A method and system for constructing a pipeline risk judgment model

Abstract

The present invention relates to the field of pipeline monitoring technology, and in particular to a method and system for constructing a pipeline risk determination model. The method comprises the following steps: using an air compressor to generate compressed air, and delivering the compressed air to a preset pipeline to obtain pipeline compressed air data; constructing an air purification model according to the pipeline compressed air data; performing air purification simulation according to the air purification model to obtain purified air data; detecting residual condensed water according to the purified air data to obtain residual condensed water data; identifying pipeline corrosion based on the residual condensed water data to obtain pipeline corrosion data; predicting the probability of pipeline perforation according to pipeline corrosion data; detecting accumulated solid particles according to the purified air data to obtain accumulated solid particle data; performing pipeline blockage analysis according to the accumulated solid particle data to obtain pipeline blockage data. The present invention improves the accuracy and real-time performance of pipeline risk determination based on pipeline monitoring technology, and significantly improves the pipeline safety recognition rate.

Description

Technical Field

[0001] The present invention relates to the technical field of pipeline monitoring, and in particular to a method and system for constructing a pipeline risk determination model. Background Art

[0002] Traditional pipeline risk assessment models usually adopt a standardized monitoring system structure consisting of pressure sensors, temperature sensors, flow meters, data acquisition terminals, central processing units and alarm devices, but rely on the setting of fixed thresholds, such as the change of a single physical parameter such as pressure, temperature, flow rate, etc. as the basis for judgment, lack of fusion processing of multi-source data, resulting in low risk identification accuracy and prone to missed or misjudgment; based on regular manual inspections and offline detection methods, it is impossible to achieve real-time monitoring and continuous tracking of pipeline operation status, resulting in delayed risk discovery and potential safety hazards; the modeling of complex mechanisms such as corrosion, blockage, and condensate accumulation is relatively rough, lacking oxygen reduction reaction, electrochemical The deep analysis of microscopic phenomena such as the corrosion process and particle deposition evolution cannot achieve accurate prediction; the lack of high-precision image recognition and 3D modeling capabilities cannot accurately obtain the internal structure changes and morphology information of the pipeline, and it is difficult to support risk positioning and visualization analysis at the structural level; the system module integration is low, the various risk identification methods are not interconnected, and there is a lack of a unified risk assessment system, which leads to one-sided overall judgment results and low intelligence level; the impact of environmental factors such as compressed air quality, water vapor content, and solid particle concentration on the operation status of the pipeline is ignored, and the causal relationship between external inducements and internal failures cannot be analyzed, making it difficult to build a risk judgment model with adaptive and dynamic learning capabilities. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a method and system for constructing a pipeline risk determination model to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for constructing a pipeline risk determination model includes the following steps:

[0005] Step S1: using an air compressor to generate compressed air, and delivering the compressed air to a preset pipeline to obtain pipeline compressed air data; constructing an air purification model according to the pipeline compressed air data; performing air purification simulation according to the air purification model to obtain purified air data;

[0006] Step S2: detecting residual condensed water according to the purified air data to obtain residual condensed water data; performing pipeline corrosion identification based on the residual condensed water data to obtain pipeline corrosion data; and predicting the probability of pipeline perforation according to the pipeline corrosion data;

[0007] Step S3: Detect the accumulated solid particles based on the purified air data to obtain the accumulated solid particle data; perform pipeline blockage analysis based on the accumulated solid particle data to obtain the pipeline blockage data;

[0008] Step S4: Construct a pipeline risk determination model based on the pipeline blockage data and the pipeline perforation probability; perform pipeline risk determination on the air purification simulation process according to the pipeline risk determination model to generate pipeline risk data; transmit the pipeline risk data to the risk monitoring system to execute the pipeline risk warning task.

[0009] By utilizing the process of generating and transporting compressed air to the pipeline, the present invention can effectively collect the air data in the pipeline, and then construct a more accurate air purification model, thereby realizing air purification simulation and obtaining purified air data. This process lays the foundation for subsequent pipeline status monitoring. In particular, it can simulate the air quality in the actual use environment and provide more real data support. By detecting the purified air data, the presence of residual condensate can be identified in a timely manner, and pipeline corrosion can be identified based on this data, which provides the necessary data support for the early detection and accurate prediction of pipeline corrosion. Based on the pipeline corrosion data obtained from this process, the pipeline perforation probability can be accurately predicted, so as to discover potential pipeline damage problems in advance and avoid the occurrence of safety hazards. The acquisition of the accumulated solid particle data and the implementation of the pipeline blockage analysis can help accurately identify the pipeline blockage situation and provide data basis for pipeline maintenance and cleaning. The analysis of these data can realize a comprehensive assessment of the pipeline operation status and provide more dimensional information for subsequent risk determination. Combining the pipeline blockage data with the perforation probability can effectively construct a pipeline risk determination model and accurately identify the pipeline risk through this model. This multi-dimensional data fusion risk determination method avoids the disadvantages of traditional models relying too much on a single physical parameter, and improves the accuracy and real-time performance of pipeline risk assessment. Finally, the risk monitoring system can obtain the pipeline risk data in a timely manner, thereby realizing the real-time monitoring and warning of pipeline risks and reducing the possibility of accidents. Through the real-time collection and fusion analysis of multi-source data, combined with advanced image recognition and three-dimensional modeling technologies, it can not only provide detailed information on the internal structure changes and morphology of the pipeline, but also dynamically learn and adaptively adjust the pipeline risks. This method breaks through the limitations of traditional pipeline monitoring, ensures the comprehensive monitoring of the pipeline operation status, can accurately capture various risks caused by environmental factors, equipment failures or material deterioration, and greatly improves the intelligent level and safety of the system. At the same time, through the causal correlation analysis of external incentives and internal failures, the system can provide more scientific decision-making support for pipeline maintenance and optimization.

[0010] Preferably, this specification also provides a construction system for a pipeline risk determination model, which is used to execute the construction method of the pipeline risk determination model as described above. The construction system of the pipeline risk determination model includes:

[0011] An air purification simulation module that uses an air compressor to generate compressed air and transports the compressed air to a preset pipeline to obtain pipeline compressed air data; constructs an air purification model based on the pipeline compressed air data; and performs air purification simulation according to the air purification model to obtain purified air data;

[0012] A pipeline corrosion identification module that detects residual condensate water based on the purified air data to obtain residual condensate water data; performs pipeline corrosion identification based on the residual condensate water data to obtain pipeline corrosion data; and predicts the pipeline perforation probability according to the pipeline corrosion data;

[0013] A pipeline blockage analysis module that detects accumulated solid particles based on the purified air data to obtain accumulated solid particle data; performs pipeline blockage analysis according to the accumulated solid particle data to obtain pipeline blockage data;

[0014] A pipeline risk determination module that constructs a pipeline risk determination model according to the pipeline blockage data and the pipeline perforation probability; performs pipeline risk determination on the air purification simulation process according to the pipeline risk determination model to generate pipeline risk data; and transmits the pipeline risk data to a risk monitoring system to execute the pipeline risk early warning task.

[0015] The construction system of the pipeline risk determination model of the present invention can implement any construction method of the pipeline risk determination model of the present invention. It is used as a medium for the operation and signal transmission between various modules to complete the construction method of the pipeline risk determination model. The internal modules of the system cooperate with each other, improving the accuracy and real-time performance of pipeline risk determination and significantly increasing the pipeline safety identification rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, objects, and advantages of the present invention will become more apparent:

[0017] Figure 1 It is a schematic flow chart of the steps of a construction method of a pipeline risk determination model of the present invention;

[0018] Figure 2 It is a detailed schematic flow chart of step S1 in the present invention;

[0019] Figure 3 It is a detailed schematic flow chart of step S16 in the present invention;

[0020] The implementation, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0021] The technical method of the present invention for the patent will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0023] It should be understood that although the terms "first", "second", etc. may be used here to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used here includes any and all combinations of one or more of the listed related items.

[0024] To achieve the above object, please refer to Figures 1 to 3 , the present invention provides a method for constructing a pipeline risk determination model, and the method includes the following steps:

[0025] Step S1: Use an air compressor to generate compressed air, and transport the compressed air to a preset pipeline to obtain pipeline compressed air data; construct an air purification model according to the pipeline compressed air data; perform air purification simulation according to the air purification model to obtain purified air data;

[0026] In this embodiment, start the air compressor and set its compression pressure to 0.7 MPa. The air compressor inhales atmospheric air and compresses the air to the set pressure range through the compression process. The volume flow rate of the compressed air is 10 m³ / min, and then it is transported to the preset pipeline through the pipeline system. At this time, use a flow meter to monitor the flow rate of the compressed air in the pipeline in real time, and record the pipeline compressed air data, including the flow rate, pressure, temperature, and humidity data of the compressed air. Based on these data, construct an air purification model, and the model needs to set input parameters such as the length and diameter of the pipeline, air flow velocity, and humidity in the compressed air. Based on these parameters, conduct air purification simulation, and the set simulation conditions include an inlet temperature of 30°C and a relative humidity of 60%. During the model calculation process, determine the impurity components in the air and remove them in combination with filtration devices (such as activated carbon, silica gel, etc.). After the simulation runs, output the purified air data, including the humidity value and impurity concentration after purification, as the input data for the subsequent steps.

[0027] Step S2: Detect the residual condensate water according to the purified air data to obtain the residual condensate water data; identify the pipeline corrosion based on the residual condensate water data to obtain the pipeline corrosion data; predict the pipeline perforation probability according to the pipeline corrosion data;

[0028] In this embodiment, according to the purified air data, extract the air humidity data and calculate the dew point temperature based on the humidity value. The dew point temperature is calculated through the relationship between humidity and temperature, and the set dew point temperature is 5°C. If the dew point temperature deviates from the preset target dew point value (for example, the target value is 8°C), a dew point deviation will occur. Then, record the dew point deviation period and collect the purified air data during this period, and use a condensate water sensor to detect the residual condensate water. The condensate water sensor can measure the volume of the condensate water accumulated in the pipeline in real time. A capacitive moisture sensor is used, and the moisture concentration range that this sensor can detect is 0.1% - 10%. The detected condensate water data is used to analyze the corrosion situation of the pipeline. By combining the residual condensate water data, identify the corrosion on the pipeline surface. The corrosion identification process includes irradiating the inner wall of the pipeline with high-energy X-rays to generate X-ray fluorescence spectra, and extracting the characteristic information of metal elements from the spectral data to identify the metal type (such as iron, copper, zinc, etc.). On this basis, conduct an electrochemical reaction simulation to generate current data, and conduct an oxygen reduction reaction simulation based on the current data and the metal element data. The simulation results provide the rust accumulation data. Based on these data, judge the severity of the pipeline corrosion and further analyze the pipeline perforation probability. The perforation probability is calculated through the pipeline corrosion data combined with a prediction model, considering the corrosion rate and the size of the corrosion area, and predict the pipeline perforation risk within a certain period of time.

[0029] Step S3: Detect the accumulated solid particles based on the purified air data to obtain the accumulated solid particle data; perform pipeline blockage analysis based on the accumulated solid particle data to obtain the pipeline blockage data;

[0030] In this embodiment, based on the purified air data, the particulate matter concentration is extracted. The concentration is measured using the light scattering method, and the particulate matter concentration range is 0.1 - 10 μg / m³. When the particle concentration exceeds a preset threshold (such as 5 g / m³), anomaly detection starts. The anomaly detection algorithm is used to analyze the concentration time series, and through the marking of timestamps, the time periods with abnormal concentration fluctuations are identified. Then, based on these abnormal concentration time periods, the inner wall image of the pipeline is captured using an industrial endoscope. When the image is captured, the resolution of the endoscope device is set to 0.01 mm. After the captured image undergoes color space conversion, the image recognition algorithm is used to detect the solid particle area in the image. The color range of the particle area is preset as gray to dark brown based on experimental data, and the particle area is accurately calibrated through this color range. For the calibrated particle area, a laser rangefinder is used to measure the thickness of the solid particles, with the accuracy set to 0.1 mm, to obtain the thickness data of the particulate matter. Then, based on the particle thickness and the three-dimensional structure of the particles, by calculating their volume, the accumulation situation of the particles is judged. The three-dimensional reconstruction algorithm is used to combine the measured particle thickness and its distribution to obtain the volume of the solid particles. Further, based on the volume and accumulation situation of the solid particles, it is evaluated whether the pipeline is blocked. This data is used for subsequent pipeline blockage analysis and generates pipeline blockage data, including the blockage location, blockage severity, and particle accumulation rate, etc.

[0031] Step S4: Construct a pipeline risk determination model based on the pipeline blockage data and the pipeline perforation probability; perform pipeline risk determination on the air purification simulation process according to the pipeline risk determination model to generate pipeline risk data; transmit the pipeline risk data to the risk monitoring system to execute the pipeline risk warning task.

[0032] In this embodiment, a pipeline risk determination model is constructed based on pipeline blockage data and pipeline perforation probability. The model combines pipeline blockage data and pipeline perforation probability data, and uses machine learning algorithms (such as decision trees or support vector machines) for training. The model input includes historical data, temperature, humidity, pressure, corrosion rate, particle concentration, etc. of the pipeline. After training, the model can score the risk of each pipeline and judge the risk level of the pipeline based on the score. The risk determination model includes two parts: pipeline material damage risk assessment and leakage risk assessment. The pipeline material damage risk assessment is based on the data of corrosion rate and corrosion area, and the leakage risk assessment is predicted based on the blockage and perforation probability of the pipeline. The generated pipeline risk data includes the total risk score, damage risk and leakage risk of the pipeline, and these data are transmitted to the risk monitoring system. The risk monitoring system uses an alarm mechanism to issue an early warning when the pipeline risk exceeds a preset threshold. The specific threshold is set to trigger an early warning when the total risk score exceeds 0.75, so as to adjust and repair the pipeline status in real time.

[0033] Preferably, step S1 specifically comprises:

[0034] Step S11: start the air compressor and set the compression pressure value to 0.6-0.8 MPa; use the air compressor to inhale a preset amount of normal pressure air, perform a compression operation to generate compressed air, and transport the compressed air to a preset pipeline to obtain pipeline compressed air data;

[0035] In this embodiment, the air compressor is started and the compression pressure value is set to 0.7MPa. After the air compressor is started, it will inhale normal pressure air from the external environment, usually at room temperature (about 25°C), and compress the air to the set pressure value of 0.7MPa through the compression process of the compressor. The compression volume of the air compressor is set to 20m³ / h to ensure that the flow rate can meet the air supply required by the pipeline. After the compression operation is completed, the compressed air is delivered to the preset pipeline through a steel pipe with a diameter of 50mm and a pipeline length of 30 meters. The pressure sensor in the pipeline is used to monitor the pressure in the pipeline in real time to ensure that the delivery pressure of the compressed air is maintained at about 0.7MPa, with an error range of ±0.02MPa. The air flow rate in the pipeline is monitored by a flow meter, and the flow value is set to 5000L per minute to ensure the stability of the air supply. All measurement data are recorded and summarized to form pipeline compressed air data, including parameters such as pressure, flow, and temperature, as input data for subsequent steps.

[0036] Step S12: Identify the pipeline transportation path according to the pipeline compressed air data;

[0037] In this embodiment, based on the pipeline compressed air data, by analyzing the air flow velocity, pressure, and temperature data inside the pipeline, a physical model is used to calculate the pipeline transportation path. This path identification process is based on the pressure gradient and flow rate changes inside the pipeline, and fluid dynamics simulation technology (such as CFD software) is adopted to establish a flow field model of the pipeline. Data fitting is performed on the data at the measurement points to simulate the flow direction and velocity of the air flow inside the pipeline, and accurately identify the flow path of the air in the pipeline. The model inputs include the geometric parameters of the pipeline (such as pipeline diameter, length, bending angle), the physical properties of the fluid (such as the density and viscosity of air), and the data such as the temperature, pressure, and flow rate of the compressed air obtained previously. Based on these inputs, the air flow direction and pressure distribution of each pipe segment are simulated and calculated, the transportation path diagram of the pipeline and the transportation capacity of each pipe segment are output, the air flow obstruction areas are marked, and the complete pipeline transportation path data is obtained.

[0038] Step S13: Detect the air impurity components according to the pipeline compressed air data;

[0039] In this embodiment, according to the pipeline compressed air data, a gas analyzer is used to detect the impurity components in the air. The gas analyzer adopts infrared spectroscopy analysis method, which can accurately detect the gas components existing in the air, such as carbon dioxide, nitrogen oxides, sulfides, etc. The detection interval for each time is 5 seconds to ensure the real-time nature of the detection data. The sensitivity of the device is set to 0.01 ppm to ensure the sensitivity to low-concentration impurities. According to the gas concentration in the air, a standardized method is used to evaluate the composition of the air impurities. A concentration threshold is set. For example, when the carbon dioxide concentration exceeds 400 ppm, the air is considered to contain higher impurities. The concentration of particulate matter in the air is measured by the light scattering method. The detection range of the particulate matter concentration is set to 0.1 - 100 μg / m³, and the detection range of the particulate matter size is 0.3 μm - 10 μm. After the detection of the impurity components is completed, the types and concentration data of the impurities are recorded and used as the basis for setting the subsequent air purification targets.

[0040] Step S14: Set the air purification target based on the air impurity components;

[0041] In this embodiment, the purification targets include the maximum allowable concentrations of specific impurity components (such as carbon dioxide, nitrogen oxides, particulate matter, etc.). According to actual requirements, it is set that the carbon dioxide concentration in the air shall not exceed 350 ppm, the particulate matter concentration shall not exceed 5 μg / m³, and the nitrogen oxide concentration shall not exceed 10 ppm. When setting these targets, national or industry standards are referred to, such as the ISO 8573-1 standard, which stipulates the maximum allowable concentrations of various impurity components in compressed air. The setting of the purification targets also takes into account the usage environment of the pipeline and the air quality requirements. For example, in a closed environment, the oxygen concentration in the air must be maintained within the range of 21% ± 0.5%, so the oxygen concentration is set not to be lower than 20.5%. These set values become the targets for the subsequent air purification process to ensure that the quality of the compressed air meets the requirements.

[0042] Step S15: Construct an air purification model according to the pipeline transportation path and the air purification targets;

[0043] In this embodiment, the construction process of this model first considers the geometric shape of the pipeline and the way of air flow, and sets the initial conditions such as the air flow rate, temperature, humidity, etc. inside the pipeline. Based on these conditions, a suitable air purification method is selected, such as using high-efficiency filters, condensation separators, drying devices, etc. According to the air flow path, the model calculates the purification requirements for each pipe section, and determines the type and parameters of the purification devices to be installed in each area. For example, the filtration efficiency of the filter (95% - 99%) and the particulate matter capture ability (≥ 0.3 μm). At the same time, the model associates the air purification devices with the pipeline path to ensure that the purification process covers all air flow paths. The model simulates the air flow process through different purification devices by numerical simulation methods (such as CFD simulation), evaluates the purification effect, and further adjusts the configuration and parameters of the purification devices to meet the set air purification targets.

[0044] Step S16: Conduct air purification simulation according to the air purification model to obtain purified air data.

[0045] In this embodiment, the simulation process first conducts numerical simulation by inputting the initial air data inside the pipeline (including flow rate, temperature, humidity, impurity component concentration, etc.) and the air purification targets. During the simulation process, the temperature change, humidity change, and pollutant removal efficiency when the air passes through each part of the pipeline are set. Using CFD simulation software, the calculation conditions are set as a time step of 1 second and a simulation period of 60 minutes to simulate the purification efficiency of the air flow between different purification devices. The efficiency parameters of each purification device are set according to experimental data. For example, the particulate matter removal efficiency of the activated carbon filter is 98%, and the moisture removal efficiency of the condensation separator is 95%. The simulation results output the purified air data, including the purified humidity, temperature, pollutant concentration, etc. These data will provide a basis for the subsequent pipeline risk assessment.

[0046] Preferably, step S16 is specifically as follows:

[0047] Step S161: Import the air purification model into the simulation software;

[0048] In this embodiment, a CFD (Computational Fluid Dynamics) software with air flow and purification simulation functions is selected, such as ANSYS Fluent or COMSOL Multiphysics. The detailed parameters of the air purification model established according to step S15 will be set. Each component in the model (such as the condensation unit, adsorption unit, filtration unit, etc.) will be input into the simulation software according to its design specifications to ensure that each component exhibits physical characteristics consistent with the actual environment during the simulation. The parameters of these components, such as the efficiency of the filter and the selection of the adsorption material, need to be accurately imported to ensure that the subsequent simulation results reflect the real air purification process. Ensure that the software can effectively read and establish the required three-dimensional pipeline model and set appropriate boundary conditions for a comprehensive simulation of air flow and purification effects.

[0049] Step S162: Set the input compressed air volume flow rate in the simulation software to 1 - 20 m³ / min and the moisture content to 0.01 - 0.2 kg / kg;

[0050] In this embodiment, the input compressed air volume flow rate and moisture content are set through the simulation software. Specifically, the input compressed air volume flow rate is set to 1 to 20 m³ / min to reflect the air flow range in the pipeline. In the simulation, a flow control device is used to ensure that the air flow meets the set range, usually adjusted according to the actual requirements of the pipeline. The moisture content of the compressed air is set to 0.01 - 0.2 kg / kg, and this value is calculated based on the environmental humidity of the compressed air and the exhaust temperature of the compressor. By setting a humidity sensor, the moisture content of the air entering the pipeline is monitored in real time. This moisture content determines the processing capacity of the subsequent condensation unit to ensure that the influence of moisture on the purification process can be accurately simulated during the simulation. After setting these parameters in the software, the changes in air flow rate and moisture content during the simulation will affect the purification effects of the condensation, adsorption, and filtration units.

[0051] Step S163: Set the temperature of the condensation unit in the simulation software to 3 - 10 °C, the material of the adsorption unit to activated alumina, and the particle size retention range of the filtration unit to 0.1 - 10 μm;

[0052] In this embodiment, specific parameters of the condensation unit, adsorption unit, and filtration unit are set in the simulation software. The temperature of the condensation unit is set to 3 to 10 °C. This temperature range is based on the actual requirements of air condensation and is used to remove moisture from compressed air. By adjusting the temperature of the cooling system of the condensation device, this setting is ensured to be achieved. The material of the adsorption unit is set to activated alumina, which is an efficient adsorption material used to adsorb moisture, oil, and other pollutants in the air. In the simulation, the adsorption capacity of this material is set to 5 g / kg (the maximum amount of adsorbed moisture), and the adsorption efficiency is 90%. The particle size retention range of the filtration unit is set to 0.1 - 10 μm. Such a setting can ensure that fine particulate matter in the air is effectively filtered. In the simulation software, calculations are performed based on the specific dimensions of the filtration unit (such as the filter mesh pore size, surface area, etc.), and particles with a size larger than the set value are simulated to be captured, ensuring the high efficiency of air purification.

[0053] Step S164: Set the environmental temperature to 25–45 °C, the system operating pressure to 0.7–1.0 MPa, and the relative humidity to 30%–90% in the simulation software;

[0054] In this embodiment, the environmental temperature is set to 25 to 45 °C, and this range reflects the environmental temperature of the air in actual operation. The operating pressure of the system is set to 0.7 to 1.0 MPa to simulate the influence of different pressures on the purification effect. The pressure value is set through the boundary conditions in the software, and the pressure input value is adjusted according to the working environment of the pipeline. Usually, 0.7 MPa is the common value. The relative humidity is set to 30% to 90%. The humidity change within this range has a significant impact on the moisture content in the air and the condensation efficiency. Through the humidity control system, the environmental humidity value is set to ensure that the moisture in the air can be effectively removed in a high-humidity environment. All these parameters will affect all aspects of the air purification process during the simulation, ensuring that the simulation results are consistent with the actual operation.

[0055] Step S165: Run the air purification simulation program in the simulation software and output the purified air data.

[0056] In this embodiment, according to all the aforementioned set parameters (including air flow rate, moisture content, condensation unit temperature, adsorption unit material, filtration unit particle size retention, ambient temperature, system pressure, humidity, etc.), the simulation program is started to simulate the processes of air flow, purification, and pollutant removal in the pipeline. The simulation software will calculate the air purification efficiency within the set time and gradually output the performance data of each component, such as the moisture removal rate of the condensation unit, the adsorption capacity of the adsorption unit, the particulate matter removal rate of the filtration unit, etc. After the simulation is completed, the output purified air data includes the purified air flow rate, humidity, pollutant concentration, and particulate matter quantity, etc., and all data are within the set threshold range. Through these data, it is possible to evaluate whether the purification system meets the air quality standard and provide a necessary basis for subsequent pipeline risk determination.

[0057] Preferably, the detection of residual condensate water in step S2 includes:

[0058] Extracting the humidity data of the purified air data;

[0059] In this embodiment, the humidity data in the purified air data is obtained. By using a humidity sensor installed in the air pipeline to detect the purified air in real time, the humidity value of the air is obtained. Humidity sensors usually use capacitive or resistive sensors, and the change of water vapor in the air affects the capacitance or resistance of the sensor, and then outputs the humidity value. The measurement range of the humidity value is usually 0% to 100% relative humidity (RH), and its accuracy needs to be adjusted according to the operating environment. The humidity data needs to be accurate to two decimal places for subsequent calculations and analyses, and usually the set accuracy is ±0.1%RH. This humidity data will be used to further calculate the dew point temperature and determine whether the system meets the air purification target.

[0060] Determining the dew point temperature based on the humidity data;

[0061] In this embodiment, according to the obtained humidity data, the dew point temperature calculation formula is used to determine the dew point temperature. The calculation of the dew point temperature depends on the temperature and humidity data, and the standard dew point calculation formula for moist air is used for calculation. The dew point temperature calculation formula is usually:

[0062] ;

[0063] Wherein, is the dew point temperature, is the ambient temperature and RH is the relative humidity. The ambient temperature is generally obtained from an ambient temperature sensor, and its range is usually set from 25°C to 45°C. Assuming the ambient temperature of the system is 30°C and the relative humidity is 60%, the dew point temperature can be calculated by the above formula. For example, when the ambient temperature is 30°C and the humidity is 60%, the dew point temperature is approximately 22.5°C. According to this calculation formula, this dew point calculation formula can be set in the simulation software and the change of the dew point can be monitored in real time.

[0064] Calculate the dew point deviation value based on the preset target dew point value and the dew point temperature;

[0065] In this embodiment, it is necessary to calculate the dew point deviation value according to the preset target dew point value and the actually measured dew point temperature. The target dew point temperature value is usually set as the design standard value of the system, for example, set as 20°C. According to the system requirements, the calculation formula of the dew point deviation value is:

[0066] ;

[0067] Where, is the dew point deviation value, is the actually measured dew point temperature, is the preset target dew point temperature. Taking the previously calculated actual dew point temperature of 22.5°C and the target dew point temperature of 20°C, the calculated dew point deviation value is 2.5°C. In the simulation software, this calculation can be implemented through built-in functions, monitoring the dew point deviation value in real time and judging whether it exceeds the set tolerance range.

[0068] Record the dew point deviation period according to the dew point deviation value, collect the purified air in the dew point deviation period, detect the residual condensate water by using a condensate water sensor, and calculate the accumulated volume of the residual condensate water to obtain the residual condensate water data.

[0069] In this embodiment, when the dew point deviation value exceeds the set threshold, it is necessary to record the dew point deviation period. The set threshold is 0.5°C. When the dew point deviation value exceeds this value, the system will automatically record the start and end times, marked as the dew point deviation period. The air data during these periods is collected and stored for subsequent analysis. The collection of air data includes parameters such as humidity, temperature, and flow rate. In the system, the data acquisition system will monitor the humidity and temperature changes in real time and record them in the database, marked as the data during the dew point deviation period. A condensate sensor installed in the pipeline is used to detect the residual condensate in the air. Condensate sensors usually use sensor probes to detect the moisture condensed in the air due to temperature changes. These sensors work based on capacitive or resistive principles. When condensate accumulates on the sensor surface, the electrical properties of the sensor change, and then a signal is transmitted. By setting the water accumulation threshold of the condensate sensor, the recording function is automatically activated when the condensate exceeds the set value. The accumulated volume of condensate is calculated by the following formula:

[0070] ;

[0071] where, is the accumulated volume of condensate, is the area measured by the condensate sensor, is the height of the water. The volume data of the accumulated water is usually obtained through real-time monitoring and accumulation and recorded by time period. Multiple condensate sensors are set at key positions of the pipeline to ensure the accuracy and real-time nature of the data. According to the above-mentioned calculation method of the water accumulation of the condensate sensor, the data of the residual condensate is obtained. This data includes information such as the volume of the residual condensate, the time period of the water accumulation, and the temperature of the condensate. This data will be used for subsequent pipeline corrosion and risk determination analysis. The data is stored in the condensate sensor system and reported to the monitoring system periodically for further analysis.

[0072] Preferably, the pipeline corrosion identification in step S2 includes:

[0073] Marking the position of the condensate pipeline based on the residual condensate data;

[0074] In this embodiment, multiple condensate sensors are installed inside the pipeline to monitor the humidity and moisture data in the pipeline in real time, ensuring that the moisture status of each part of the pipeline can be comprehensively and accurately obtained. The accuracy of each sensor reaches ±0.01 L, and it can meticulously record the accumulation of condensate. To ensure the comprehensiveness and accuracy of data collection, the sensors are evenly distributed in the pipeline, covering different monitoring nodes. The sensors generate a humidity distribution map of the pipeline path by monitoring the humidity changes at each node in the pipeline in real time. These data are processed by algorithms to calculate the accumulation amount of condensate and determine the areas where moisture is relatively concentrated in the pipeline according to the humidity change trend. Usually, these areas are the parts with higher humidity, indicating the accumulation positions of condensate. Through the data processing algorithm, these high-humidity areas are accurately identified and extracted, thereby marking the specific pipeline positions of the condensate. These position marking data form a marking map of the condensate pipeline position, which serves as an important reference for subsequent pipeline corrosion analysis, condensate accumulation area monitoring, and pipeline risk assessment. These accurate markings provide accurate data support for condensate analysis and help to detect potential corrosion risk areas in advance.

[0075] Irradiate the position of the condensate pipeline with high-energy X-rays and generate X-ray fluorescence spectra;

[0076] In this embodiment, according to the analysis of the condensate data in the early stage, the positions where condensate accumulates in the pipeline are identified. Subsequently, high-energy X-rays are used to irradiate these marked areas for in-depth analysis of the internal state of the pipeline. To ensure that the pipeline wall can be penetrated and effective imaging information can be obtained, the energy of the X-rays is usually set between 50 and 100 keV. This energy range can not only ensure sufficient penetration but also effectively avoid damaging the pipeline material. The X-ray source used usually irradiates the condensate accumulation area directionally to ensure that the entire target area is covered. During the irradiation process, the X-ray detector installed outside the pipeline captures the photons passing through the pipeline in real time, and then generates the fluorescence spectrum on the pipeline surface. The peak information in the fluorescence spectrum reflects the elemental composition of the metal surface inside the pipeline. By comparing with the standard elemental spectrum diagram, the types and contents of different metal elements in the pipeline can be determined. The generated spectral data are transmitted to the computer system in real time and interpreted through peak analysis technology to extract accurate elemental information. These data provide basic support for subsequent corrosion analysis and risk assessment, and help to identify whether there are metal damages, corrosion, or other abnormalities on the pipeline surface. During the whole process, the irradiation intensity and time of the X-rays are strictly controlled to ensure that no adverse effects are produced on the pipeline and the surrounding environment, and at the same time, the analysis accuracy is maximally improved.

[0077] Speculate on the types of metal elements on the pipeline surface based on the X-ray fluorescence spectra to obtain metal element data;

[0078] In this embodiment, by analyzing each spectral peak in the X-ray fluorescence spectrum, the types and concentrations of metal elements on the pipeline surface can be inferred. Each peak in the X-ray fluorescence spectrum represents the characteristic fluorescence signal of a specific element. For example, the characteristic peak of iron is usually near 6.4 keV, and the characteristic peak of copper is around 8.0 keV. To accurately infer the composition of the metal on the pipeline surface, it is first necessary to perform precise spectral peak analysis on the acquired spectral data, identify the characteristic peaks of each element, and match them with the known standard spectral peaks. This process relies on the high resolution of the spectrum to distinguish between adjacent elemental spectral peaks. By comparing the spectral peaks with the standard characteristic peaks of elements in the database, the main metal components on the pipeline surface can be identified, and then the type and concentration of the metal can be inferred. The analysis accuracy of each element can be controlled within 0.1%, which ensures the high accuracy of the inferred metal element information. The analysis results not only provide the necessary elemental data for subsequent corrosion reaction simulations but also help evaluate the corrosion resistance of the pipeline material, thus providing a reliable theoretical basis for the long-term use of the pipeline. During the whole process, interference peaks also need to be corrected to ensure the accuracy and consistency of the final results.

[0079] Based on the metal element data and the residual condensate data, an electrochemical reaction simulation is carried out to generate current data;

[0080] In this embodiment, by combining the metal element data with the residual condensate data, an electrochemical reaction simulation is performed to generate current data. First, the metal element data obtained from the X-ray fluorescence spectrum analysis, including the type, concentration, and distribution of the metal, are used as input data. These metal element data, together with the environmental data such as humidity, temperature, pH value, and ion concentration in the condensate water, participate in the electrochemical reaction simulation. The simulation process is based on electrochemical kinetic models, which describe in detail the interactions between metals and ions in water (such as hydrogen ions, chloride ions, etc.) and their corrosion behaviors. Specifically, the metal corrosion rate in the electrochemical reaction process is closely related to the current intensity. Therefore, it is necessary to calculate the current data to reflect this rate. The input environmental data have an important impact on the reaction results. For example, an increase in temperature accelerates the reaction rate, a change in pH value affects the solubility of the metal, and humidity directly affects the moisture content and corrosion rate on the metal surface. In the simulation, parameters such as electrolyte concentration and temperature need to be determined through precise sensor data acquisition to ensure the accuracy of the input data. The electrochemical simulation uses specific kinetic models, such as the Nernst equation and the Tafel equation, for calculation and solution, and finally outputs the current intensity data related to metal corrosion. These current data reflect the changes in the reaction rate on the metal surface. Through these current data, important basic data can be provided for subsequent oxygen reduction reaction simulations and the corrosion behavior of the pipeline under different environmental conditions can be analyzed in depth to help predict the service life and corrosion risk of the pipeline.

[0081] Based on the current data and metal element data, an oxygen reduction reaction is simulated to obtain oxygen reduction data; based on the oxygen reduction data, rust accumulation analysis is carried out to obtain rust accumulation data;

[0082] In this embodiment, the current data provides the rate information of the metal corrosion reaction, while the metal element data provides the specific type, concentration and distribution of the metal, all of which have an important impact on the oxidation-reduction reaction. In the oxygen reduction reaction model, the metal surface reacts with oxygen molecules in water to produce an oxidation-reduction reaction. During this process, factors such as temperature, humidity, and the state of the metal surface (such as whether an oxide film is formed) will significantly affect the progress of the reaction. Therefore, environmental factors such as temperature and humidity need to be accurately collected through real-time monitoring data to ensure the accuracy of the model input. During the simulation, the relationship between the current density and the reaction rate is considered in detail. The current density reflects the corrosion rate of the metal surface. By calculating the rate of each reaction step, the model obtains the overall result of the oxidation-reduction reaction, including the reaction rate and the generation of products. The products of the oxygen reduction reaction are mainly metal oxides or rust, and the generation of these products affects the corrosion depth of the metal and the metal loss rate. During this process, the simulation not only calculates the oxidation rate of the metal, but also obtains the overall trend of metal corrosion through the accumulation of reaction products. The obtained oxygen reduction data provides an important basis for the subsequent rust accumulation analysis. By analyzing the accumulation of reaction products, the accumulation of rust on the pipeline surface can be inferred, including the thickness and distribution of the rust. Finally, the obtained rust accumulation data provides accurate quantitative data for further evaluating the corrosion degree of the pipeline and predicting its service life. This process provides a theoretical basis for the accurate simulation and evaluation of pipeline corrosion, helping to formulate reasonable pipeline maintenance and protection strategies.

[0083] Detect the pipeline corrosion area at the position of the condensate pipeline according to the rust accumulation data to obtain pipeline corrosion data.

[0084] In this embodiment, according to the results of rust accumulation analysis, the corrosion area of the pipeline is detected by setting a threshold for corrosion accumulation. The setting of the threshold is based on the corrosion resistance standard of the pipeline material and factors such as temperature and humidity in the working environment. Taking a carbon steel pipeline as an example, the usually set corrosion accumulation threshold is 0.2 grams per square centimeter. If the rust accumulation amount in a certain area exceeds this value, then this area is marked as a corrosion risk area. According to these thresholds, first, the accumulation data calculated by the corrosion model is combined with the sensor data (such as temperature, humidity, etc.) inside and outside the pipeline to identify the areas where the accumulation amount exceeds the set threshold. The corrosion degree of each area is usually expressed as a percentage. For example, if the corrosion accumulation amount in a certain area is 0.4 grams per square centimeter, it means that the corrosion degree of this area is 50%, that is, the corroded metal area accounts for half of the surface area of this area. Further, through the layout diagram of the pipeline, these corrosion areas are spatially marked and the impact on the pipeline strength and safety is analyzed. For the pipeline part that has been marked as a corrosion area, using the design standard of the pipeline, calculate whether its current strength meets the safety requirements. For example, if the wall thickness of the pipeline is reduced to less than 70% of the original design thickness, it will result in insufficient load-bearing capacity of the pipeline and a risk of leakage. The final corrosion data is usually expressed in percentage form, indicating the corrosion degree of each area of the pipeline, and further predicting the remaining service life of these areas. Through this process, an accurate basis can be provided for the maintenance and protection measures of the pipeline, such as strengthening or replacing the most severely corroded areas, so as to reduce the risk of pipeline failure.

[0085] Preferably, the prediction of the pipeline perforation probability in step S2 includes:

[0086] Calculating the corrosion rate according to the pipeline corrosion data;

[0087] In this embodiment, the calculation of the corrosion rate mainly relies on the pipeline corrosion data collected by corrosion sensors. These sensors are usually installed at key positions of the pipeline, such as elbows, joints or high-risk areas of the pipeline, to regularly record the corrosion degree of the pipeline. The corrosion rate is usually expressed in mm / year, and the calculation method is to compare the wall thickness change of the pipeline within a specific time interval. For example, by comparing the wall thickness of a certain section of the pipeline in the initial state with the wall thickness after use within a certain period of time (such as 1 year), the speed of metal loss is calculated. If the initial wall thickness of a certain section of the pipeline is 10 mm and the wall thickness is reduced to 9 mm after 1 year of use, the calculated corrosion rate is 1 mm / year. In addition, environmental factors such as environmental temperature, humidity, and fluid characteristics (such as flow rate, fluid composition) need to be considered when calculating the corrosion rate, as these factors have a significant impact on the metal corrosion rate. Therefore, accurately obtaining and integrating these environmental data can improve the calculation accuracy of the corrosion rate. The changes in surface roughness and corrosion accumulation amount also need to be used as reference factors for calculation, because these data can reflect the metal loss situation on the pipeline surface, thus helping to calculate a more accurate corrosion rate. Through these detailed analysis methods, an accurate corrosion rate can be obtained, providing an important basis for subsequent pipeline maintenance and risk assessment.

[0088] Identify the corrosion area on the inner wall of the pipeline according to the pipeline corrosion data;

[0089] In this embodiment, after obtaining the pipeline corrosion data, the corrosion accumulation amount on the pipeline surface is obtained by sensors real-time monitoring the metal loss situation on the pipeline surface and inside. These sensors usually record data such as metal loss, surface roughness, and corrosion accumulation amount during the corrosion process, and these data can help evaluate the corrosion degree of the pipeline. To effectively identify the corrosion area on the inner wall of the pipeline, a corrosion threshold is set, and this threshold is usually 5% of the metal loss. If the metal loss in a certain pipeline area exceeds this threshold, it indicates that obvious corrosion has occurred in this area and has reached the level that requires attention. The specific corrosion data is usually expressed as a percentage, indicating the corrosion depth of this area. For example, if the initial wall thickness of a certain section of the pipeline is 6 mm and the corrosion depth of this section of the pipeline is 0.3 mm, then the corrosion degree of this section of the pipeline is 5% (i.e., 0.3 mm / 6 mm), and this corrosion degree reaches the set threshold, identifying this section of the pipeline as a corrosion area. By combining these corrosion data with the calculated corrosion rate, a comprehensive evaluation can be carried out for each area of the pipeline, and then the pipeline areas with serious corrosion can be identified, providing an effective basis for maintenance and detection. The accuracy of corrosion identification depends on the comprehensive analysis of sensor data, environmental factors, and corrosion rate to ensure timely discovery of pipeline corrosion problems and take necessary protective measures.

[0090] Use a laser confocal microscope to perform microscopic three-dimensional topography imaging on the corrosion area of the inner wall of the pipeline to obtain a three-dimensional topography image of the inner wall of the pipeline;

[0091] In this embodiment, the laser confocal microscope uses laser scanning technology. By precisely irradiating the corrosion area of the inner wall of the pipeline with a light beam, it captures the light signals scattered back from the surface, thereby generating detailed microscopic three-dimensional images. To ensure the accuracy of imaging, first, it is necessary to ensure that the surface of the inner wall of the pipeline is clean, without any impurities or contaminants, because uneven surfaces or the presence of contaminants affect the imaging quality. During the imaging process, the laser beam gradually scans the corrosion area and records the light signals reflected or scattered from the surface. With the high resolution of the laser confocal microscope (usually 0.1μm), it can accurately capture the minute structural changes on the inner wall of the pipeline, including corrosion pits, surface protrusions, metal spalling, and other subtle surface morphological changes. These high-precision three-dimensional topography images provide detailed information on the corrosion area of the pipeline, providing reliable basic data for subsequent corrosion assessment and maintenance decision-making. Through this technology, it is possible to gain an in-depth understanding of the microscopic corrosion situation on the inner wall of the pipeline, helping researchers and engineers accurately evaluate the scope, depth, and development trend of corrosion, and thus conduct more effective risk prediction and protective measure planning.

[0092] Calculate the Gaussian curvature based on the three-dimensional topography image of the inner wall of the pipeline; Mark the bulging area of the pipe wall in the three-dimensional topography image of the inner wall of the pipeline based on the Gaussian curvature;

[0093] In this embodiment, the curvature value of each point is extracted from the three-dimensional image, and then the Gaussian curvature on the entire surface is calculated. Specifically, for each corrosion area, the Gaussian curvature value of the local area is obtained by mathematically processing its three-dimensional coordinate points. The calculation process of the Gaussian curvature requires the use of digital image processing tools or specialized graphic calculation software, such as Matlab or CAD software. The calculation results are used to identify the protruding or concave areas on the inner wall of the pipeline and provide a basis for subsequent marking of the bulging area of the pipe wall. Based on the calculated Gaussian curvature value, mark the bulging area of the pipe wall in the three-dimensional topography image of the inner wall of the pipeline. The bulging area usually shows an area with a relatively large local curvature value, indicating that this part of the pipe wall has expanded due to corrosion or material deterioration. According to the calculation results of the Gaussian curvature, set a threshold. If the curvature value of a certain area exceeds this threshold, it indicates that this area is the bulging area of the pipe wall. For example, if the Gaussian curvature is greater than 0.5mm^-1, then this area is considered the bulging area. Through data processing in the three-dimensional image, these areas can be accurately marked as a basis for further evaluating the stability of the pipeline and conducting repairs.

[0094] Identify the minimum wall thickness of the pipe wall in the bulging area of the pipe wall; Continuously monitor the wall thickness reduction rate in the bulging area of the pipe wall;

[0095] In this embodiment, the thinnest part of each bulge area is determined by using the thickness data of the pipe wall in the three-dimensional topography image. Generally, the identification of the minimum wall thickness requires considering the actual size of the pipeline and the corrosion mode. For example, the depth of the corrosion area is related to factors such as the type and flow rate of the fluid inside the pipeline. By scanning each point on the inner wall of the pipeline, the thinnest part is identified and its thickness data is recorded. This value is crucial for evaluating whether there is a risk of pipeline rupture. The value of the minimum wall thickness of the pipe wall is usually accurate to 0.1 mm after the decimal point. For example, if the minimum wall thickness of a certain bulge area is 3.5 mm, it indicates the weakness degree of this area. By installing regular monitoring equipment, the wall thickness reduction rate of the bulge area on the pipe wall is continuously monitored. This process requires installing wall thickness sensors, which should regularly record the wall thickness changes in the bulge area and calculate the change amount of the wall thickness per unit time. The wall thickness reduction rate is usually in the unit of mm / year, and the calculation method is to track the thickness changes of the inner wall of the pipeline based on the sensor data. For example, if the wall thickness of a certain bulge area decreases by 0.2 mm within 6 months, the wall thickness reduction rate is 0.4 mm / year. This rate is of great significance for evaluating the aging degree of the pipeline and future corrosion risks.

[0096] Predict the probability of pipeline perforation based on the minimum wall thickness of the pipe wall and the wall thickness reduction rate.

[0097] In this embodiment, a pipeline perforation risk assessment model is established by combining the physical properties of the pipeline material, operating conditions, and corrosion data accumulated during long-term operation. This model mainly focuses on two key factors: the minimum wall thickness of the pipeline and the wall thickness reduction rate. In practical applications, during the long-term use of the pipeline, the metal wall thickness gradually decreases due to factors such as corrosion and friction. By monitoring the minimum wall thickness and wall thickness reduction rate of the pipeline, the remaining service life and perforation risk of the pipeline can be effectively evaluated. The model conducts a preliminary assessment based on the minimum wall thickness of the pipeline. The minimum wall thickness is usually a key indicator for the pipeline to withstand internal and external pressures. If the minimum wall thickness of the pipeline is lower than the specified safety standard (such as 3 mm), even if the wall thickness of other areas is normal, there is a potential risk of perforation. Then, the model also needs to consider the wall thickness reduction rate, which is data obtained by continuously monitoring the corrosion process of the pipeline over a certain period of time. If the wall thickness reduction rate of the pipeline exceeds the set threshold (such as 0.5 mm / year), it indicates that the corrosion process accelerates, and the probability of pipeline perforation also increases. By combining these factors with comprehensive factors such as the compressive strength of the pipeline material, temperature and humidity of the working environment, and fluid characteristics, the model can calculate the probability of pipeline perforation. During this process, the calculated perforation probability can not only reflect the current risk status of the pipeline but also predict the possibility of pipeline perforation in the future for a period of time, thus providing a scientific basis for pipeline maintenance, repair, and replacement. Through this method, pipeline safety management can be carried out more accurately, and preventive measures can be taken before potential problems occur, greatly improving the operating safety of the pipeline.

[0098] Preferably, step S3 is specifically as follows:

[0099] Step S31: Extract the particulate matter concentration according to the purified air data;

[0100] In this embodiment, when analyzing the purified air data, the particulate matter concentration data is collected by an air quality monitoring device. The particulate matter concentration is usually expressed in μg / m³. Monitoring devices such as PM2.5 sensors and PM10 sensors will regularly record the changes in the particulate matter concentration. These devices are usually installed near the pipeline and can measure the content of fine particulate matter (such as PM2.5, PM10, etc.) in the air in real time. The data acquisition system records the collected data according to the timestamp to ensure the timeliness of the concentration data. According to the collected data, the particulate matter concentration value at each time point is extracted. For example, if at a certain time point, the PM2.5 concentration is 50 μg / m³ and the PM10 concentration is 100 μg / m³, then these concentration values will be stored and used for the analysis of subsequent steps.

[0101] Step S32: Extract the concentration measurement timestamp based on the particulate matter concentration; sort the concentration measurement timestamps by time and construct a concentration time series; identify the abnormal change trend of the concentration time series; extract the abnormal concentration change time based on the abnormal change trend;

[0102] In this embodiment, based on the collected particulate matter concentration data, the timestamps corresponding to the concentration data are first extracted. These timestamps represent the measurement time of the particulate matter concentration, and the unit is usually seconds or minutes. By sorting the concentration measurement timestamps by time, the time series data of the particulate matter concentration is constructed. Next, by performing trend analysis on the changes in the concentration time series, potential abnormal fluctuations are identified. The identification of the abnormal change trend can be based on a preset threshold. For example, when the concentration fluctuation exceeds ±10%, the fluctuation is considered abnormal. The abnormal concentration change time refers to the time period during which the abnormal fluctuation occurs. By comparing with the reference concentration data, the specific change time is determined. For example, if the particulate matter concentration suddenly increases from 50 μg / m³ to 150 μg / m³ within a certain period of time and the change lasts for more than 1 hour, then this change time period can be marked as the abnormal concentration change time.

[0103] Step S33: Take a pipeline image based on the abnormal concentration change time; perform color space conversion on the pipeline image to obtain a color space converted pipeline image; identify the solid particle area of the color space converted pipeline image based on a preset solid particle color range, and measure the thickness of the solid particles in the solid particle area using a laser rangefinder to obtain the solid particle thickness;

[0104] In this embodiment, after identifying the abnormal concentration change time period, a pipeline image is captured to further analyze the accumulation of solid particles. When capturing the pipeline image, a high-resolution camera is used to ensure the image quality, and the shooting conditions need to ensure uniform light to reduce the impact of shadows on image analysis. Next, the color space of the captured pipeline image is transformed, usually from the RGB color space to the HSV (hue, saturation, value) color space. After the transformation, the image is processed using a preset color range of solid particles to identify the area of solid particles in the image. This color range is set based on experimental data. For example, the color of the solid particles is within a specific interval in the HSV color space, such as the hue is 10 - 30, the saturation is 0.4 - 0.7, and the value is 0.2 - 0.6. Finally, a laser rangefinder is used to measure the thickness of the identified solid particle area. The laser rangefinder can accurately obtain the thickness value of the solid particles by emitting laser light and measuring the reflection time, and the measurement accuracy usually reaches 0.1 mm. Through this process, the thickness data of the solid particles is obtained, which serves as the basis for subsequent analysis.

[0105] Step S34: Perform three-dimensional reconstruction on the solid particle area to obtain the three-dimensional structure of the solid particles, and measure the volume of the solid particles in the three-dimensional structure of the solid particles;

[0106] In this embodiment, after obtaining the thickness data of the solid particle area, three-dimensional reconstruction technology is used to reconstruct the solid particle area. A laser scanning device or a stereo vision system is used to capture multiple images from different angles, and combined with the thickness data, computer vision technology is used to generate a three-dimensional model of the solid particles. During the three-dimensional reconstruction process, it is necessary to ensure a stable distance between the scanning device and the pipeline surface to obtain accurate depth information. Through three-dimensional reconstruction software, the multi-angle images and thickness data are synthesized to obtain the three-dimensional structure of the solid particles. After the reconstruction is completed, a volume measurement tool is used to calculate the volume of the solid particles. The volume is usually achieved by segmenting each layer in the three-dimensional structure, and the unit is cubic millimeters (mm³). This volume data provides a quantitative basis for subsequent accumulation judgment.

[0107] Step S35: Perform solid particle accumulation judgment based on the thickness of the solid particles and the volume of the solid particles to obtain the accumulated solid particle data;

[0108] In this embodiment, an accumulation standard is set and analyzed by comparing the volume of the solid particles with the set threshold. The threshold can be set according to the working environment of the pipeline and the properties of the particulate matter. For example, when the volume of the solid particles in a certain area exceeds 1000 mm³ and the thickness exceeds 5 mm, it is considered that solid particle accumulation has occurred in this area. By counting the accumulation data of all areas, the solid particle accumulation situation in the overall pipeline can be obtained. These data will serve as the basis for subsequent pipeline blockage analysis.

[0109] Step S36: Conduct pipeline blockage analysis based on the accumulated solid particle data and record the pipeline blockage data.

[0110] In this embodiment, a blockage threshold is set, and the maximum accumulation amount of solid particles is set according to the design standard and usage environment of the pipeline. For example, if the accumulated volume of solid particles in a certain section of the pipeline exceeds 5000 mm³, then this section of the pipeline is considered to have a blockage risk. In the pipeline blockage analysis, combined with the data of the accumulated solid particles, the blockage risk level of the pipeline is further evaluated, which is divided into three levels: high, medium, and low. The setting of the blockage risk level is based on the comparison between the actually measured accumulation amount and the set standard. For example, if the accumulation amount is greater than 5000 mm³, it is a high risk; if the accumulation amount is between 1000 mm³ and 5000 mm³, it is a medium risk; if the accumulation amount is less than 1000 mm³, it is a low risk. Finally, record the pipeline blockage data, including the accumulation amount, blockage risk level, and relevant time information, to provide data support for pipeline maintenance and repair.

[0111] Preferably, step S36 is specifically as follows:

[0112] Step S361: Mark the positions of solid particle accumulation in the pipeline according to the accumulated solid particle data;

[0113] In this embodiment, the coordinate positioning method is adopted to perform spatial mapping of the spatial coordinate information of the solid particles in the three-dimensional reconstruction structure with the structural blueprint of the pipeline, so as to clarify the specific pipe section positions corresponding to each accumulation area. Use a laser positioning device in cooperation with an inertial navigation module to obtain the spatial position data of each section of the pipeline, which is expressed in a three-dimensional coordinate system. For example, the X, Y, and Z axes respectively represent the length direction, height direction, and width direction of the pipeline. Align the central coordinates and boundary contour data of all accumulation areas with the three-dimensional coordinate system of the pipeline to form a "particle accumulation position index table". In this index table, each record includes position coordinates (for example, X = 4.3 m, Y = 1.5 m, Z = 0.9 m), accumulation volume (for example, 1500 mm³), and thickness value (such as 7 mm). In this way, the positions of solid particle accumulation are accurately marked on the spatial structure of the pipeline.

[0114] Step S362: Use an industrial endoscope to obtain the particle accumulation images at the positions of solid particle accumulation in the pipeline;

[0115] In this embodiment, the accumulated position of the marked solid particles is used as input control data, and an industrial endoscope is operated to enter the target pipe section to perform image acquisition operations. An industrial endoscope probe with a 360-degree rotation ability is used, and a light source module is integrated at the front end of the probe, with the brightness set to not less than 1000 lumens to ensure sufficient exposure conditions in a closed pipeline environment. The forward direction of the endoscope is controlled by a mechanical guiding device according to a preset path, and the guiding path is constructed based on the position coordinates obtained in S361. The image acquisition frequency is set to 30 frames per second to ensure that the image resolution reaches 1920×1080 pixels. During the acquisition process, the image frames and the three-dimensional coordinate information of the probe are synchronously recorded to facilitate subsequent binding of the images and spatial positions. The finally obtained image data includes a color image sequence with clear details of the accumulation, and the image format is uncompressed PNG format to retain the image pixel information.

[0116] Step S363: Binarize the particle accumulation image and mark the particle accumulation area;

[0117] In this embodiment, grayscale processing is performed to convert the color image into a grayscale image, and the weighted average method is used to calculate the grayscale value, that is, grayscale value = 0.299×R + 0.587×G + 0.114×B. Then, image binarization operation is performed, and a fixed threshold is set. For example, the threshold is set to 128 (the pixel value range is 0–255). The area with a pixel value greater than 128 in the grayscale image is marked as the foreground (value is 1), and the area less than or equal to 128 is marked as the background (value is 0). After the binary image is generated, a contour extraction algorithm (such as Canny edge detection or findContours in OpenCV) is used to extract the particle accumulation area. The outer boundary of each particle accumulation area is determined through contour closure detection, and the contour area is calculated at the same time. Areas with a contour area less than the threshold (for example, less than 500 pixel points) are removed to remove image noise and misidentified areas. Finally, the particle accumulation areas in each frame of the image are uniquely numbered and marked and output in the form of a binary mask image.

[0118] Step S364: Obtain the pipeline position space; map the particle accumulation area to the pipeline position space and identify the particle accumulation area at the pipeline elbow;

[0119] In this embodiment, the spatial coordinates recorded in step S361 are matched with the image data marked in step S363. First, the probe coordinates during image acquisition are extracted, and the particle accumulation area in the image is mapped to the three-dimensional pipeline space through the inverse projection method. The specific method is to establish a projection conversion matrix between the pixel points and the three-dimensional coordinate system of the pipeline. This matrix is determined by the focal length of the endoscope, the position of the imaging center, and the attitude angle during acquisition (obtained through a gyroscope). Using the triangulation method, the two-dimensional boundary of the particle accumulation area in the image is mapped into a three-dimensional space point set and matched with the position in the pipeline structure model. According to the three-dimensional model structure diagram, the elbow area is identified. The elbow area is defined as a pipe segment in the pipeline space with a non-linear turn, where the turning angle of the center line is between 45° and 135°, and the continuous curvature change is greater than a set value (such as 10 degrees / meter). The mapped particle accumulation area is judged to coincide with these elbow spaces, and the particle accumulation point set overlapping with the elbow area is extracted and marked as the "elbow particle accumulation area".

[0120] Step S365: Calculate the compactness of the particle accumulation area in the pipeline elbow;

[0121] In this embodiment, the space volume ratio method is used to calculate the compactness. The compactness is defined as the ratio of the actual volume of the particles to the volume of the available flow space in the elbow. First, the effective flow space volume V_total of the elbow section is extracted, which is calculated through three-dimensional modeling based on the inner diameter of the pipeline (for example, the inner diameter is 150 mm) and the length of the elbow section (for example, 300 mm). For example, V_total = π×(75 mm)²×300 mm = 5,301,437 mm³. Then, the total volume sum V_aggregated of all particle accumulations in this area is extracted, for example, 2,300,000 mm³. The compactness calculation formula is D = V_aggregated / V_total, and the result is 0.43. The compactness value is reserved to two decimal places. The compactness of all elbow particle accumulation areas is calculated one by one using this method, and an "elbow particle accumulation compactness table" is formed.

[0122] Step S366: Determine the pipeline blockage based on the compactness of the particle accumulation area in the pipeline elbow to obtain pipeline blockage data.

[0123] In this embodiment, a clogging determination criterion is set, and the quantitative grading threshold method is adopted. The tightness thresholds are set as follows: when D > 0.6, it is determined as "severe clogging"; when 0.3 < D ≤ 0.6, it is determined as "moderate clogging"; when D ≤ 0.3, it is determined as "mild or no clogging". For example, when the tightness of a certain elbow area is 0.72, it is marked as "severe clogging"; when the tightness of another area is 0.45, it is marked as "moderate clogging". During the determination process, the clogging result is recorded together with the elbow position coordinates to generate a "Pipeline Clogging Record Table", which includes fields: three-dimensional coordinates of the clogging position, corresponding tightness value, clogging level label (such as: severe / moderate / mild). All determination results are stored as the output data of the "Pipeline Risk Determination Model" and used as the basis for subsequent cleaning scheduling and maintenance plan formulation.

[0124] Preferably, step S4 is specifically as follows:

[0125] Step S41: Construct a pipeline risk determination model based on the pipeline clogging data and the pipeline perforation probability;

[0126] In this embodiment, based on the collected pipeline clogging data and pipeline perforation probability data, a multi-dimensional pipeline risk determination data set is first constructed. The pipeline clogging data is collected by the flow velocity sensors and particulate matter concentration sensors arranged inside the pipeline. The collection frequency is set to 1 Hz, and the flow velocity changes and particulate matter concentration are continuously recorded for 7 days. The recorded data format is CSV, and the fields include timestamp, flow velocity (unit: m / s), particulate matter concentration (mg / m³), and pipe section number. The pipeline perforation probability is obtained by analyzing the pipeline material fatigue data and historical maintenance records. The fatigue data is from the strain sensor detection of the pipeline once per hour, and the strain threshold is set to 200 με. If it exceeds, it is marked as a fatigue section. The historical records are read in JSON format, including the time, location, cause, and treatment method of each maintenance event. Based on the above data, a risk determination model is constructed using hierarchical logic rules. For example, the pipeline clogging warning threshold is set as a flow velocity drop exceeding 20% and a particulate matter concentration higher than 150 mg / m³. Combining with the perforation probability grading (higher than 0.6 is defined as high risk), different types of data are input into a Python script in a dictionary structure, and the NumPy and Pandas libraries are used to complete data fusion, and a multi-condition determination table is constructed according to the conditional screening logic, thereby establishing a determination process based on rule judgment rather than a fitting model.

[0127] Step S42: Determine the pipeline material damage risk for the air purification simulation process according to the pipeline risk determination model, and generate pipeline material damage risk data;

[0128] In this embodiment, the above - constructed pipeline risk determination model is used to analyze the operating state of pipeline materials in real - time during the air purification simulation process, generating pipeline material damage risk data. In the air purification simulation system, the initial parameters of the pipeline material are set as Q235 steel, with its yield strength set at 235 MPa, elastic modulus of 2.1×10¹¹ Pa, and Poisson's ratio of 0.3. The purification simulation is carried out by the CFD software FLUENT. The wind speed is set at 12 m / s, the temperature is 45 °C, and the particulate matter concentration is set at 180 mg / m³. The static analysis of the pipeline structure stress state is carried out, and finite - element simulation is performed through ANSYS Workbench to obtain the equivalent stress values at different time nodes. The equivalent stress data obtained from the simulation is compared with the material fatigue limit (set at 160 MPa). If the equivalent stress exceeds the fatigue limit for 3 consecutive time steps (the time - step length is 10 s), it is determined that there is a risk of material damage. The risk location, risk level, and simulation time point are integrated into structured data and recorded in CSV format. The fields include "pipe segment number", "maximum stress value (MPa)", "risk level", "occurrence time point", etc., as the pipeline material damage risk data.

[0129] Particularly importantly, step S42 includes the following steps:

[0130] Step S421: When any of the following situations occurs, it is determined as a high - risk of pipeline material damage, and high - risk data of pipeline material damage is generated: the air flow velocity in the pipeline deviates from the design range by more than ±15%, the pipeline surface temperature exceeds the design - bearable range, and the gas pressure in the pipeline deviates from the safe working pressure by more than 10%.

[0131] In this embodiment, when determining the high - risk of pipeline material damage, key data such as the air flow velocity in the pipeline, the pipeline surface temperature, and the gas pressure in the pipeline need to be collected first. These data can be monitored in real - time through sensors installed in the pipeline. The air flow velocity sensor should be accurate to millimeters per second, the pipeline surface temperature sensor should have an accuracy of ±2 °C, and the gas pressure sensor should support a measurement range of ±0.5 MPa. These parameters are recorded through a data acquisition system and compared with the preset design range. For example, the design range of the air flow velocity is 5 - 15 m / s. When the actually measured air flow velocity deviates from this range by more than ±15%, it is determined that the air flow velocity is abnormal. The set thresholds for temperature and pressure should be set according to the tolerance standards of the pipeline material. For example, the pipeline surface temperature should not exceed 80 °C, and the gas pressure should not exceed 5 MPa. At this time, if any data exceeds the set range through system judgment, it is immediately marked as a high - risk of pipeline material damage, and the corresponding high - risk data is generated. These data will be saved and transmitted to the subsequent risk assessment system for further processing.

[0132] Step S422: When the following conditions occur simultaneously, it is determined that the risk of severe pipeline material damage is serious, and data on the risk of severe pipeline material damage is generated: obvious cracks or deformations appear on the outer surface of the pipeline, the pipeline leakage detection exceeds the safety standard twice in a row, the corrosion depth of the inner wall of the pipeline exceeds 10 mm, and the fluctuations of the air flow temperature and pressure in the pipeline exceed the set threshold and continue for more than 30 minutes;

[0133] In this embodiment, when determining the severe risk of pipeline material damage, it is necessary to check the external structure and internal conditions of the pipeline in sequence. First, the outer surface of the pipeline is inspected using a visualization detection tool, such as an ultrasonic detector or a high-precision camera, which can effectively detect cracks, deformations, or corrosion on the pipeline surface. The sensitivity of ultrasonic detection should be set to 0.5 mm to detect tiny cracks. Then, a leakage detection system is used to monitor pipeline leakage. The accuracy requirement for leakage detection is at least 0.1 L / min. If the leakage amount exceeds the set safety standard (such as exceeding 1 L / min) in two consecutive detection results, it is considered that the pipeline has a leakage risk. For corrosion detection, an endoscope device combined with an electrochemical sensor is used to inspect the inner wall of the pipeline. If the corrosion depth exceeds 10 mm, it is marked as a severe risk. At the same time, pressure sensors and temperature sensors are used to continuously monitor the air flow in the pipeline. If the temperature and pressure fluctuations exceed the set threshold (such as the temperature fluctuation exceeds ±5°C and the pressure fluctuation exceeds ±0.2 MPa) and continue for more than 30 minutes, the system will automatically determine that the risk of severe pipeline material damage exists. These abnormal data will be marked and stored as severe risk data in the risk database for subsequent processing.

[0134] Step S423: Integrate the high-risk data of pipeline material damage and the severe-risk data of pipeline material damage to obtain pipeline material damage risk data.

[0135] In this embodiment, when integrating the high-risk data and severe-risk data of pipeline material damage, it is first necessary to ensure the accuracy and timeliness of all data. This process requires extracting the high-risk data of pipeline material damage and the severe-risk data of pipeline material damage from the risk assessment system. The content in the two datasets should include parameters such as air flow velocity, temperature, pressure, leakage volume, corrosion depth, etc. The system will standardize these data. For example, for the air flow velocity, it will be compared with the design range. If it exceeds the range of ±15%, it is a high risk, and if it exceeds the relevant safety standards, it is a severe risk. For the treatment of the leakage volume, if it exceeds the set threshold of 1 L / min, it will be classified as a severe risk. All detection results will be sorted according to the set thresholds, and the high-risk data and severe-risk data will be integrated through data fusion technology. This integration process uses a weighted algorithm, which assigns different weights according to the importance of each data point, so as to finally generate comprehensive pipeline material damage risk data. The integrated data will be used for subsequent risk prediction and pipeline maintenance decision-making.

[0136] Step S43: Determine the pipeline leakage risk for the air purification simulation process according to the pipeline risk determination model, and generate pipeline leakage risk data;

[0137] In this embodiment, for the pressure flow field results in the air purification simulation process, the pipeline leakage risk is evaluated according to the risk determination model to generate leakage risk data. First, the maximum internal pressure of each pipe segment is extracted based on the flow field simulation results, with the unit of Pa. The simulation time period is set to 24 hours, and the time step is 1 s. After extracting the pressure data once per second, the change rate of the maximum pressure and the average pressure of each pipe segment is obtained through Python processing. Combining the material compressive strength (the compressive limit of Q235 steel is set to 310 MPa) and the wall thickness (assumed to be 3 mm), the safety factor is calculated as the compressive strength / the maximum pressure. If the safety factor is less than 1.2, it is marked as having a potential leakage risk. Further, the pressure gradient of the same pipe segment is calculated, and the calculation formula is ΔP / Δx, where ΔP is the pressure change value and Δx is the length of the corresponding pipe segment (unit: m). If ΔP is greater than 5×10 5 Pa / m, combined with the rules of the determination model, it is confirmed that there is a high-risk area of leakage caused by stress concentration. Finally, the leakage risk data is summarized, and the recorded fields include "pipe segment number", "maximum pressure value (Pa)", "safety factor", "pressure gradient", "leakage risk level", and the unified format is a tabular JSON structure and is stored in the central server for subsequent integration.

[0138] Especially importantly, step S43 includes the following steps:

[0139] Step S431: When any of the following situations occurs, it is determined that the pipeline leakage risk is relatively high, and pipeline leakage high-risk data is generated: the gas pressure in the pipeline deviates from the normal operating pressure range by more than ±10%, the pipeline surface temperature exceeds the designed temperature range, or the gas flow velocity in the pipeline exceeds the safe operation upper limit by 10%.

[0140] In this embodiment, when it is determined that the pipeline leakage risk is relatively high, data collection is first carried out through the pressure sensor, temperature sensor, and gas flow velocity sensor installed in the pipeline system. These sensors should meet high-precision requirements. For example, the accuracy of the pressure sensor is ±0.1 MPa, the accuracy of the temperature sensor is ±1 °C, and the accuracy of the gas flow velocity sensor is ±0.5 m / s. The collected data is transmitted to the monitoring system in real time. During the judgment process, if the gas pressure deviates from the set normal operating pressure range by more than ±10% (for example, if the normal operating pressure is 5 MPa, exceeding 5.5 MPa or being lower than 4.5 MPa is determined as abnormal), it is determined as a high risk. At the same time, if the pipeline surface temperature exceeds the designed temperature range (such as the upper limit of the designed temperature is 80 °C, exceeding this range is regarded as abnormal) or the gas flow velocity exceeds the set safe operation upper limit by 10% (for example, the safe upper limit of the gas flow velocity is 15 m / s, exceeding 16.5 m / s is regarded as abnormal), it is determined that the pipeline leakage risk is relatively high. All abnormal data will generate pipeline leakage high-risk data in the data recording system and be stored as the basis for subsequent processing.

[0141] Step S432: When the following situations occur simultaneously, it is determined that the pipeline leakage risk is serious, and pipeline leakage serious-risk data is generated: cracks or deformations appear on the pipeline surface, and the crack width exceeds 0.5 mm; the pipeline leakage detection system shows that the leakage volume continues to increase, and the leakage volume measured three consecutive times exceeds 5 L / min; the internal gas flow temperature and pressure in the pipeline fluctuate beyond the set threshold and the duration exceeds 20 minutes.

[0142] In this embodiment, when determining that the pipeline leakage risk is serious, first, the outer surface of the pipeline is inspected by visual detection means (such as high-resolution cameras and ultrasonic detection equipment) to identify cracks and deformations. The detection of cracks shall ensure that the crack width exceeds 0.5 mm (the crack width measured by the ultrasonic detection equipment is used as the judgment criterion). Then, an online leakage detection system is used to monitor the leakage volume of the pipeline. If the leakage detection results show that the leakage volume exceeds 5 L / min for three consecutive times, the leakage volume is considered abnormal. In the monitoring of the air flow temperature and pressure, temperature sensors and pressure sensors are used to record the temperature and pressure of the air flow in the pipeline, and the real-time data is compared with the set safety threshold. For example, if the fluctuation range of the air flow temperature exceeds ±5°C, or the pressure fluctuation exceeds ±0.2 MPa, and this abnormal fluctuation lasts for more than 20 minutes, it is determined that the pipeline leakage risk is serious. All detection data and results will be marked as pipeline leakage serious risk data and stored for subsequent analysis.

[0143] Step S433: Integrate the pipeline leakage high-risk data and the pipeline leakage serious risk data to obtain the pipeline leakage risk data.

[0144] In this embodiment, when integrating the pipeline leakage high-risk data and the pipeline leakage serious risk data, first, all relevant data are extracted from the monitoring system, including pressure, temperature, air flow velocity, crack width, leakage volume, temperature and pressure fluctuation conditions, etc. Each data point will be classified according to the preset threshold. For example, data such as the air flow velocity exceeding 10% of the set upper limit, the pressure fluctuation exceeding ±10%, and the crack width exceeding 0.5 mm are marked as pipeline leakage high-risk data. At the same time, data with a leakage volume exceeding 5 L / min, a temperature fluctuation exceeding ±5°C, or a pressure fluctuation exceeding ±0.2 MPa and lasting for more than 20 minutes are marked as pipeline leakage serious risk data. Then, the data of these two risk categories are merged through a data integration algorithm to ensure that each abnormal data point is accurately classified. Finally, all the integrated data will form a complete pipeline leakage risk data for subsequent risk analysis and pipeline maintenance decision-making.

[0145] Step S44: Integrate the pipeline material damage risk data and the pipeline leakage risk data to obtain the pipeline risk data;

[0146] In this embodiment, the previously obtained pipeline material damage risk data and pipeline leakage risk data are integrated to generate unified pipeline risk data. During the data integration process, the unique pipe segment number is first used as the primary key, and horizontal data alignment is performed through the merge function in the Pandas library of Python. Field concatenation is performed on fields such as "maximum stress value", "risk level", "maximum pressure value", "safety factor", "leakage risk level", etc. under each pipe segment number. A risk comprehensive scoring mechanism is set, with the material damage risk level set to 13 points and the leakage risk level set to 13 points. The total risk level is the sum of the two scores. If the total score is greater than 4, it is a high risk; if it is equal to 3, it is a medium risk; if it is less than 3, it is a low risk. The integrated data table retains all original fields and adds fields "comprehensive risk level" and "evaluation timestamp", with the timestamp accurate to the second. After integration, abnormal data (such as missing values for pressure or stress) is removed and stored uniformly as a Parquet format file for easy reading and analysis in big data platforms such as Hadoop or Spark later.

[0147] Step S45: Transmit the pipeline risk data to the risk monitoring system to execute the pipeline risk warning task.

[0148] In this embodiment, the above-integrated pipeline risk data is transmitted to the risk monitoring system deployed in the industrial control network to execute the risk warning task. The monitoring system is based on an embedded edge computing terminal (model: NVIDIA Jetson Xavier NX). The system adopts a timed data polling mechanism internally and automatically retrieves new risk data files every 30 seconds. Through the configured FTP data interface, the system specifies the access path of the pipeline risk data storage, and uses the SFTP protocol to complete the secure file transmission. After the transmission is successful, the data parsing module is automatically called to read the pipe number and comprehensive risk level by field and map them to the risk visualization interface in real time. The risk warning interface is built using HTML5 + WebSocket technology. Each pipe segment is displayed as a graphical node. When the comprehensive risk level is high, the node flashes red; when it is medium, it is yellow; when it is low risk, it is green and always on. The system updates the risk level trend chart every 15 minutes and sends high-risk warnings to the preset operation and maintenance terminals through the SMS gateway. The SMS content includes field information such as "pipe segment number", "risk type", "detection time", "recommended maintenance time window", etc. All transmission records, parsing logs, and warning trigger times are recorded in the system log file, and the log retention period is set to 30 days, in the format of a UTF-8 encoded text file.

[0149] Preferably, this specification also provides a construction system for a pipeline risk determination model, which is used to execute the construction method of the pipeline risk determination model as described above. The construction system for the pipeline risk determination model includes:

[0150] An air purification simulation module generates compressed air using an air compressor and conveys the compressed air to a preset pipeline to obtain pipeline compressed air data; constructs an air purification model based on the pipeline compressed air data; performs air purification simulation according to the air purification model to obtain purified air data;

[0151] A pipeline corrosion identification module detects residual condensate based on the purified air data to obtain residual condensate data; performs pipeline corrosion identification based on the residual condensate data to obtain pipeline corrosion data; predicts the pipeline perforation probability according to the pipeline corrosion data;

[0152] A pipeline blockage analysis module detects accumulated solid particles based on the purified air data to obtain accumulated solid particle data; performs pipeline blockage analysis according to the accumulated solid particle data to obtain pipeline blockage data;

[0153] A pipeline risk determination module constructs a pipeline risk determination model according to the pipeline blockage data and the pipeline perforation probability; performs pipeline risk determination on the air purification simulation process according to the pipeline risk determination model to generate pipeline risk data; transmits the pipeline risk data to a risk monitoring system to execute the pipeline risk warning task.

[0154] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Accordingly, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be encompassed within the present invention.

[0155] The above description is only a specific implementation manner of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a pipeline risk determination model, characterized in that The following steps are involved: Step S1: using an air compressor to generate compressed air, and delivering the compressed air to a preset pipeline to obtain pipeline compressed air data; constructing an air purification model according to the pipeline compressed air data; performing air purification simulation according to the air purification model to obtain purified air data; Step S2: detecting residual condensed water according to the purified air data to obtain residual condensed water data; performing pipeline corrosion identification based on the residual condensed water data to obtain pipeline corrosion data; and predicting the probability of pipeline perforation according to the pipeline corrosion data; Step S3: Detecting accumulated solid particles according to the purified air data to obtain accumulated solid particle data; performing pipeline blockage analysis according to the accumulated solid particle data to obtain pipeline blockage data. Step S3 is specifically as follows: Step S31: extracting particle concentration according to purified air data; Step S32: extracting concentration measurement timestamps based on the particle concentration; sorting the concentration measurement timestamps by time and constructing a concentration time series; identifying the abnormal change trend of the concentration time series; extracting the abnormal concentration change time based on the abnormal change trend; Step S33: photographing a pipeline image based on the abnormal concentration change time; performing color space conversion on the pipeline image to obtain a color space converted pipeline image; performing solid particle area recognition on the color space converted pipeline image based on a preset solid particle color range, and measuring the solid particle thickness of the solid particle area using a laser rangefinder to obtain the solid particle thickness; Step S34: reconstructing the solid particle region in three dimensions to obtain a three-dimensional structure of the solid particles, and measuring the volume of the solid particles in the three-dimensional structure of the solid particles; Step S35: judging the accumulation of solid particles according to the thickness and volume of solid particles, and obtaining accumulated solid particle data; Step S36: Perform pipeline blockage analysis based on the accumulated solid particle data and record the pipeline blockage data. Step S36 is specifically as follows: Step S361: marking the location of solid particle accumulation in the pipeline according to the accumulated solid particle data; Step S362: using an industrial endoscope to obtain a particle accumulation image of a location where solid particles accumulate in the pipeline; Step S363: performing binary conversion on the particle accumulation image and marking the particle accumulation area; Step S364: acquiring the pipeline position space; mapping the particle accumulation area to the pipeline position space, and identifying the particle accumulation area of the pipeline elbow; Step S365: calculating the compactness of the particle accumulation area of the pipeline elbow; Step S366: determining the pipeline blockage in the particle accumulation area of the pipeline elbow according to the tightness, and obtaining pipeline blockage data; Step S4: construct a pipeline risk determination model based on pipeline blockage data and pipeline perforation probability; perform pipeline risk determination on the air purification simulation process based on the pipeline risk determination model to generate pipeline risk data; The pipeline risk data is transmitted to the risk monitoring system to perform pipeline risk early warning tasks.

2. The method for constructing the pipeline risk judgment model according to claim 1, characterized in that Step S1 is specifically as follows: Step S11: Start the air compressor and set the compression pressure value to 0.6 - 0.8 MPa; Use the air compressor to inhale a preset amount of atmospheric air, perform a compression operation to generate compressed air, and transport the compressed air to a preset pipeline to obtain pipeline compressed air data; Step S12: Identify the pipeline transportation path according to the pipeline compressed air data; Step S13: Detect the air impurity components according to the pipeline compressed air data; Step S14: Set the air purification target based on the air impurity components; Step S15: Construct an air purification model according to the pipeline transportation path and the air purification target; Step S16: Perform air purification simulation according to the air purification model to obtain purified air data.

3. The method for constructing a pipeline risk judgment model according to claim 2, wherein Step S16 specifically includes: Step S161: Import the air purification model into the simulation software; Step S162: Set the input compressed air volume flow rate to 1 - 20 m³ / min and the moisture content to 0.01 - 0.2 kg / kg in the simulation software; Step S163: Set the temperature of the condensation unit to 3 - 10 °C, the material of the adsorption unit to activated alumina, and the particle size retention range of the filtration unit to 0.1 - 10 μm in the simulation software; Step S164: Set the ambient temperature to 25–45 °C, the system operating pressure to 0.7–1.0 MPa, and the relative humidity to 30%–90% in the simulation software; Step S165: Run the air purification simulation program in the simulation software and output the purified air data.

4. The method for constructing a pipeline risk judgment model according to claim 1, wherein, The detection of residual condensate water in Step S2 includes: Extract the humidity data of the purified air data; Determine the dew point temperature based on the humidity data; Calculate the dew point deviation value based on the preset target dew point value and the dew point temperature; Record the dew point deviation period according to the dew point deviation value, collect the purified air during the dew point deviation period, detect the residual condensate water using a condensate water sensor, and calculate the accumulated volume of the residual condensate water to obtain the residual condensate water data.

5. The method for constructing a pipeline risk judgment model according to claim 1, wherein The pipeline corrosion identification in Step S2 includes: Mark the position of the condensate water pipeline based on the residual condensate water data; Irradiate the position of the condensate water pipeline with high-energy X-rays and generate an X-ray fluorescence spectrum; Speculate the type of metal elements on the pipeline surface based on the X-ray fluorescence spectrum to obtain the metal element data; Perform an electrochemical reaction simulation according to the metal element data and the residual condensate water data to generate current data; Perform an oxygen reduction reaction simulation based on the current data and the metal element data to obtain oxygen reduction data; Perform rust accumulation analysis based on the oxygen reduction data to obtain rust accumulation data; Detect the pipeline corrosion area at the position of the condensate water pipeline according to the rust accumulation data to obtain the pipeline corrosion data.

6. The method for constructing a pipeline risk determination model according to claim 1, wherein The prediction of the pipeline perforation probability in Step S2 includes: Calculate the corrosion rate according to the pipeline corrosion data; Identify the inner wall corrosion area of the pipeline according to the pipeline corrosion data; Use a laser confocal microscope to perform microscopic three-dimensional morphology imaging on the inner wall corrosion area of the pipeline to obtain a three-dimensional morphology image of the pipeline inner wall; Calculate the Gaussian curvature based on the three-dimensional morphology image of the pipeline inner wall; Mark the bulging area of the pipe wall in the three-dimensional morphology image of the pipeline inner wall based on the Gaussian curvature; Identify the minimum wall thickness of the pipe wall in the bulging area of the pipe wall; Continuously monitor the wall thickness reduction rate in the bulging area of the pipe wall; Predict the pipe perforation probability based on the minimum wall thickness of the pipe wall and the wall thickness thinning rate.

7. The method for constructing a pipeline risk judgment model according to claim 1, wherein Step S4 is specifically as follows: Step S41: Construct a pipe risk determination model based on the pipe blockage data and the pipe perforation probability; Step S42: Determine the pipe material damage risk for the air purification simulation process according to the pipe risk determination model, and generate pipe material damage risk data; Step S43: Determine the pipe leakage risk for the air purification simulation process according to the pipe risk determination model, and generate pipe leakage risk data; Step S44: Integrate the pipe material damage risk data and the pipe leakage risk data to obtain the pipe risk data; Step S45: Transmit the pipe risk data to the risk monitoring system to perform the pipe risk early warning task.

8. A construction system for a pipeline risk judgment model, characterized in that, For constructing the pipe risk determination model according to the method described in claim 1, the pipe risk determination model construction system includes: An air purification simulation module that uses an air compressor to generate compressed air, conveys the compressed air to a preset pipe to obtain pipe compressed air data; constructs an air purification model according to the pipe compressed air data; performs air purification simulation according to the air purification model to obtain purified air data; A pipe corrosion identification module that detects residual condensate water according to the purified air data to obtain residual condensate water data; performs pipe corrosion identification based on the residual condensate water data to obtain pipe corrosion data; predicts the pipe perforation probability according to the pipe corrosion data; A pipe blockage analysis module that detects accumulated solid particles according to the purified air data to obtain accumulated solid particle data; performs pipe blockage analysis according to the accumulated solid particle data to obtain pipe blockage data; A pipe risk determination module that constructs a pipe risk determination model according to the pipe blockage data and the pipe perforation probability; determines the pipe risk for the air purification simulation process according to the pipe risk determination model to generate pipe risk data; transmits the pipe risk data to the risk monitoring system to perform the pipe risk early warning task.

Images