Dynamic early warning and maintenance strategy adjustment method based on residual life

Abstract

The invention relates to a dynamic early warning and maintenance strategy adjusting method based on residual life, and belongs to the technical field of equipment management in the oil refining and chemical engineering field. The method comprises the following steps: acquiring multi-source operation data of equipment through an online acquisition module, and constructing a structured equipment state parameter set; based on the parameter set, establishing a residual life admission judgment model and generating a residual life accounting result of the current equipment; further adopting a corrosion rate-residual life two-factor coupling model to generate an equipment risk influence rate curve, and constructing a dynamic risk prevention and control boundary; and according to the risk level of boundary division, a dynamic maintenance strategy adjustment module is linked to perform early warning response, strategy execution, response recording and self-learning optimization are realized through a closed-loop management module, real-time evolution of a fortification value and a maintenance plan along with an equipment state is realized, and the accuracy of risk response and the intelligent level of equipment maintenance are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the equipment management technology in the field of oil refining and chemical industry, and particularly to a dynamic early warning and inspection and repair strategy adjustment method based on residual life, which is used for coping with the corrosion risk caused by the deterioration of crude oil. BACKGROUND

[0002] With the continuous deterioration of crude oil resources in the oil refining and chemical industry, the content of corrosion components such as sulfur, acid value and nitrogen in the raw material is continuously increasing, which leads to the intensified corrosion risk of equipment and pipelines. As early as 2008, the industry proposed an evaluation method for the set-off value after the deterioration of crude oil, which sets static boundary values for parameters such as equipment material selection and corrosion resistance to carry out protection management. However, the traditional set-off method ignores dynamic factors such as material performance degradation and corrosion rate change of equipment in the long-term service process, which leads to lagging prevention and control measures, and there is a risk of safety hazards and economic losses. Therefore, a new technology is urgently needed, which can realize real-time sensing of equipment state changes, dynamic adjustment of set-off boundaries based on residual life, and systematic prevention and control combined with early warning management platform to generate dynamic early warning of equipment residual life and inspection and repair strategy. SUMMARY

[0003] To solve the above problems in the prior art, the present application provides a dynamic early warning and inspection and repair strategy adjustment method based on residual life, The object of the present application can be achieved by the following technical solutions: S1: obtaining current operation data through an online acquisition module, and performing feature processing on the operation data according to equipment state fusion technology to construct a structured equipment state set of the current equipment; S2: establishing a residual life access determination model with the operation data as parameters based on the structured equipment state set, and generating a residual life accounting result of the current equipment according to the structured equipment state set; S3: according to the residual life accounting result, combining the current operation state of the equipment, using a corrosion rate-residual life double-factor coupling model to construct a residual life-corrosion rate-equipment risk influence rate curve, and generating a risk prevention and control boundary of the current adjustment strategy based on the residual life-corrosion rate-equipment risk influence rate curve; S4: performing risk early warning on the risk level set by the risk prevention and control boundary through a dynamic inspection and repair strategy adjustment module, establishing a mapping relationship between the risk early warning risk level and the inspection and repair adjustment strategy, and realizing the linkage of risk early warning and inspection and repair strategy based on a dynamic boundary setting mechanism.

[0004] Specifically, the online acquisition module comprises a distributed control system and a laboratory information management system, the distributed control system acquires real-time running process parameters of equipment through a sensor network distributed in a facility to be inspected, and the laboratory information management system acquires corrosion-related test data through a data interface connected with an analysis device.

[0005] Specifically, the structured equipment state set is obtained by standardizing and fusing the running data acquired by the online acquisition module, so as to provide a unified data input interface for the remaining life access determination model.

[0006] Specifically, the corrosion rate-remaining life double-factor coupling model is constructed based on a quadratic regression algorithm, a variable cost sensitivity factor and a dynamic risk influence rate optimization mechanism are added, a joint mapping relationship between the corrosion rate and the remaining life and the risk influence rate is established to generate an equipment risk evaluation curve, and the specific process comprises the following steps: , A cost sensitivity factor is introduced, a dynamic risk influence rate optimization mechanism, wherein λ is a cost sensitivity factor corresponding to the type of the equipment, a, b, c, d, e and f are fitting coefficients calculated based on historical equipment failure data, Δv is an actual corrosion change trend in a current period, ΔR is a life change trend, γ1 and γ2 are dynamic risk adjustment coefficients, if the corrosion rate is accelerated, Δv>0, the risk is adjusted upward, and if the remaining life is extended, ΔR>0, the risk is adjusted downward.

[0007] Specifically, the risk prevention and control boundary is constructed in the form of a two-dimensional risk matrix, wherein the horizontal axis is the corrosion rate and the vertical axis is the remaining life, a dynamic early warning and closed-loop processing mechanism is constructed by dividing corresponding prevention threshold values and inspection and repair rules.

[0008] Specifically, the remaining life access determination model integrates a multivariate support vector machine and a data-driven random process model to realize small-sample high-precision prediction, and the specific formula comprises: , wherein k is a fold number, n i is the number of test set samples, y j is a real label and a predicted label. The optimization model structure is as follows: , wherein Loss(n, xj) is the contribution of the feature xj to the loss function at the node n.

[0009] Specifically, the dynamic maintenance strategy adjustment module obtains the risk warning by identifying the risk level of the equipment currently in and combining the remaining life trend and corrosion rate dynamic parameter analysis.

[0010] Specifically, the mapping relationship is used to establish the corresponding rules between the risk level and the maintenance strategy, and different risk levels correspond to different maintenance response strategies, including inspection frequency, detection method, maintenance cycle and resource scheduling level, which supports dynamically adjusting the mapping rules through historical operation data.

[0011] Specifically, the dynamic maintenance strategy module further comprises a closed-loop management module, which realizes intelligent risk closed-loop management of the whole life cycle of the equipment by tracking and recording risk warning response, maintenance execution process and result feedback information.

[0012] Specifically, the dynamic boundary setting mechanism dynamically compresses and relaxes the security boundary of the equipment according to the risk level to build a dynamic threshold curve, and the static security is converted to adaptive risk control through the dynamic threshold curve.

[0013] The beneficial effects of the present application are: Integrate various monitoring parameters, including process parameters obtained through the DCS interface, laboratory data provided by LIMS and online thickness measurement data, construct a material selection adaptability-corrosion rate-remaining life three-dimensional evaluation system, compared with the traditional single parameter monitoring, the system can fully reflect the equipment state, when the material selection meets the guide but the remaining life is less than 1 year, the system can identify the potential risk and trigger the warning, avoid ignoring the overall risk due to the normal single parameter, through the life prediction model based on real-time data, dynamically calculate the remaining life, timely reflect the equipment state change, compared with the static prediction model relying on historical data, the system updates the real-time data, the life prediction error is controlled in a lower range, when the corrosion rate increases suddenly, the system can update the remaining life prediction within 24 hours, significantly improve the prediction accuracy, provide reliable basis for maintenance decision, the real-time monitoring and rapid response warning system, through the red, yellow and blue three-level warning mechanism, the risk information is pushed to the operation center and the maintenance personnel in real time. Compared with the traditional monthly inspection mode, the system accesses the DCS real-time data and online monitoring, the risk identification time is improved to within 24 hours, the response speed is improved by more than 10 times, when the corrosion rate is more than 0.35mm / a, the system immediately pushes the red warning and triggers the emergency maintenance, significantly reduces the unplanned downtime and safety accident risk. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to facilitate those skilled in the art to understand, the present application will be further described below in combination with the drawings.

[0015] Figure 1A flowchart of a dynamic early warning and inspection and maintenance strategy adjustment method based on residual life according to the present application.

[0016] Figure 2 A dynamic adjustment flowchart of a dynamic early warning and inspection and maintenance strategy adjustment method based on residual life according to the present application. DETAILED DESCRIPTION

[0017] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined application purposes, the specific embodiments, structures, features and effects according to the present application are described in detail below in combination with the drawings and preferred embodiments.

[0018] Please refer to Figure 1 A dynamic early warning and inspection and maintenance strategy adjustment method based on residual life, comprising the following steps: S1: obtaining current operation data through an online acquisition module, and performing feature processing on the operation data according to a device state fusion technology to construct a structured device state set of the current device; S2: establishing a residual life access determination model with the operation data as parameters based on the structured device state set, and generating a residual life accounting result of the current device according to the structured device state set; S3: constructing a residual life and corrosion rate-device risk influence rate curve by using a corrosion rate-residual life double-factor coupling model according to the residual life accounting result and in combination with the current operation state of the device, and generating a risk prevention and control boundary of the current adjustment strategy based on the residual life and corrosion rate-device risk influence rate curve; S4: performing risk early warning on the risk level set by the risk prevention and control boundary through a dynamic inspection and maintenance strategy adjustment module, establishing a mapping relationship according to the risk early warning risk level and the inspection and maintenance adjustment strategy, and realizing the linkage of risk early warning and inspection and maintenance strategy based on a dynamic boundary setting mechanism.

[0019] Specifically, the online acquisition module comprises a distributed control system and a laboratory information management system, the distributed control system acquires real-time operation process parameters of the device through a sensor network distributed in the facility to be inspected, and the laboratory information management system acquires corrosion-related test data through a data interface connected with an analysis device.

[0020] In the atmospheric tower shell corrosion management embodiment, the system obtains the actual thickness of the shell as 15 mm through the online thickness measurement interface, and the assay data interface shows that the sulfur content of the crude oil reaches 2.5%, which is 0.5% higher than the normal value. The process operating parameter interface records that the operating temperature is 350℃ and the pressure is 0.58MPa. These data provide real-time and comprehensive basic information for subsequent evaluation. Based on the collected data, the system analyzes the shell corrosion rate and measures the current corrosion rate as 0.3mm / a, which exceeds the risk threshold set by the enterprise of 0.25mm / a. Through calculation, the system evaluates that the remaining life of the shell is about 39 years, but due to the excessive corrosion rate, further risk evaluation is triggered. The dynamic defense module marks the equipment as a medium-high risk area according to the corrosion rate and remaining life data, combined with the risk matrix. The system automatically tightens the defense threshold from 0.25mm / a to 0.2mm / a, forming an adaptive risk prevention and control boundary to ensure the timely discovery of potential corrosion risks. The intelligent early warning module generates a yellow warning, prompting "excessive corrosion rate, suggest increasing detection frequency". The warning information is pushed in real time through the operation center display screen and the maintenance personnel's mobile phone APP, ensuring that relevant personnel can learn about the equipment status in a timely manner. The decision support module suggests conducting thickness measurement inspection once a month according to the medium-high risk level, and optimizes the process parameters, such as reducing the operating temperature to 340℃ to slow down corrosion. The closed-loop management module records abnormal events, including the triggering time, risk level and rectification measures. The thickness measurement data of the next month shows that the corrosion rate has decreased to 0.22mm / a, and the risk level has decreased to medium. The system adjusts to quarterly inspection, optimizing resource allocation. Through real-time data collection and dynamic defense, the system accurately identifies the corrosion risk of the atmospheric tower shell, timely adjusts the defense boundary and inspection and maintenance strategy, avoiding potential corrosion leakage accidents. The optimized operating parameters effectively slow down the corrosion rate and extend the service life of the equipment by about 2 years, saving about 1,000,000 yuan of unplanned downtime cost. At the same time, the detection frequency is adjusted from monthly to quarterly, reducing maintenance costs and improving production efficiency. This embodiment demonstrates the system's ability to ensure safe operation of equipment through multi-parameter evaluation and rapid response in complex operating environments, providing a reliable risk management solution for the refining industry.

[0021] Specifically, the structured equipment state set provides a unified data input interface for the remaining life access determination model by standardizing and fusing the operating data obtained by the online collection module.

[0022] Specifically, the corrosion rate-remaining life double-factor coupling model is constructed based on a quadratic regression algorithm. By adding a variable cost sensitivity factor and a dynamic risk influence rate optimization mechanism, a joint mapping relationship between corrosion rate and remaining life and risk influence rate is established to generate an equipment risk evaluation curve, which specifically includes: a basic quadratic regression model, , A cost sensitivity factor is introduced, a dynamic risk impact rate optimization mechanism, wherein λ is a cost sensitivity factor corresponding to the device type, a, b, c, d, e, and f are fitting coefficients calculated from historical device failure data, Δv is the actual corrosion trend in the current period, ΔR is the life trend, γ1 and γ2 are dynamic risk adjustment coefficients If the corrosion rate increases Δv>0, the risk is adjusted upward; if the remaining life is extended ΔR>0, the risk is adjusted downward.

[0023] Specifically, the risk prevention and control boundary is constructed in the form of a two-dimensional risk matrix, wherein the horizontal axis is the corrosion rate and the vertical axis is the remaining life, and a dynamic early warning and closed-loop processing mechanism is constructed by dividing the corresponding defense threshold and inspection and repair rules.

[0024] Specifically, the remaining life access determination model integrates a multivariate support vector machine and a data-driven random process model to achieve high-precision prediction with small samples, and the specific formula includes: , wherein k is the number of folds, n i is the number of test set samples, y j is the true label and the predicted label. The optimization model structure is: , wherein Loss(n, xj) is the contribution of feature xj to the loss function at node n.

[0025] Specifically, the dynamic inspection and repair strategy adjustment module obtains the risk warning by identifying the risk level of the device currently in and combining the remaining life trend and corrosion rate dynamic parameter analysis.

[0026] Specifically, the mapping relationship is used to establish the corresponding rules between the risk level and the inspection and repair strategy, and different risk levels correspond to different maintenance response strategies, including inspection frequency, detection method, repair period, and resource scheduling level, which supports dynamically adjusting the mapping rules through historical operation data.

[0027] In the remaining life warning embodiment of the oil pipeline, the online thickness measurement interface displays that the actual thickness of the pipeline is 6.5 mm, and the analysis data interface shows that the acid value of the crude oil reaches 3.5 mgKOH / g, which exceeds the normal range. The process operation parameter interface records that the running pressure is 3.8 MPa, and the flow rate is slightly high. These data provide a comprehensive corrosion risk assessment basis for the system, analyze the pipeline corrosion rate, and measure the current corrosion rate as 0.4 mm / a, which is far beyond the enterprise risk threshold of 0.25 mm / a. Based on the wall thickness data, the system assesses that the remaining life of the pipeline is about 9 years, but due to the significantly excessive corrosion rate, a high-risk warning is triggered, and combined with the risk matrix, the pipeline is marked as a high-risk state, and the defense threshold is tightened to 0.15 mm / a. At the same time, the system analyzes the material selection adaptability and suggests replacing the pipeline with a stainless steel material with higher acid resistance, forming a dynamic risk prevention and control boundary, prompting that the “pipeline remaining life is insufficient and needs to be repaired immediately”, and the warning information is pushed in real time through the operation center and the maintenance personnel's mobile devices, ensuring fast response, immediate shutdown and replacement of the damaged pipeline section, and optimization of process parameters, such as reducing the flow rate to reduce corrosion. The closed-loop management module records the rectification process, including downtime, replacement of materials and subsequent monitoring data. After the pipeline is replaced, the corrosion rate is reduced to 0.1 mm / a, the remaining life is extended to more than 20 years, the risk level is reduced to low, and the regular inspection is restored. Through real-time monitoring and dynamic defense, high corrosion risks of oil pipelines are quickly identified, red alerts and emergency repairs are triggered in time, pipeline leakage accidents are avoided, production safety and environmental compliance are ensured. After replacing the pipeline, the corrosion rate is significantly reduced, the equipment service life is extended by about 12 years, and the potential economic loss is saved by about 3 million yuan. Closed-loop management ensures that the rectification process is traceable, and the maintenance and repair resources are optimized.

[0028] Specifically, the dynamic maintenance and repair strategy module further includes a closed-loop management module, which realizes intelligent risk closed-loop management of the entire life cycle of the equipment by tracking and recording risk warning response, maintenance and repair execution process and result feedback information.

[0029] Specifically, the dynamic boundary setting mechanism dynamically compresses and relaxes the defense boundary of the equipment according to the risk level to construct a dynamic threshold curve, and converts the static defense to adaptive risk control through the dynamic threshold curve.

[0030] In the reactor head corrosion prevention and control embodiment, the online thickness measurement interface displays that the actual thickness of the head is 12 mm, and the test analysis data interface shows that the nitrogen content is 0.4%, which is higher than the normal value. The process operation parameter interface records that the running temperature is 400 DEG C and the pressure is 2.4 MPa. These data provide accurate corrosion risk assessment basis for the system, and the measured corrosion rate of the head is 0.28 mm / a, which is slightly higher than the enterprise risk threshold 0.25 mm / a. Based on the wall thickness data, the remaining life is about 33 years, but because the corrosion rate is over standard, the medium risk assessment is triggered, and combined with the risk matrix, the head is marked as medium risk state, and the defense threshold is adjusted to 0.22 mm / a, forming a dynamic risk prevention and control boundary, ensuring timely capture of corrosion trend changes, generating a blue warning, prompting "corrosion rate close to threshold, suggest increasing detection frequency", the warning information is pushed in real time through the operation center and the maintenance personnel's mobile phone APP, promoting rapid response, the decision support module suggests increasing the thickness measurement frequency to once a month, and recommends adding corrosion inhibitor to reduce the corrosion rate, the closed-loop management module records the rectification measures and effects: after adding the corrosion inhibitor, the corrosion rate is reduced to 0.18 mm / a, the risk level is reduced to low level, and the detection frequency is adjusted to once every quarter, optimizing maintenance resources, through multi-parameter data acquisition and dynamic defense, the system effectively identifies the corrosion risk of the reactor head, the blue warning and process optimization measures timely slow down the corrosion rate, prolong the service life of the head by about 3 years, and save the detection and maintenance cost of about 500,000 yuan. The closed-loop management module records the rectification process to form a traceable data archive to support subsequent optimization.

[0031] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make slight changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any simple modification, equivalent change and modification of the above embodiments based on the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. A method for dynamic pre-warning and maintenance strategy adjustment based on residual life, characterized in that, The method comprises the following steps: S1: obtaining current operation data through an online acquisition module, and performing feature processing on the operation data according to a device state fusion technology to construct a structured device state set of the current device; S2: establishing a residual life access judgment model with the operation data as a parameter based on the structured device state set, and generating a residual life accounting result of the current device according to the structured device state set; S3: constructing a residual life and corrosion rate-device risk influence rate curve by using a corrosion rate-residual life double-factor coupling model according to the residual life accounting result and the current operation state of the device, and generating a risk prevention and control boundary of the current adjustment strategy based on the residual life and corrosion rate-device risk influence rate curve; S4: performing risk early warning on the risk level set by the risk prevention and control boundary through a dynamic inspection and maintenance strategy adjustment module, establishing a mapping relationship between the risk early warning risk level and the inspection and maintenance adjustment strategy, and realizing the linkage of risk early warning and inspection and maintenance strategy based on a dynamic boundary setting mechanism.

2. The method of claim 1, wherein, The online acquisition module comprises a distributed control system and a laboratory information management system, the distributed control system acquires real-time operation process parameters of the device through a sensor network distributed in the facility to be inspected, and the laboratory information management system acquires corrosion-related test data through a data interface connected with an analysis device.

3. The method of claim 1, wherein, The structured device state set is obtained by standardizing and fusing the operation data acquired by the online acquisition module, and provides a unified data input interface for the residual life access judgment model.

4. The method of claim 1, wherein, The corrosion rate-residual life double-factor coupling model is constructed based on a quadratic regression algorithm, and a device risk evaluation curve is generated by establishing a joint mapping relationship of corrosion rate and residual life on risk influence rate.

5. The method of claim 4, wherein, The risk prevention and control boundary is constructed in the form of a two-dimensional risk matrix, with corrosion rate as the horizontal axis and residual life as the vertical axis, and a dynamic early warning and closed-loop processing mechanism is constructed by dividing corresponding defense threshold values and inspection and maintenance rules.

6. The method of claim 2, wherein, The residual life access judgment model is based on a minimum pressure-bearing wall thickness calculation formula and a residual life evaluation formula, a multi-parameter fitting is performed on the operation data by a dynamic interpolation method, and the residual service life of the device is evaluated according to the fitted operation data.

7. The method of claim 1, wherein, The dynamic inspection and maintenance strategy adjustment module obtains the risk early warning by identifying the current risk level of the device and combining residual life trend analysis and corrosion rate dynamic parameter analysis.

8. The method of claim 1, wherein, The mapping relationship is used to establish the corresponding rules between the risk level and the inspection and maintenance strategy, different risk levels correspond to different maintenance response strategies, including inspection frequency, detection method, maintenance cycle and resource scheduling level, and the mapping rules can be dynamically adjusted through historical operation data.

9. The method of claim 2, wherein, The dynamic inspection and maintenance strategy module further comprises a closed-loop management module, which realizes intelligent risk closed-loop management of the device throughout its life cycle by tracking and recording risk early warning response, inspection and maintenance execution process and result feedback information.

10. The method of claim 4, wherein, The dynamic boundary setting mechanism dynamically compresses and relaxes the security boundary of the device according to the risk level to build a dynamic threshold curve, and the static security is converted to adaptive risk control through the dynamic threshold curve.

Images