk monitoring results, which are determined by the risk level and risk assessment value as monitoring points, to accurately reflect the current risk level of pipeline operation.

[0054] In some embodiments, combined with Figure 1 ,like Figure 2 The diagram shown is a structural schematic of another pipeline risk monitoring system provided in this application embodiment. The pipeline risk monitoring system further includes a management module 103.

[0055] The management module 103 can be configured to generate visual risk reports based on risk levels and adjusted risk assessment values ​​to improve the intelligence level of pipeline risk monitoring.

[0056] Visualized risk reports can present risk levels and adjusted risk assessment values ​​in the form of charts, clearly showing the risk levels and adjusted risk assessment values ​​of monitoring points. This prompts relevant personnel to take corresponding measures, thereby optimizing resource allocation in a targeted manner and improving the system's accuracy and monitoring adaptability in the face of multiple parameters.

[0057] In some embodiments of this application, the temperature influence coefficient is derived from the following formula:

[0058]

[0059] Where T represents the temperature influence coefficient, t i This represents the temperature parameter of any single temperature sensor, and n represents the number of temperature sensors.

[0060] Considering that high temperatures at the monitoring point can affect the pipeline's operating status, and that high temperatures can introduce errors into the acquired data—for example, at high temperatures, the internal material of the pressure sensor 107 will expand or contract due to temperature changes, leading to errors in the collected data—using higher sensitivity and higher frequency for data acquisition would result in the accumulation of random errors, thus affecting data analysis. To accurately determine the monitoring point's acquisition mode based on the temperature influence coefficient, the acquisition module 101 can be configured as follows: when T is less than 25°C, the monitoring point's acquisition mode is set to Level 1; when T is greater than or equal to 25°C and less than 50°C, the monitoring point's acquisition mode is set to Level 2; and when T is greater than or equal to 50°C, the monitoring point's acquisition mode is set to Level 3. The sensitivity and acquisition frequency decrease sequentially from Level 1 to Level 3.

[0061] Specifically, the temperature sensor is a wireless, passive sensor. Multiple (e.g., 5) temperature sensors are evenly distributed at the monitoring points. Based on the magnitude of the temperature influence coefficient, the data acquisition and operation mode of the monitoring points is dynamically adjusted to adapt to different temperature environments. Level 1 mode offers the highest level of sensitivity and acquisition frequency, such as acquiring data every 10 minutes. Level 2 mode offers medium sensitivity and acquisition frequency, such as acquiring data every 20 minutes. Level 3 mode offers the lowest sensitivity and acquisition frequency, such as acquiring data every 30 minutes.

[0062] Based on this, the pipeline risk monitoring system can support operating parameter sensors to operate at low sensitivity and sampling frequency at high temperatures, avoiding the accumulation of random errors. This allows the system to adapt to environmental changes, thereby ensuring the accuracy and stability of subsequent analysis.

[0063] In some embodiments of this application, combined with Figure 1 ,like Figure 3 The diagram shown is a structural schematic of another pipeline risk monitoring system provided in this application embodiment. The pipeline risk monitoring system further includes a verification module 104.

[0064] The verification module 104 can verify the data acquisition operation mode, determining whether the data acquisition operation mode of the monitoring point is accurate based on historical data acquisition operation mode data. For example, the verification module 104 can be configured to determine a historical operation verification set with the same data acquisition operation mode as the monitoring point based on historical data acquisition operation mode data, extract historical operation features from the historical operation verification set, including historical operating sensitivity and historical acquisition frequency, and determine the operation similarity based on the historical operating sensitivity and historical acquisition frequency. The operation similarity is derived from the following formula:

[0065]

[0066] Where f0 represents the acquisition frequency determined by the monitoring point, f1 represents the historical acquisition frequency, q0 represents the working sensitivity determined by the monitoring point, and q1 represents the historical working sensitivity.

[0067] Furthermore, the verification module 104 can be pre-set with a runtime similarity threshold A. s The verification module 104 can also be configured to verify that A is greater than or equal to A. s If A is less than A, then the data collection and operation mode of the monitoring point is considered accurate. s If so, it is determined that the data collection and operation mode of the monitoring point is inaccurate.

[0068] Based on this, the pipeline risk monitoring system can accurately determine the data acquisition and operation mode of monitoring points by comparing the currently determined data acquisition and operation mode with historical data acquisition and operation mode data. This further improves the system's adaptability, enabling it to maintain efficient operation under different environmental conditions and avoiding errors caused by solely relying on the temperature influence coefficient. Furthermore, by continuously accumulating and updating historical data acquisition and operation modes, the system's adaptability is improved. Utilizing operational similarity to dynamically determine the accuracy of the data acquisition and operation mode of monitoring points reduces reliance on human experience, thereby improving the accuracy and efficiency of the judgment.

[0069] In some embodiments of this application, the verification module 104 can also use a clustering algorithm to re-determine the monitoring point's acquisition operation mode when it is determined that the acquisition operation mode of the monitoring point is inaccurate. For example, the verification module 104 can also be configured to, when it is determined that the acquisition operation mode of the monitoring point is inaccurate, obtain representative data for each acquisition operation mode based on historical acquisition operation mode data, combine the representative data and the operating characteristics of the monitoring point to establish an aggregated dataset, obtain the feature vector of each data in the aggregated dataset, determine the expected number of clusters k as 3 (corresponding to the three acquisition operation modes), initialize the parameters of the Gaussian distribution, calculate the probability that each data in the aggregated dataset belongs to each Gaussian distribution, obtain the responsibility value, and select the cluster with the largest responsibility value as the acquisition operation mode of the monitoring point. The representative data includes all parameters of the accurate acquisition operation mode corresponding to the historical temperature influence coefficient, including the response time, measurement accuracy, measurement range, working sensitivity, and acquisition frequency of all historical temperature sensors. The operating characteristics of the monitoring point represent the working sensitivity and acquisition frequency corresponding to the acquisition operation mode of the monitoring point that has been determined to be inaccurate.

[0070] Based on this, the pipeline risk monitoring system can support the verification of monitoring point acquisition operation modes through historical data. When the initial assessment is inaccurate, a clustering algorithm is used to re-determine the monitoring point's acquisition operation mode, improving system reliability and ensuring the accuracy of adaptive adjustments to different monitoring environments. Furthermore, by re-determining the monitoring point's acquisition operation mode through clustering algorithms, the Gaussian mixture model allows the system to adjust parameters based on the natural distribution of data, thus adapting to different acquisition operation modes and avoiding errors caused by relying solely on preliminary judgments. This improves the system's accuracy and adaptability.

[0071] In some embodiments of this application, considering that wired sensors are typically used in pipeline monitoring, they are easily limited by the complexity of wiring and maintenance costs, especially in long-distance and complex environments. Traditional wired sensors have limitations, leading to insufficient scalability and adaptability of the monitoring system. The operating parameter sensors in the embodiments of this application can be wireless passive sensors. Multiple operating parameter sensors may include a humidity sensor 105, a flow sensor 106, and a pressure sensor 107.

[0072] In this way, the operation parameter sensor adopts a wireless passive sensor, which avoids the wiring limitations and maintenance problems of wired sensors. In addition, the wireless passive sensor has good flexibility. By acquiring a variety of parameters from the wireless passive sensor, the adaptability of the pipeline monitoring system can be improved.

[0073] For example, combined Figure 1 ,like Figure 4 The diagram shown illustrates the structure of another pipeline risk monitoring system provided in this application embodiment. The acquisition module 101 and processing module 102 can be connected to the humidity sensor 105, flow sensor 106, and pressure sensor 107, respectively. The acquisition module 101 can acquire the parameters collected by the humidity sensor 105, flow sensor 106, and pressure sensor 107. The processing module 102 can set the operating modes of the humidity sensor 105, flow sensor 106, and pressure sensor 107 based on the acquisition and operation modes of the monitoring points.

[0074] For example, the processing module 102 can be specifically configured to set the acquisition operation mode of the monitoring point to the working mode of humidity sensor 105, flow sensor 106 and pressure sensor 107, and acquire the humidity parameter of humidity sensor 105, the flow parameter of flow sensor 106 and the pressure parameter of pressure sensor 107 at the same time; establish the y-axis of the time-parameter function with the unit quantity of humidity parameter, the unit quantity of flow parameter and pressure parameter, establish the x-axis of the time-parameter function in the time sequence, and substitute all humidity parameter, flow parameter and pressure parameter at the same time to obtain the time-parameter function.

[0075] In the above embodiments, considering that the data collected by the humidity sensor 105, flow sensor 106 and pressure sensor 107 are all affected by temperature, the pipeline risk monitoring system can accurately set the data collection and operation mode to the working mode of the humidity sensor 105, flow sensor 106 and pressure sensor 107, so that the three types of sensors can dynamically adapt to the temperature environment and improve the accuracy of the collected data.

[0076] Furthermore, the operating mode includes the corresponding operating sensitivity and sampling frequency for each data acquisition mode. Based on the operating mode, it can acquire the humidity parameter of humidity sensor 105, the flow parameter of flow sensor 106, and the pressure parameter of pressure sensor 107 at the same time. For example, if the primary mode is set to operating mode, then in this operating mode, the humidity sensor 105, flow sensor 106, and pressure sensor 107 will acquire data once every 10 minutes. A time-parameter function x-axis is established with 10-minute intervals, starting from 0 and progressing to the right as 10 minutes, 20 minutes, 30 minutes, etc., with the interval between two values ​​representing the sampling frequency. A time-parameter function y-axis is established using the units of humidity parameter, flow parameter, and pressure parameter, with the humidity parameter unit being 2 g / m³. 3 The unit for the flow rate parameter is 0.5 m³ / s, and the unit for the pressure parameter is 10. 6 N / m 2 The y-axis values, expressed in units of humidity parameters, start from 0 and increase sequentially upwards to 2g / m³. 3 4g / m 3 and 6g / m 3 The units of flow rate and pressure are similar, and will not be repeated here.

[0077] In this way, the y-axis integrates the units of humidity, flow, and pressure parameters. The time-parameter function contains multiple parameters, which can clearly reflect the regularity of the changes in humidity, flow, and pressure parameters of different types of sensors over time, thereby reflecting the dynamic changes of the monitoring point and improving the system's comprehensive processing capability for multi-parameter data.

[0078] In some embodiments of this application, in order to accurately obtain the slope and parameter change length of the time-parameter function, and thus determine the risk level of the monitoring point based on the slope and parameter change length, the processing module 102 may be specifically configured to: connect all adjacent humidity parameters with straight lines, connect all adjacent flow parameters with straight lines, and connect all adjacent pressure parameters with straight lines; count the number of all straight line connections and the number of all lines with a slope greater than 0; obtain the sum of the lengths of all humidity parameter changes with a slope greater than 0, the sum of the lengths of all flow parameter changes with a slope greater than 0, and the sum of the lengths of all pressure parameter changes with a slope greater than 0; obtain the sum of all lengths of changes; and determine the ratio coefficient, which is derived from the following formula:

[0079]

[0080] Where B represents the ratio coefficient, Q a W represents the number of lines with a slope greater than 0, Q represents the number of lines connecting all of them, and W represents the number of lines connecting all of them. a W represents the total length of the change in humidity parameter. b W represents the total length of the changes in flow parameters. c The length of the pressure parameter variation represents the total length of the variation, and W represents the total length of all variations.

[0081] Furthermore, the processing module 102 can determine the risk level of the monitoring point as Level 1 risk when B is greater than or equal to 1.5; determine the risk level of the monitoring point as Level 2 risk when B is less than 1.5 but greater than or equal to 1; and determine the risk level of the monitoring point as Level 3 risk when B is less than 1. The risk level of Level 1 risk, Level 2 risk, and Level 3 risk decreases in that order.

[0082] In the above embodiments, by obtaining the slope and the length of parameter change of the time-parameter function, the regularity of the changes in humidity, flow, and pressure parameters over time can be measured. A slope greater than 0 indicates an upward linear trend. Obtaining the sum of the lengths of changes in humidity, flow, and pressure parameters with slopes greater than 0 reflects the overall upward change in the parameters within the time-parameter function. Combining both factors to determine the risk level of the monitoring point avoids misjudgments caused by a single standard.

[0083] For example, if only the slope of the time-parameter function is considered, there may be multiple segments of lines with a slope greater than 0, while only one segment has a slope less than 0. However, the sum of the lengths of the segments with slopes greater than 0 is negligible compared to the sum of the lengths of the segments with slopes less than 0. Therefore, judging the risk level of the current monitoring point solely based on the slope will not reveal the overall change. Combining both approaches allows for a more accurate assessment of the monitoring point's risk level, reducing errors in human judgment and improving the accuracy and stability of the system's comprehensive processing of multiple parameters.

[0084] In some embodiments of this application, the characteristic value of the parameter set is the mean of all parameters in the parameter set. To accurately determine the risk prediction value, the processing module 102 may be specifically configured to: acquire a historical parameter set and use it as a model training set; use cross-validation combined with grid search to find the parameters for building a random forest model, derive the random forest model; fit the random forest model using the model training set; and process the mean of the parameter set according to the random forest model to obtain the risk prediction value.

[0085] Specifically, the historical parameter set contains all parameter data from the same type of sensor at different historical moments. Using this historical parameter set as the model training set improves the generalization ability of the random forest model. The parameters for building the random forest model are found by combining cross-validation with grid search. Cross-validation divides the data into several parts and trains the model multiple times to verify its stability and performance. Grid search searches for parameter combinations to build the random forest model. Using the model training set to fit the random forest model reduces the risk of overfitting, thereby improving the accuracy and stability of the random forest model. Furthermore, substituting the features of the parameter set into the trained random forest model can accurately derive risk prediction values. The higher the risk prediction value, the higher the urgency of the risk at the monitoring point, requiring attention and relevant measures, thus enhancing the reliability and stability of the system for pipeline safety management.

[0086] In some embodiments of this application, the statistical results of risk assessment indicators include the number of smoothing indicators and the number of containment indicators. To accurately determine the statistical results of risk assessment indicators, the processing module 102 may be specifically configured to: compare each parameter in the parameter set with standard parameter data to determine the magnitude relationship between each parameter and the standard parameter data; identify parameters less than or equal to the standard parameter data as smoothing indicators; identify parameters greater than the standard parameter data as containment indicators; and count the number of smoothing indicators and the number of containment indicators.

[0087] For example, the standard parameter data for humidity sensor 105 can be 11 g / m³. 3 The standard parameter data for flow sensor 106 can be 2 m³ / s, and the standard parameter data for pressure sensor 107 can be 4*10. 6N / m 2 The specific adjustments can be made based on the actual pipe diameter, pipe length, and pump station operating parameters.

[0088] In some embodiments of this application, in order to accurately determine the risk adjustment coefficient based on the statistical results of risk assessment indicators, the processing module 102 may be specifically configured to: determine the ratio M between the number of smoothing indicators and the number of containment indicators; when M < 1, use the first preset adjustment coefficient as the risk adjustment coefficient of the risk prediction value; when 1 ≤ M ≤ 2, use the second preset adjustment coefficient as the risk adjustment coefficient of the risk prediction value; when 2 < M, use the third preset adjustment coefficient as the risk adjustment coefficient of the risk prediction value.

[0089] The first preset adjustment coefficient, the second preset adjustment coefficient, and the third preset adjustment coefficient can be manually set in the processing module 102 based on experience.

[0090] For example, the first preset adjustment coefficient can be 1.5, the second preset adjustment coefficient can be 0.8, and the third preset adjustment coefficient can be 0.6. The risk prediction value is adjusted according to the risk adjustment coefficient N, and the risk prediction value is U. The adjusted risk assessment value is then determined to be N*U. When M < 1, it indicates that the number of containment indicators is greater than the number of smoothing indicators, and the majority of the data exceeds the standard parameters, indicating a relatively serious risk at the monitoring point. Therefore, the risk assessment value needs to be appropriately increased to indicate the urgency of the risk at the monitoring point. When 1 ≤ M ≤ 2, it indicates that the number of smoothing indicators is equal to or exceeds the number of containment indicators, and the monitoring point is at a moderate risk. Therefore, the risk assessment value needs to be appropriately decreased. When 2 < M, it indicates that the number of smoothing indicators is the majority, requiring a further decrease in the risk assessment value. Therefore, the decrease is the largest, ensuring the comprehensiveness and accuracy of the system's multi-parameter processing.

[0091] In the above embodiments, the corresponding preset adjustment coefficient is selected according to the ratio of different indicators, which realizes the dynamic adjustment of the risk prediction value, avoids the error caused by relying solely on the random forest model, and ensures the accuracy of the adjustment.

[0092] Based on the embodiments described above, the pipeline risk monitoring system can acquire temperature parameters from multiple temperature sensors and dynamically adjust the data acquisition and operation mode of the monitoring points based on the temperature influence coefficient. When the ambient temperature changes, the system can adaptively adjust the data acquisition and operation mode according to the temperature influence coefficient, reducing parameter errors caused by temperature fluctuations and thus improving the accuracy and flexibility of monitoring.

[0093] Furthermore, the pipeline risk monitoring system can verify the data collection and operation mode of the monitoring points by collecting historical data. When the judgment is inaccurate, a clustering algorithm is used to redetermine the data collection and operation mode of the monitoring points, which improves the reliability of the system and ensures the accuracy of adaptive adjustments to different monitoring environments.

[0094] Furthermore, the pipeline risk monitoring system can comprehensively process parameters from multiple operating parameter sensors through time-parameter functions and random forest models to assess the risk level of monitoring points. This helps to promptly identify potential risks and improves pipeline safety management. Based on the risk level and adjusted risk assessment value, a visualized risk report is generated, which not only enhances the intelligence level of pipeline monitoring but also provides data support for pipeline safety management.

[0095] Furthermore, the pipeline risk monitoring system can use wireless passive sensors, avoiding the wiring limitations and maintenance difficulties of traditional wired sensors. Wireless passive sensors also have good flexibility, and by acquiring various parameters from wireless passive sensors, the adaptability of the pipeline monitoring system is improved.

[0096] like Figure 5 The diagram shown is a flowchart illustrating a pipeline risk monitoring method provided in an embodiment of this application. Figure 5 The pipeline risk monitoring method shown can be executed by a pipeline risk monitoring device. The pipeline risk monitoring device can be configured with software units to implement the functions of each module of the aforementioned pipeline risk monitoring system. These may include software units implementing the acquisition module, verification module, processing module, and management module. Figure 5 The pipeline risk monitoring methods shown include: S201-S204.

[0097] S201. Collect temperature parameters from multiple temperature sensors evenly distributed at monitoring points, determine the temperature influence coefficient based on the temperature parameters from multiple temperature sensors, and determine the data collection and operation mode of the monitoring points based on the temperature influence coefficient.

[0098] S202. Based on the parameters collected by multiple operating parameter sensors in the data acquisition mode at the same time, establish a time-parameter function, and determine the risk level of the monitoring point according to the slope of the time-parameter function and the length of the parameter change.

[0099] S203. Based on the parameter sets of each operating parameter sensor among multiple operating parameter sensors, determine the risk prediction value of the monitoring point, and determine the risk adjustment coefficient according to the statistical results of the risk assessment indicators in the parameter sets of each operating parameter sensor. Based on the risk prediction value and the risk adjustment coefficient, determine the risk assessment value of the monitoring point.

[0100] The parameter set includes parameters collected by the operating parameter sensors at different times. The risk assessment value and the risk adjustment coefficient are positively correlated.

[0101] S204. Pipeline risk monitoring results based on risk level and risk assessment value as the monitoring points.

[0102] In some embodiments of this application, the determination of the temperature influence coefficient based on the temperature parameters of multiple temperature sensors in S201 above can be achieved based on the following formula:

[0103]

[0104] Where T represents the temperature influence coefficient, t i This represents the temperature parameter of any single temperature sensor, and n represents the number of temperature sensors.

[0105] In the above S201, when the pipeline risk monitoring device determines the acquisition and operation mode of the monitoring point based on the temperature influence coefficient, the embodiments of this application provide an optional implementation method, including: S2011-S2013.

[0106] S2011. When T is less than 25, the data acquisition and operation mode of the monitoring point is determined to be the first-level mode.

[0107] S2012. When T is greater than or equal to 25 and less than 50, the data acquisition and operation mode of the monitoring point is determined to be the secondary mode.

[0108] S2013. When T is greater than or equal to 50, the data collection and operation mode of the monitoring point is determined to be the third-level mode.

[0109] The sensitivity and acquisition frequency of the Level 1, Level 2, and Level 3 modes decrease sequentially.

[0110] In some embodiments of this application, the pipeline risk monitoring method further includes: S205-S208.

[0111] S205. Based on the historical data of the collected operation mode, determine the historical operation verification set with the same data collection operation mode as the monitoring point, and extract the historical operation features from the historical operation verification set.

[0112] Among them, historical operational characteristics include historical operational sensitivity and historical data collection frequency.

[0113] S206. Determine the operational similarity based on historical operating sensitivity and historical acquisition frequency.

[0114] For example, the similarity can be derived from the following formula:

[0115]

[0116] Where f0 represents the acquisition frequency corresponding to the acquisition operation mode of the monitoring point, f1 represents the historical acquisition frequency, q0 represents the working sensitivity corresponding to the acquisition operation mode of the monitoring point, and q1 represents the historical working sensitivity.

[0117] S207. When A is greater than or equal to the similarity threshold, the monitoring point's data collection and operation mode is judged to be accurate.

[0118] S208. When A is less than the similarity threshold, the monitoring point's data collection and operation mode is judged to be inaccurate.

[0119] In some embodiments of this application, the pipeline risk monitoring method further includes: S209-S211.

[0120] S209. When it is determined that the data collection and operation mode of the monitoring point is inaccurate, representative data for each data collection and operation mode shall be obtained based on historical data collection and operation mode data.

[0121] S210. Combine the operational characteristics of representative data and monitoring points to establish an aggregated dataset, and obtain the feature vector of each data point in the aggregated dataset.

[0122] S211. Determine the expected number of clusters k as 3, initialize the parameters of the Gaussian distribution, calculate the probability that each data point in the aggregated dataset belongs to each Gaussian distribution, obtain the responsibility value, and select the cluster with the largest responsibility value as the collection and operation mode of the monitoring point.

[0123] In some embodiments of this application, the operating parameter sensors are wireless passive sensors. Multiple operating parameter sensors include humidity sensors, flow sensors, and pressure sensors. In S202 above, when the pipeline risk monitoring device establishes a time-parameter function based on parameters collected simultaneously by multiple operating parameter sensors operating in the data acquisition mode, this application provides an optional implementation method, including: S2021-S2023.

[0124] S2021. Set the data acquisition and operation mode of the monitoring point to the working mode of humidity sensor, flow sensor and pressure sensor.

[0125] S2022. Obtain the humidity parameters of the humidity sensor, the flow parameters of the flow sensor, and the pressure parameters of the pressure sensor at the same time.

[0126] S2023. Establish the y-axis of the time-parameter function using the units of humidity, flow, and pressure parameters, and establish the x-axis of the time-parameter function in time sequence. Substitute all humidity, flow, and pressure parameters at the same time into the x-axis to obtain the time-parameter function.

[0127] In some embodiments of this application, when the pipeline risk monitoring device determines the risk level of the monitoring point in S202 above, that is, when the pipeline risk monitoring device determines the risk level of the monitoring point based on the slope of the time-parameter function and the length of the parameter change, the embodiments of this application provide an optional implementation method, including: S2024-S2028.

[0128] S2024. Connect all adjacent humidity parameters with a straight line, connect all adjacent flow parameters with a straight line, and connect all adjacent pressure parameters with a straight line.

[0129] S2025. Count the number of all straight line connections and the number of all straight lines with a slope greater than 0. Obtain the total length of the change in humidity parameter with a slope greater than 0, the total length of the change in flow parameter with a slope greater than 0, and the total length of the change in pressure parameter with a slope greater than 0. Obtain the total length of all changes and determine the ratio coefficient.

[0130] For example, the ratio coefficient can be obtained from the following formula:

[0131]

[0132] Where B represents the ratio coefficient, Q a W represents the number of lines with a slope greater than 0, Q represents the number of lines connecting all of them, and W represents the number of lines connecting all of them. a W represents the total length of the change in humidity parameter. b W represents the total length of the changes in flow parameters. c The length of the pressure parameter variation represents the total length of the variation, and W represents the total length of all variations.

[0133] S2026. When B is greater than or equal to 1.5, the risk level of the monitoring point is determined to be Level 1 risk.

[0134] S2027. When B is less than 1.5 and greater than or equal to 1, the risk level of the monitoring point is determined to be Level II risk.

[0135] S2028. When B is less than 1, the risk level of the monitoring point is determined to be Level 3 risk.

[0136] The risk levels decrease sequentially from Level 1 to Level 2 and Level 3.

[0137] In some embodiments of this application, the characteristic value of the parameter set is the mean of all parameters in the parameter set. In S203 above, when the pipeline risk monitoring device determines the risk prediction value of the monitoring point based on the parameter set of each of the multiple operating parameter sensors, this application provides an optional implementation method, including: S2031-S2032.

[0138] S2031. Obtain the historical parameter set and use it as the model training set. Use cross-validation combined with grid search to find the parameters for building the random forest model, obtain the random forest model, and use the model training set to fit the random forest model.

[0139] S2032. Process the mean of the parameter set according to the random forest model to obtain the risk prediction value.

[0140] In some embodiments of this application, the statistical results of risk assessment indicators include the number of smoothing indicators and the number of containment indicators. In S203 above, when the pipeline risk monitoring device determines the risk adjustment coefficient based on the statistical results of risk assessment indicators in the parameter set of each operating parameter sensor, this application embodiment provides an optional implementation method, including: S2033-S2036.

[0141] S2033. Compare each parameter in the parameter set with the standard parameter data to determine the size relationship between each parameter and the standard parameter data.

[0142] S2034. Parameters that are less than or equal to the standard parameter data are determined as smoothing indices.

[0143] S2035. Parameters that are greater than the standard parameter data shall be identified as control indicators.

[0144] S2036, the number of statistical smoothing indicators and the number of containment indicators.

[0145] In some embodiments of this application, the statistical results of risk assessment indicators include the number of smoothing indicators and the number of containment indicators. In S203 above, when the pipeline risk monitoring device determines the risk assessment value of the monitoring point based on the risk prediction value and the risk adjustment coefficient, this application provides an optional implementation method, including: S2037-S20310.

[0146] S2037. Determine the ratio M between the number of smoothing indicators and the number of restraining indicators.

[0147] S2038. When M < 1, the first preset adjustment coefficient is used as the risk adjustment coefficient for the risk prediction value.

[0148] S2039. When 1≤M≤2, the second preset adjustment coefficient shall be used as the risk adjustment coefficient of the risk prediction value.

[0149] S20310. When 2 < M, the third preset adjustment coefficient is used as the risk adjustment coefficient for the risk prediction value.

[0150] The above primarily describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the aforementioned functions, the pipeline risk monitoring device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0151] This application embodiment can divide the pipeline risk monitoring device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0152] like Figure 6 The diagram shown is a structural schematic of a pipeline risk monitoring device provided in an embodiment of this application. Figure 6 The pipeline risk monitoring device shown includes: a data acquisition unit 301 and a processing unit 302.

[0153] The acquisition unit 301 is configured to acquire temperature parameters from multiple temperature sensors evenly distributed at monitoring points, determine a temperature influence coefficient based on the temperature parameters of the multiple temperature sensors, and determine the acquisition operation mode of the monitoring points based on the temperature influence coefficient.

[0154] The processing unit 302 is configured to establish a time-parameter function based on the parameters collected by multiple operating parameter sensors at the same time while operating in the acquisition mode, and determine the risk level of the monitoring point according to the slope of the time-parameter function and the length of the parameter change.

[0155] The processing unit 302 is also configured to determine the predicted risk value of a monitoring point based on the parameter sets of each of the multiple operating parameter sensors, and to determine the risk adjustment coefficient based on the statistical results of the risk assessment indicators in the parameter sets of each operating parameter sensor. Finally, based on the predicted risk value and the risk adjustment coefficient, the risk assessment value of the monitoring point is determined. The parameter set includes parameters collected by the operating parameter sensors at different times. The risk assessment value and the risk adjustment coefficient are positively correlated.

[0156] Processing unit 302 is also configured to monitor pipeline risk results based on risk level and risk assessment value as monitoring points.

[0157] Alternatively, the temperature influence coefficient can be derived from the following formula:

[0158]

[0159] Where T represents the temperature influence coefficient, t i This represents the temperature parameter of any single temperature sensor, and n represents the number of temperature sensors.

[0160] The acquisition unit 301 is specifically configured as follows: when T is less than 25, the acquisition operation mode of the monitoring point is determined to be Level 1 mode. When T is greater than or equal to 25 and less than 50, the acquisition operation mode of the monitoring point is determined to be Level 2 mode. When T is greater than or equal to 50, the acquisition operation mode of the monitoring point is determined to be Level 3 mode. The working sensitivity and acquisition frequency decrease sequentially from Level 1 mode to Level 2 mode and Level 3 mode.

[0161] Optionally, the system further includes a verification unit 303. Verification unit 303 is configured to determine a historical operation verification set with the same acquisition operation mode as the monitoring point based on historical acquisition operation mode data, extract historical operation features from the historical operation verification set, including historical operating sensitivity and historical acquisition frequency, and determine the operation similarity based on the historical operating sensitivity and historical acquisition frequency. The operation similarity is derived from the following formula:

[0162]

[0163] Where f0 represents the acquisition frequency corresponding to the acquisition operation mode of the monitoring point, f1 represents the historical acquisition frequency, q0 represents the working sensitivity corresponding to the acquisition operation mode of the monitoring point, and q1 represents the historical working sensitivity.

[0164] If A is greater than or equal to the operational similarity threshold, the monitoring point's data collection and operation mode is considered accurate. If A is less than the operational similarity threshold, the monitoring point's data collection and operation mode is considered inaccurate.

[0165] Optionally, the verification unit 303 is further configured to, when the judgment that the acquisition operation mode of the monitoring point is inaccurate, obtain representative data for each acquisition operation mode based on historical acquisition operation mode data, combine the representative data and the operation characteristics of the monitoring point to establish an aggregated dataset, obtain the feature vector of each data in the aggregated dataset, determine the expected number of clusters k as 3, initialize the parameters of the Gaussian distribution, calculate the probability that each data in the aggregated dataset belongs to each Gaussian distribution, obtain the responsibility value, and select the cluster with the largest responsibility value as the acquisition operation mode of the monitoring point.

[0166] Optionally, the operating parameter sensors are wireless passive sensors, and the multiple operating parameter sensors include a humidity sensor, a flow sensor, and a pressure sensor.

[0167] The processing unit 302 is specifically configured to: set the acquisition operation mode of the monitoring point to the working mode of the humidity sensor, flow sensor, and pressure sensor; acquire the humidity parameter of the humidity sensor, the flow parameter of the flow sensor, and the pressure parameter of the pressure sensor at the same time; establish the y-axis of the time-parameter function using the units of the humidity parameter, the unit of the flow parameter, and the unit of the pressure parameter; establish the x-axis of the time-parameter function in the order of time; and substitute all the humidity parameter, flow parameter, and pressure parameter at the same time to obtain the time-parameter function.

[0168] Optionally, processing unit 302 is specifically configured to: connect all adjacent humidity parameters with straight lines, connect all adjacent flow parameters with straight lines, and connect all adjacent pressure parameters with straight lines. It then counts the number of all straight lines and the number of lines with a slope greater than 0, obtains the sum of the lengths of the changes in humidity parameters with a slope greater than 0, the sum of the lengths of the changes in flow parameters with a slope greater than 0, and the sum of the lengths of the changes in pressure parameters with a slope greater than 0, obtains the total length of all changes, and determines the ratio coefficient, which is derived from the following formula:

[0169]

[0170] Where B represents the ratio coefficient, Q a W represents the number of lines with a slope greater than 0, Q represents the number of lines connecting all of them, and W represents the number of lines connecting all of them. a W represents the total length of the change in humidity parameter. b W represents the total length of the changes in flow parameters. c The length of the pressure parameter variation represents the total length of the variation, and W represents the total length of all variations.

[0171] When B is greater than or equal to 1.5, the risk level of the monitoring point is determined to be Level 1 risk. When B is less than 1.5 but greater than or equal to 1, the risk level of the monitoring point is determined to be Level 2 risk. When B is less than 1, the risk level of the monitoring point is determined to be Level 3 risk. The risk level decreases sequentially from Level 1 to Level 2 to Level 3 risk.

[0172] Optionally, the feature value of the parameter set is the mean of all parameters in the parameter set. The processing unit 302 is specifically configured to: acquire a historical parameter set and use it as a model training set; use cross-validation combined with grid search to find the parameters for building the random forest model, derive the random forest model; fit the random forest model using the model training set; and process the mean of the parameter set according to the random forest model to obtain the risk prediction value.

[0173] Optionally, the statistical results of the risk assessment indicators include the number of smoothing indicators and the number of containment indicators. Processing unit 302 is specifically configured to: compare each parameter in the parameter set with standard parameter data to determine the magnitude relationship between each parameter and the standard parameter data; identify parameters less than or equal to the standard parameter data as smoothing indicators; identify parameters greater than the standard parameter data as containment indicators; and count the number of smoothing indicators and the number of containment indicators.

[0174] Optionally, the processing unit 302 is specifically configured as follows:

[0175] Determine the ratio M between the number of smoothing indicators and the number of containment indicators. When M < 1, the first preset adjustment coefficient is used as the risk adjustment coefficient for the risk forecast value. When 1 ≤ M ≤ 2, the second preset adjustment coefficient is used as the risk adjustment coefficient for the risk forecast value. When 2 < M, the third preset adjustment coefficient is used as the risk adjustment coefficient for the risk forecast value.

[0176] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0178] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0179] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do 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 pipeline risk monitoring system, characterized in that, include: Acquisition module and processing module; The acquisition module is configured to acquire temperature parameters from multiple temperature sensors evenly distributed at monitoring points, determine a temperature influence coefficient based on the temperature parameters from the multiple temperature sensors, and determine the acquisition operation mode of the monitoring points based on the temperature influence coefficient. The processing module is configured to establish a time-parameter function based on parameters collected at the same time by multiple operating parameter sensors operating in the acquisition mode, and determine the risk level of the monitoring point according to the slope of the time-parameter function and the length of parameter change. The processing module is further configured to determine the risk prediction value of the monitoring point based on the parameter set of each of the plurality of operating parameter sensors, and to determine the risk adjustment coefficient based on the statistical results of the risk assessment indicators in the parameter set of each operating parameter sensor, and to determine the risk assessment value of the monitoring point based on the risk prediction value and the risk adjustment coefficient. The parameter set includes parameters collected by the operating parameter sensor at different times; the risk assessment value and the risk adjustment coefficient are positively correlated; The processing module is also configured to determine the pipeline risk monitoring result of the monitoring point based on the risk level and the risk assessment value. The temperature influence coefficient is derived from the following formula: ； in, Indicates the temperature effect coefficient. This represents the temperature parameter of any temperature sensor. Indicates the number of temperature sensors; The acquisition module is specifically configured as follows: when If the value is less than 25, the data acquisition and operation mode of the monitoring point will be determined as Level 1 mode; when If the value is greater than or equal to 25 and less than 50, then the data acquisition and operation mode of the monitoring point is determined to be the secondary mode; when If the value is greater than or equal to 50, the data acquisition and operation mode of the monitoring point will be determined as Level 3. The operating sensitivity and acquisition frequency of the first-level mode, the second-level mode, and the third-level mode decrease sequentially. The operating parameter sensor is a wireless passive sensor, and the plurality of operating parameter sensors include a humidity sensor, a flow sensor, and a pressure sensor; The processing module is specifically configured as follows: The data acquisition and operation mode of the monitoring point is set to the working mode of the humidity sensor, the flow sensor, and the pressure sensor. The humidity parameter of the humidity sensor, the flow parameter of the flow sensor, and the pressure parameter of the pressure sensor are obtained at the same time. The y-axis of the time-parameter function is established using the units of humidity, flow rate, and pressure. The x-axis of the time-parameter function is established in time sequence. The time-parameter function is obtained by substituting all the humidity, flow rate, and pressure parameters at the same time. Connect all adjacent humidity parameters with a straight line, connect all adjacent flow rate parameters with a straight line, and connect all adjacent pressure parameters with a straight line. Count the number of all straight lines and the number of all lines with a slope greater than 0. Obtain the sum of the lengths of all humidity parameter changes with a slope greater than 0, the sum of the lengths of all flow parameter changes with a slope greater than 0, and the sum of the lengths of all pressure parameter changes with a slope greater than 0. Determine the ratio coefficient, which is derived from the following formula: ； in, Represents the ratio coefficient. This represents the number of lines with a slope greater than 0. This indicates the total number of straight line connections. This represents the total length of the change in humidity parameters. This represents the total length of the changes in flow parameters. This represents the total length of the change in pressure parameters. This represents the sum of the lengths of all changes; when If the value is greater than or equal to 1.5, the risk level of the monitoring point is determined to be Level 1 risk. when If the value is less than 1.5 and greater than or equal to 1, then the risk level of the monitoring point is determined to be Level II risk. when If the value is less than 1, the risk level of the monitoring point is determined to be Level 3 risk. The risk levels of the Level 1 risk, Level 2 risk, and Level 3 risk decrease sequentially.

2. The pipeline risk monitoring system according to claim 1, characterized in that, Also includes: Verification module; The verification module is configured to determine a historical operation verification set with the same acquisition operation mode as the monitoring point based on historical acquisition operation mode data, extract historical operation features from the historical operation verification set, the historical operation features including historical operating sensitivity and historical acquisition frequency, and determine the operation similarity based on the historical operating sensitivity and historical acquisition frequency, the operation similarity being derived by the following formula: ； in, This indicates the acquisition frequency corresponding to the acquisition operation mode of the monitoring point. Indicates the historical data collection frequency. This indicates the operating sensitivity corresponding to the data acquisition and operation mode of the monitoring point. Indicates sensitivity to historical work; when If the similarity threshold is greater than or equal to the operating similarity threshold, then the data collection and operation mode of the monitoring point is determined to be accurate. when If the similarity is less than the specified operating threshold, the data collection and operation mode of the monitoring point is determined to be inaccurate.

3. The pipeline risk monitoring system according to claim 2, characterized in that, The verification module is further configured as follows: When it is determined that the data collection operation mode of the monitoring point is inaccurate, representative data for each data collection operation mode is obtained based on the historical data collection operation mode data. The representative data and the operation characteristics of the monitoring point are combined to establish an aggregated dataset. The feature vector of each data in the aggregated dataset is obtained, the expected number of clusters k is determined to be 3, the parameters of the Gaussian distribution are initialized, the probability of each data in the aggregated dataset belonging to each Gaussian distribution is calculated, the responsibility value is obtained, and the cluster with the largest responsibility value is selected as the data collection operation mode of the monitoring point.

4. The pipeline risk monitoring system according to claim 1, characterized in that, The characteristic value of the parameter set is the mean of all parameters in the parameter set; The processing module is specifically configured as follows: Obtain a historical parameter set and use it as a model training set. Use cross-validation combined with grid search to find the parameters for building a random forest model, and obtain the random forest model. Use the model training set to fit the random forest model, and process the mean of the parameter set according to the random forest model to obtain the risk prediction value.

5. The pipeline risk monitoring system according to claim 1, characterized in that, The statistical results of the risk assessment indicators include the number of smoothing indicators and the number of containment indicators; The processing module is specifically configured as follows: The parameters in the parameter set are compared with the standard parameter data to determine the size relationship between each parameter and the standard parameter data; Parameters that are less than or equal to the standard parameter data are determined as smoothing indices; Parameters that are greater than the standard parameter data are identified as containment indicators; The number of the smoothing indicators and the number of the containment indicators are counted.

6. The pipeline risk monitoring system according to claim 5, characterized in that, The processing module is specifically configured as follows: Determine the ratio M between the number of smoothing indicators and the number of restraint indicators; When M < 1, the first preset adjustment coefficient is used as the risk adjustment coefficient for the predicted risk value. When 1≤M≤2, the second preset adjustment coefficient is used as the risk adjustment coefficient for the predicted risk value; When 2 < M, the third preset adjustment coefficient is used as the risk adjustment coefficient for the predicted risk value.

7. A pipeline risk monitoring method, characterized in that, The pipeline risk monitoring system according to any one of claims 1-6 comprises: The temperature parameters of multiple temperature sensors evenly distributed at the monitoring points are collected, and a temperature influence coefficient is determined based on the temperature parameters of the multiple temperature sensors. The collection and operation mode of the monitoring points is then determined based on the temperature influence coefficient. Based on the parameters collected by multiple operating parameter sensors operating in the aforementioned acquisition mode at the same time, a time-parameter function is established. The risk level of the monitoring point is determined according to the slope of the time-parameter function and the length of the parameter change. Based on the parameter sets of each of the multiple operating parameter sensors, the predicted risk value of the monitoring point is determined, and the risk adjustment coefficient is determined according to the statistical results of the risk assessment indicators in the parameter sets of each operating parameter sensor. Based on the predicted risk value and the risk adjustment coefficient, the risk assessment value of the monitoring point is determined. The parameter set includes the parameters collected by the operating parameter sensors at different times. The risk assessment value and the risk adjustment coefficient are positively correlated. The risk level and the risk assessment value are used to determine the pipeline risk monitoring result of the monitoring point.

Images