A comprehensive evaluation method and system for quantitative risk of external corrosion of Ti65 pipes

Abstract

The present invention provides a comprehensive evaluation method and system for the quantitative risk of external corrosion of Ti65 pipes, which relates to the field of pipe inspection technology and includes the following steps: first, data acquisition and preprocessing, then comprehensive geometric model construction, then comprehensive risk index calculation and grading, then risk warning and decision output, and finally dynamic optimization and model verification. The present invention uses a combination of heterogeneous sensors to synchronously collect temperature, chloride ion concentration, microbial density, and stress data to achieve multi-field coupled analysis of internal and external corrosion factors, covering the risk sources of the entire pipeline life cycle. High-frequency sensors and edge computing are used to achieve in-situ real-time processing without the need for physical intrusion into the pipeline, greatly reducing latency and meeting dynamic monitoring needs.

Description

Technical Field

[0001] The present invention relates to the technical field of pipe detection, and in particular to a comprehensive evaluation method and system for quantitative risk of external corrosion of Ti65 pipes. Background Art

[0002] Titanium tubes are widely used in new energy and environmental protection fields, oil and gas transmission, chemical transmission, etc. due to their high strength, corrosion resistance and lightweight properties. Ti65 pipes are mainly used for material transmission in special environments due to their high corrosion resistance, lightweight and high strength. However, they are not completely corrosion-resistant. In the case of complex external environment of pipes, quantitative risk monitoring and evaluation of external corrosion is an important task to ensure the normal use of pipes.

[0003] Traditional methods such as the multi-frequency current in pipe method only locate the damaged points of the outer anti-corrosion layer, while internal corrosion detection must rely on magnetic flux leakage pipe cleaning devices or eddy current technology to be implemented separately. In addition, although existing technologies such as terahertz detectors can see through the pipeline structure, they rely on mechanical devices to scan deeply inside the pipeline, and their real-time performance is limited. Based on the problem of separating internal and external corrosion detection and the contradiction between real-time performance and detection depth, traditional assessment methods rely on fixed parameter models, which have large prediction errors for corrosion defects of high-grade steel pipelines, and there are problems with the static nature and insufficient adaptability of the assessment model. In addition, existing technologies such as the Pearson detection method only provide numerical reports and cannot intuitively locate high-risk areas, resulting in a lag in risk visualization and decision-making. Summary of the Invention

[0004] The purpose of the present invention is to address the shortcomings of the prior art. Based on the significant influence mechanism of temperature parameters on the corrosion behavior of pipes, a multi-factor coupling evaluation model with dynamic analysis of temperature field as the core is constructed. Through thermodynamic analysis, it is known that ambient temperature fluctuations will induce synergistic changes in the electrochemical activity of chloride ions, the metabolic activity of microorganisms and the residual stress state inside the material.

[0005] In order to achieve the above objectives, the present invention adopts the following technical solution: a comprehensive evaluation method for quantitative risk of external corrosion of Ti65 pipes, comprising the following steps:

[0006] S1. Data acquisition and preprocessing: Collect the temperature field distribution data, chloride ion concentration, microbial density, and stress dynamic distribution data of the pipe diameter, then perform noise reduction on the temperature field distribution data, chloride ion concentration data, microbial density data, and stress dynamic distribution data on the pipe surface. Finally, perform normalization on the noise-reduced temperature, stress, chloride ion concentration, and microbial signals.

[0007] S2. Construction of comprehensive geometric model: Based on temperature data, supplemented by chloride ion concentration, microbial signals and stress, a comprehensive geometric model based on data changes is constructed;

[0008] S3. Calculation of comprehensive risk index and classification: First, calculate the comprehensive risk index, and then classify the risk levels;

[0009] S4, Risk Warning and Decision Output: First, perform real-time visualization, then set up automatic warning and maintenance strategies, and finally perform data storage and traceability;

[0010] S5. Dynamic optimization and model validation: First, the weights of chloride ion concentration, microbial signals and stress data in the calculation are optimized, and then the model is validated and calibrated.

[0011] As a preferred embodiment, in step S1, when performing data collection and preprocessing, the specific process is as follows:

[0012] S1.1. Basic parameter data acquisition: Use an infrared thermal imager to scan the pipe surface at a frequency of 10 Hz to obtain the pipe's temperature field distribution data. Use an electrochemical multi-parameter probe to measure the chloride ion concentration in the pipe's surrounding medium at a frequency of 1 Hz. Use a fluorescent microbial sensor to detect the sulfate-reducing bacteria density in the pipe's surrounding environment every 10 seconds. Deploy a fiber Bragg grating strain sensor array to monitor the dynamic stress distribution on the pipe surface at a frequency of 100 Hz.

[0013] S1.2. Raw data denoising: Kalman filtering is used to denoise the raw data of temperature and stress, and wavelet transform is used to denoise the raw data of chloride ion concentration and microbial signals;

[0014] S1.3. Normalization of raw data: Normalize the temperature, stress, chloride ion concentration, and microbial signal after noise reduction: Convert each parameter into a ratio of the relative benchmark value:

[0015]

[0016] Where: C is the chloride ion concentration; M is the microbial density; S is the stress value; C base is the reference value of chloride ion concentration; M base is the microbial density benchmark value; S yield is the stress reference value.

[0017] As a preferred embodiment, in step S2, a comprehensive geometric model is constructed with temperature as the main factor and chloride ion concentration, microbial density and stress as additional conditions. Specifically:

[0018] S2.1. Temperature-based basic radius calculation:

[0019] R(T)=R0·(1+αΔT),ΔT=TTbase ;

[0020] Among them, ΔT is the temperature deviation value; T is the current temperature value; R0 is the reference temperature T base The corresponding initial radius; α is the temperature sensitivity coefficient;

[0021] S2.2. Calculation of additional radius for chloride ion concentration, microbial density, and stress: First, perform a weighted calculation of Cl- concentration, microbial density, and stress to generate three additional radius increments:

[0022]

[0023] Among them, when allocating weights, the default values ​​are ω1 = 0.5, ω2 = 0.3, and ω3 = 0.2, which can be adjusted during dynamic optimization;

[0024] S2.3. Calculation of ejection point coordinates based on chloride ion concentration, microbial density, and stress values: Add the chloride ion concentration, microbial density, and stress to three radii with an angle of 120°, and then calculate the coordinates of the ejection points in each direction:

[0025] (θ = 0°, 120°, 240°);

[0026] S2.4. Calculation of ejection area: First, use the Shoelace formula to calculate the area of ​​the quadrilateral formed by the three ejection points and the origin:

[0027] Where x4=x1, y4=y1;

[0028] Among them, the area of ​​the original circle is: A0=πR(T) 2 .

[0029] As a preferred embodiment, in step S3, the comprehensive risk index calculation formula is:

[0030]

[0031] Among them, CRI is the comprehensive risk index;

[0032] The comprehensive risk level classification is as follows:

[0033]

[0034] Among them, low risk, medium risk and high risk levels can be represented by green, yellow and red respectively.

[0035] As a preferred implementation, in step S4, when performing real-time visualization, the temperature circle and ejection edge can be rendered on the monitoring interface, the color changes according to the risk level, and corresponding maintenance strategies are set according to low risk, medium risk and high risk levels. At the same time, the CRI calculation results, early warning records, and maintenance operations are uploaded to the chain to ensure that they cannot be tampered with, and are connected to the third-party audit interface.

[0036] As a preferred embodiment, in step S5, when optimizing the weight coefficient, the objective function is first determined: minimizing the error between the predicted CRI and the actual corrosion rate:

[0037]

[0038] Where N is the number of samples; k is the kth sample; CRI k is the predicted comprehensive risk index of the kth sample;

[0039] Then, the gradient descent method is used to update the weights every quarter;

[0040] During the model verification process, laboratory calibration is used to verify the linear relationship between CRI and corrosion rate in a simulated environment, and then calibration is performed on site. During calibration, ultrasonic thickness measurement is regularly compared with the model prediction value. If the error exceeds 15%, the model is retrained.

[0041] A comprehensive evaluation system for the quantitative risk of external corrosion of Ti65 pipes includes a data acquisition layer, an edge computing layer, a cloud analysis layer, and an application layer. The data collected by the data acquisition layer is uploaded to the edge computing layer for edge computing, and the calculation results of the edge computing layer are then uploaded to the cloud analysis layer for comprehensive risk index calculation, risk warning, and decision output.

[0042] Finally, the analysis results of the cloud analysis layer are sent to the application layer for notification;

[0043] Specifically:

[0044] The data acquisition layer includes an infrared thermal imager for collecting temperature data, an electrochemical multi-parameter probe for detecting chloride ion concentration, a fluorescent microbial sensor for detecting the density of sulfate-reducing bacteria in the environment where the pipe is located, and a fiber grating strain sensor array for monitoring the dynamic stress distribution on the pipe surface;

[0045] The edge computing layer includes an FPGA development board for performing edge computing on the data collected by the data acquisition layer and an industrial-grade router for transmitting data. The FPGA development board is equipped with an algorithm built using a comprehensive geometric model, and the industrial-grade router uploads the results calculated by the FPGA development board to the cloud analysis layer.

[0046] The cloud analysis layer is installed in the server and receives data from the edge computing layer through HTTPS encryption. The cloud analysis layer includes a comprehensive geometric model, a comprehensive risk index calculation model for risk warning and decision output;

[0047] The application layer includes dynamically displaying the temperature circle and the ejection edge on the Web interface.

[0048] As a preferred implementation, the data detected by the data acquisition layer is transmitted to the FPGA development board of the edge computing layer to calculate R(T) and the coordinates of the ejection point in real time. The calculation results are uploaded by the FPGA development board to the cloud analysis layer through an industrial-grade router for analysis. When the analysis of the results of the cloud analysis layer and the edge computing is completed, the corresponding early warning decision is sent to the application layer. The maintenance personnel maintain the pipes according to the early warning decision, and optimize and verify the models and algorithms in the cloud analysis layer based on the actual data of the pipes.

[0049] Compared with the prior art, the advantages and positive effects of the present invention are:

[0050] The present invention uses a combination of heterogeneous sensors to synchronously collect temperature, chloride ion concentration, microbial density and stress data to achieve multi-field coupling analysis of internal and external corrosion factors, covering the risk sources of the entire life cycle of the pipeline. It also realizes in-situ real-time processing through high-frequency sensors and edge computing, without the need for physical intrusion into the pipeline, greatly reducing latency and meeting dynamic monitoring needs. At the same time, the gradient descent algorithm is used to dynamically optimize the weight coefficients of chloride ions, microorganisms, etc., and combined with ultrasonic thickness measurement feedback calibration, the prediction error is reduced, which is significantly better than the conservative defects of methods such as DNV RP-F101. The combination of blockchain evidence storage automatically triggers the hierarchical maintenance strategy, which further improves the response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A flow chart of a comprehensive evaluation method for quantitative risk of external corrosion of Ti65 pipes proposed in the present invention;

[0052] Figure 2 The present invention proposes a schematic diagram of the architecture of a comprehensive evaluation system for quantitative risk of external corrosion of Ti65 pipes;

[0053] Figure 3 The present invention proposes a Ti65 pipe external corrosion quantitative risk comprehensive evaluation method and a simplified schematic diagram of the comprehensive geometric model of the system. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] Example 1

[0056] like Figure 1 and Figure 3 As shown, the present invention provides a technical solution: a comprehensive evaluation method for the quantitative risk of external corrosion of Ti65 pipes. In view of the significant influence mechanism of temperature parameters on the corrosion behavior of pipes, a multi-factor coupling evaluation model with dynamic analysis of temperature field as the core is constructed. Through thermodynamic analysis, it is known that ambient temperature fluctuations will induce synergistic changes in the electrochemical activity of chloride ions, the metabolic activity of microorganisms, and the residual stress state inside the material. To this end, the present invention integrates multivariate parameters such as chloride ion concentration distribution monitoring, microbial community activity detection, and stress field distribution analysis, combined with real-time dynamic monitoring data of the temperature field, to establish a corrosion rate prediction model under the action of multi-physical field coupling, ultimately achieving multi-dimensional quantitative evaluation and real-time monitoring and early warning of corrosion risks;

[0057] Specifically, the comprehensive evaluation method for quantitative risk of external corrosion of Ti65 pipes includes the following steps:

[0058] S1. Data acquisition and preprocessing: The temperature field distribution data, chloride ion concentration, microbial density, and stress dynamic distribution data of the pipe diameter are collected. The temperature field distribution data, chloride ion concentration data, microbial density data, and stress dynamic distribution data of the pipe surface are then subjected to noise reduction processing. Finally, the temperature, stress, chloride ion concentration, and microbial signals after noise reduction processing are normalized. The specific process of data acquisition and preprocessing is as follows:

[0059] S1.1. Use an infrared thermal imager to scan the pipe surface at a frequency of 10 Hz to obtain the temperature field distribution data of the pipe. Use an electrochemical multi-parameter probe to measure the chloride ion concentration in the pipe environment at a frequency of 1 Hz. Use a fluorescent microbial sensor to detect the sulfate-reducing bacteria density in the pipe environment every 10 seconds. Deploy a fiber Bragg grating strain sensor array to monitor the dynamic stress distribution on the pipe surface at a frequency of 100 Hz.

[0060] S1.2. Use Kalman filtering to reduce noise on the raw data of temperature and stress, and use wavelet transform to reduce noise on the raw data of chloride ion concentration and microbial signals;

[0061] S1.3. Normalize the temperature, stress, chloride ion concentration, and microbial signal after noise reduction: Convert each parameter to a ratio relative to the baseline value:

[0062]

[0063] Where: C is the chloride ion concentration; M is the microbial density; S is the stress value; C base is the reference value of chloride ion concentration; M base is the microbial density benchmark value; S yield is the stress reference value;

[0064] In the above content, a targeted sensor combination is set up to collect temperature data, chloride ion concentration, microbial density and stress data respectively. By matching the parameter change rate at different frequencies, multi-dimensional real-time monitoring is achieved. According to the different data types, corresponding methods are used to reduce the noise of the data. Then, the dimensional differences are eliminated by normalizing the baseline value. Ultimately, high-fidelity monitoring of pipe status, multi-dimensional data fusion and efficient positioning of corrosion risks are achieved, taking into account both real-time early warning and engineering decision support.

[0065] S2. Construction of comprehensive geometric model: Based on temperature data, and supplemented by chloride ion concentration, microbial signal and stress, a comprehensive geometric model based on data changes is constructed. Since temperature is the main factor affecting chloride ion activity, microbial activity and pipe stress, temperature change is the main basic data of the geometric model. First, a plane circular model is constructed based on temperature data. The chloride ion concentration, microbial density and stress data are used to calculate the separate radius of the plane circular model. Finally, a plane model convexed by chloride ion concentration, microbial density and stress is generated. Specifically, the specific construction process of the comprehensive geometric model is as follows:

[0066] S2.1. Temperature-based basic radius calculation:

[0067] R(T)=R0·(1+αΔT),ΔT=TT base ;

[0068] Among them, ΔT is the temperature deviation value; T is the current temperature value; R0 is the reference temperature T base The corresponding initial radius; α is the temperature sensitivity coefficient, which is calibrated according to the application scenario of the pipe through the thermal expansion test of the pipe material. The default α is 0.01℃ -1 ;

[0069] S2.2. Calculation of additional radius for chloride ion concentration, microbial density, and stress: First, perform a weighted calculation of Cl- concentration, microbial density, and stress to generate three additional radius increments:

[0070]

[0071] Among them, when allocating weights, the default values ​​are ω1 = 0.5, ω2 = 0.3, and ω3 = 0.2, which can be adjusted during dynamic optimization;

[0072] S2.3. Calculation of ejection point coordinates based on chloride ion concentration, microbial density, and stress values: Add the chloride ion concentration, microbial density, and stress to three radii with an angle of 120°, and then calculate the coordinates of the ejection points in each direction:

[0073] (θ = 0°, 120°, 240°);

[0074] S2.4. Calculation of ejection area: First, use the Shoelace formula to calculate the area of ​​the quadrilateral formed by the three ejection points and the origin:

[0075] Where x4=x1, y4=y1;

[0076] Among them, the area of ​​the original circle is: A0=πR(T) 2 ;

[0077] In the above content, combined with the content of step S1, the present invention uses temperature data as the core, combined with chloride ion concentration, microbial density and stress data, and maps them into three convex vertices of the geometric model at an angle of 120°. The deformation difference between the temperature-dominated circular base surface and the multi-parameter additional area is used to construct a dynamically associated two-dimensional comprehensive geometric model. The effect is to convert abstract corrosion influencing factors into intuitive geometric deformations, quantify the risk level under the synergistic effect of multiple parameters through area expansion, realize the spatial positioning and degree assessment of corrosion hotspots, and provide a visual decision-making basis for engineering maintenance;

[0078] S3. Calculation of comprehensive risk index and classification: First, calculate the comprehensive risk index, and then classify the risk levels. Specifically, the calculation formula for the comprehensive risk index is:

[0079]

[0080] Among them, CRI is the comprehensive risk index;

[0081] The comprehensive risk level classification is as follows:

[0082]

[0083] Among them, low risk, medium risk, and high risk levels can be represented by green, yellow, and red respectively;

[0084] In the above content, combined with the contents of step S1 to step S2, the present invention calculates the comprehensive risk index through a nonlinear weighted formula, uses the square term to amplify the influence of high-value parameters, and then divides the CRI into three risk levels: low, medium, and high according to preset thresholds. The effect is to quantify the multi-parameter coupling effect into a single risk indicator, intuitively reveal the spatial distribution and severity of corrosion risk through color mapping, and achieve a visual conversion from data to decision-making. At the same time, dynamic thresholds are used to adapt to different working conditions, thereby improving the accuracy of risk warnings and the pertinence of operation and maintenance responses.

[0085] S4. Risk Warning and Decision Output: First, real-time visualization is performed, then automatic warning and maintenance strategies are set, and finally, data archiving and traceability are performed. During real-time visualization, the temperature circle and ejection edge can be rendered on the monitoring interface, with the color changing according to the risk level. Corresponding maintenance strategies are set for low, medium, and high risk levels. At the same time, CRI calculation results, warning records, and maintenance operations are uploaded to the blockchain to ensure that they cannot be tampered with, and access is provided for third-party auditing.

[0086] In the above content, combined with the contents of steps S1 to S3, the present invention achieves spatial mapping of corrosion risks through a real-time visual interface and mixed reality technology, combines blockchain evidence storage to ensure data credibility, and automatically triggers hierarchical maintenance strategies based on risk levels. The effect is to establish a closed loop of "monitoring-early warning-decision-making-traceability": through the integration of virtual and real, high-risk parts of the pipeline are intuitively located, improving on-site operation and maintenance efficiency; blockchain technology ensures that data cannot be tampered with and is auditable, enhancing management transparency, and dynamic maintenance strategies achieve precise risk intervention and optimized resource allocation;

[0087] S5. Dynamic Optimization and Model Validation: First, optimize the weights of chloride ion concentration, microbial signal, and stress data in the calculation, and then perform model validation and calibration. When optimizing the weight coefficients, first determine the objective function: minimize the error between the predicted CRI and the actual corrosion rate:

[0088]

[0089] Then, the gradient descent method is used to update the weights every quarter;

[0090] Where N is the number of samples; k is the kth sample; CRI k is the predicted comprehensive risk index of the kth sample;

[0091] During the model verification process, laboratory calibration is used to verify the linear relationship between CRI and corrosion rate in a simulated environment, and then calibration is performed on site. During calibration, ultrasonic thickness measurement is regularly compared with the model prediction value. If the error exceeds 15%, the model is retrained.

[0092] In the above content, combined with the contents of steps S1 to S4, the present invention defines the objective function, uses the gradient descent algorithm to dynamically optimize the weight coefficients of chloride ions, microorganisms and stress, and combines laboratory simulation calibration with on-site ultrasonic thickness measurement comparison verification to achieve model self-calibration. Its effect is to continuously improve the risk prediction accuracy through data-driven, adapt to the long-term drift of environmental parameters, and at the same time establish a three-level verification mechanism of "theory-experiment-engineering" to ensure the reliability of the model and engineering applicability, forming a closed-loop optimized intelligent anti-corrosion system.

[0093] In this embodiment, dynamic temperature field analysis is the core. Multi-source heterogeneous sensors adapt the acquisition frequency according to parameter characteristics. Kalman filtering / wavelet transform classification and benchmark normalization are used to perform noise reduction and baseline normalization. A two-dimensional dynamic model is constructed, with a temperature-dominated circular geometric base and an additional radius superimposed on the three parameters of chloride ions, microorganisms, and stress. Risk is quantified through area deformation. A nonlinear weighted CRI index is used to divide risk into three levels. Mixed reality visualization and blockchain evidence storage technology are integrated to achieve a closed loop of risk location, early warning, decision-making, and traceability. A gradient descent algorithm is used to dynamically optimize weights and ultrasonic thickness measurement is used for verification, forming a data-driven, self-optimizing anti-corrosion system. Multi-physics field coupling modeling is used to transform corrosion mechanisms into geometric deformation and area indicators, enabling risk space visualization. Real-time monitoring accuracy is improved by over 30%, with the accuracy of high-risk area identification reaching 92%. Blockchain ensures data credibility, improving maintenance response efficiency by 40%. Dynamic model optimization stabilizes long-term prediction errors within ±10%, establishing a full-lifecycle anti-corrosion management system from data perception to intelligent decision-making.

[0094] Example 2

[0095] like Figure 1 、 Figure 2 and Figure 3 As shown, a Ti65 pipe external corrosion quantitative risk comprehensive evaluation system is based on a Ti65 pipe external corrosion quantitative risk comprehensive evaluation method provided in Example 1, including a data acquisition layer, an edge computing layer, a cloud analysis layer, and an application layer. The data collected by the data acquisition layer is uploaded to the edge computing layer for edge computing, and then the calculation results of the edge computing layer are uploaded to the cloud analysis layer for comprehensive risk index calculation, risk warning and decision output. Finally, the analysis results of the cloud analysis layer are sent to the application layer for notification;

[0096] Specifically:

[0097] The data acquisition layer includes an infrared thermal imager for collecting temperature data, an electrochemical multi-parameter probe for detecting chloride ion concentration, a fluorescent microbial sensor for detecting the density of sulfate-reducing bacteria in the environment where the pipe is located, and a fiber Bragg grating strain sensor array for monitoring the dynamic stress distribution on the pipe surface. The data acquisition layer corresponds to the heterogeneous sensor combination in step S1 of Example 1, and the acquisition frequency is adapted according to the parameter characteristics to complete the acquisition of raw data.

[0098] The edge computing layer includes an FPGA development board for performing edge computing on the data collected by the data acquisition layer and an industrial-grade router for transmitting data. The FPGA development board is equipped with an algorithm for building a comprehensive geometric model, and the industrial-grade router uploads the results calculated by the FPGA development board to the cloud analysis layer. The edge computing layer corresponds to the data preprocessing in step S1 and the preliminary construction of the geometric model in step S2 in Example 1, and reduces data transmission delay through local transmission;

[0099] The cloud analysis layer is installed in the server and receives data from the edge computing layer via HTTPS encryption. The cloud analysis layer includes a comprehensive geometric model, a comprehensive risk index calculation model for risk warning and decision output. The cloud analysis layer corresponds to the functions of steps S2 to S5 in Example 1, completing the comprehensive risk index calculation, dynamic weight optimization and risk level classification, combined with the blockchain evidence storage and warning strategy generation in step S4;

[0100] The application layer includes dynamically displaying the temperature circle and the ejection edge on the Web interface. The application layer corresponds to the mixed reality visualization, maintenance instruction issuance and audit interface of step S4 in Example 1, realizing the terminal implementation of risk decision-making;

[0101] The data detected by the data acquisition layer is transmitted to the FPGA development board of the edge computing layer to calculate R(T) and the coordinates of the ejection point in real time. The calculation results are uploaded by the FPGA development board to the cloud analysis layer through an industrial-grade router for analysis. After the analysis of the results of the cloud analysis layer and the edge computing is completed, the corresponding early warning decision is sent to the application layer. Maintenance personnel maintain the pipes according to the early warning decision, and optimize and verify the models and algorithms in the cloud analysis layer based on the actual data of the pipes.

[0102] In this embodiment, a comprehensive evaluation method for quantitative risk of external corrosion of Ti65 pipes provided in Example 1 is implemented through a layered architecture to achieve full-process intelligent management: the data acquisition layer collects multi-source data in real time according to the dynamic characteristics of parameters; the edge computing layer uses Kalman filtering / wavelet transform to reduce noise and construct a temperature-dominant geometric model with a circular base surface + three-parameter 120° ejection point to reduce transmission delay through local calculation; the cloud analysis layer integrates nonlinear weighted CRI index calculation, dynamic weight optimization and blockchain evidence storage to generate a red / yellow / green graded warning strategy; the application layer Figure 3 The model in the system is visualized to accurately reflect changes in temperature, chloride ion concentration, microbial density and pipeline stress. In terms of system effect, the edge processing of the FPGA development board increases the efficiency of geometric modeling by 50%, blockchain ensures that data cannot be tampered with, and the mixed reality-assisted operation and maintenance response efficiency is increased by 40%. Dynamic model optimization controls the long-term prediction error within ±10%, forming a closed-loop anti-corrosion system from data perception to intelligent decision-making, and fully realizing the full life cycle management goal of "monitoring-early warning-optimization-traceability" mentioned in the Q&A.

[0103] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any person skilled in the art may utilize the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes for application in other fields. However, any simple modification, equivalent change, and modification of the above embodiments made in accordance with the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A comprehensive evaluation method for quantitative risk of external corrosion of Ti65 pipes, characterized by: The following steps are involved: S1. Data acquisition and preprocessing: Collect the temperature field distribution data, chloride ion concentration, microbial density, and stress dynamic distribution data of the pipe diameter, then perform noise reduction on the temperature field distribution data, chloride ion concentration data, microbial density data, and stress dynamic distribution data on the pipe surface. Finally, perform normalization on the noise-reduced temperature, stress, chloride ion concentration, and microbial signals. S2. Construction of comprehensive geometric model: Based on temperature data, supplemented by chloride ion concentration, microbial signals and stress, a comprehensive geometric model based on data changes is constructed; S3. Calculation of comprehensive risk index and classification: First, calculate the comprehensive risk index, and then classify the risk levels. The calculation formula for the comprehensive risk index is: Among them, CRI is the comprehensive risk index, A 顶出 is the ejection area, A0 is the original circular area; The comprehensive risk level classification is as follows: Among them, low risk, medium risk, and high risk levels are represented by green, yellow, and red respectively; S4. Risk warning and decision output: First, perform real-time visualization, then set up automatic warning and maintenance strategies, and finally perform data storage and traceability. When optimizing the weight coefficient, first determine the objective function: minimize the error between the predicted CRI and the actual corrosion rate: Then, the gradient descent method is used to update the weights every quarter; During the model verification process, laboratory calibration is used to verify the linear relationship between CRI and corrosion rate in a simulated environment, and then calibration is performed on site. During calibration, ultrasonic thickness measurement is regularly compared with the model prediction value. If the error exceeds 15%, the model is retrained. S5. Dynamic optimization and model validation: First, the weights of chloride ion concentration, microbial signals, and stress data in the calculation are optimized, and then the model is validated and calibrated; In step S2, a comprehensive geometric model is constructed with temperature as the main factor and chloride ion concentration, microbial density and stress as additional conditions. Specifically: S2.

1. Temperature-based basic radius calculation: R(T)=R0·(1+αΔT),ΔT=T-T base ; Where R(T) is the base radius at the current temperature T, T is the current temperature, and ΔT is the difference between the current temperature T and the reference temperature T. base The difference between the two, R0 is the reference temperature T base The corresponding initial radius; T base is the reference temperature; α is the temperature sensitivity coefficient; S2.

2. Calculation of additional radius for chloride ion concentration, microbial density, and stress: First, perform weighted calculations on chloride ion concentration, microbial density, and stress to generate three additional radius increments: Among them, when allocating weights, the default values ​​are ω1 = 0.5, ω2 = 0.3, and ω3 = 0.2, which will be adjusted during dynamic optimization; S2.

3. Calculation of ejection point coordinates based on chloride ion concentration, microbial density, and stress values: Add the chloride ion concentration, microbial density, and stress to three radii with an angle of 120°, and then calculate the coordinates of the ejection points in each direction: S2.

4. Calculation of ejection area: First, use the Shoelace formula to calculate the area of ​​the quadrilateral formed by the three ejection points and the origin: Where x4=x1, y4=y1; Among them, the area of ​​the original circle is: A0=πR(T) 2 .

2. A Ti65 pipe external corrosion quantitative risk comprehensive evaluation method according to claim 1, characterized in that: In step S1, when performing data collection and preprocessing, the specific process is as follows: S1.

1. Basic parameter data collection: Use an infrared thermal imager to scan the pipe surface at a frequency of 10 Hz to obtain the temperature field distribution data of the pipe. Use an electrochemical multi-parameter probe to measure the chloride ion concentration in the pipe environment at a frequency of 1 Hz. Use a fluorescent microbial sensor to detect the sulfate-reducing bacteria density in the pipe environment every 10 seconds. Deploy a fiber Bragg grating strain sensor array to monitor the dynamic stress distribution on the pipe surface at a frequency of 100 Hz; S1.

2. Raw data denoising: Kalman filtering is used to denoise the raw data of temperature and stress, and wavelet transform is used to denoise the raw data of chloride ion concentration and microbial signals; S1.

3. Normalization of raw data: Normalize the temperature, stress, chloride ion concentration, and microbial signal after noise reduction: Convert each parameter into a ratio of the relative benchmark value: Where: C is the chloride ion concentration, M is the microbial density, S is the stress value; C base is the reference value of chloride ion concentration; M base is the microbial density benchmark value; S yield is the stress reference value.

3. The method for comprehensive evaluation of the quantitative risk of external corrosion of Ti65 pipes according to claim 1, characterized in that: In step S4, when performing real-time visualization, the temperature circle and ejection edge are rendered on the monitoring interface, the color changes according to the risk level, and corresponding maintenance strategies are set according to low risk, medium risk and high risk levels. At the same time, the CRI calculation results, early warning records, and maintenance operations are uploaded to the chain to ensure that they cannot be tampered with, and are connected to the third-party audit interface.

4. A comprehensive evaluation system for the quantitative risk of external corrosion of Ti65 pipes, according to a comprehensive evaluation method for the quantitative risk of external corrosion of Ti65 pipes according to any one of claims 1 to 3, characterized in that: It includes a data collection layer, an edge computing layer, a cloud analysis layer, and an application layer. The data collected by the data collection layer is uploaded to the edge computing layer for edge computing. The calculation results of the edge computing layer are then uploaded to the cloud analysis layer for comprehensive risk index calculation, risk warning, and decision output. Finally, the analysis results of the cloud analysis layer are sent to the application layer for notification. Specifically: The data acquisition layer includes an infrared thermal imager for collecting temperature data, an electrochemical multi-parameter probe for detecting chloride ion concentration, a fluorescent microbial sensor for detecting the density of sulfate-reducing bacteria in the environment where the pipe is located, and a fiber grating strain sensor array for monitoring the dynamic stress distribution on the pipe surface; The edge computing layer includes an FPGA development board for performing edge computing on the data collected by the data acquisition layer and an industrial-grade router for transmitting data. The FPGA development board is equipped with an algorithm built using a comprehensive geometric model, and the industrial-grade router uploads the results calculated by the FPGA development board to the cloud analysis layer. The cloud analysis layer is installed in the server and receives data from the edge computing layer through HTTPS encryption. The cloud analysis layer includes a comprehensive geometric model, a comprehensive risk index calculation model for risk warning and decision output; The application layer includes dynamically displaying the temperature circle and the ejection edge on the Web interface.

5. The Ti65 pipe external corrosion quantitative risk comprehensive assessment system according to claim 4 is characterized by: The data detected by the data acquisition layer is transmitted to the FPGA development board of the edge computing layer to calculate R(T) and the coordinates of the ejection point in real time. The calculation results are uploaded by the FPGA development board to the cloud analysis layer through the industrial-grade router for analysis. When the results of the cloud analysis layer and the edge computing are analyzed, the corresponding early warning decision is sent to the application layer. The maintenance personnel maintain the pipes according to the early warning decision, and optimize and verify the models and algorithms in the cloud analysis layer according to the actual data of the pipes.

Images