Intelligent maintenance decision method and system for refining and chemical equipment, electronic device and storage medium

Abstract

The application discloses a kind of intelligent maintenance decision methods of refining equipment, comprising the following steps: S110 obtains the real-time monitoring data of each measuring point of refining equipment;S120 input the real-time monitoring data obtained into early warning model, determine whether equipment appears early fault warning, if yes, then enter step S130, if no, then return to step S110;S130 obtains equipment health degree evaluation grade and equipment dynamic risk evaluation grade according to health degree evaluation model and dynamic risk evaluation model respectively;And S140 carries out maintenance decision according to the evaluation grade obtained.The application also discloses a kind of intelligent maintenance decision system of refining equipment, electronic equipment and storage medium.The application combines early warning model of equipment based on real-time monitoring data driving, health degree evaluation model based on failure possibility influence factor and data driving, and dynamic risk evaluation model based on health degree real-time evaluation and fault consequence, so that accurate maintenance decision can be made.

Description

Technical Field

[0001] This invention relates to the field of risk monitoring technology for refining and chemical equipment, and in particular to an intelligent maintenance decision-making method, system, electronic device, and storage medium for refining and chemical equipment. Background Technology

[0002] Equipment is the primary carrier of risk in refining and chemical enterprises. While the overall operation of refining and chemical units is stable, unplanned shutdowns occur frequently, and abnormal production fluctuations are continuous. Statistics show that unplanned shutdowns caused by equipment issues remain the main factor in recent years. Specifically, the failure rate of large units has increased significantly, instrument malfunctions have led to more unplanned shutdowns, and power system failures, power fluctuations, and high-pressure heat exchanger leaks are frequent occurrences. Potential equipment risks have still not been assessed in a timely manner or effectively managed.

[0003] Traditional equipment risk assessment techniques rely on the consequences and probability of failure modes. The probability of failure is derived from historical data using probabilistic statistical analysis, resulting in a static risk model. This method is ill-suited for assessing operational risks. To effectively control equipment risk, it is essential not only to understand the probability and severity of failures but also to grasp the volatility of equipment failures and establish a dynamic risk model. Dynamic risk assessment techniques for equipment are still in their early stages both domestically and internationally, with few mature methods available.

[0004] Traditional health status assessments employ the ABCD classification method based on vibration intensity, failing to consider the significant differences between equipment. Health indices derived from statistical data are difficult to normalize and accurately measure, and cannot address issues such as equipment failure and downtime in zone A versus continued good operation in zone D. Furthermore, there is a lack of direct equipment health status assessment results to guide production activities. Acquiring engineering case data is time-consuming, engineering simulation experiments are costly, and research hypotheses are difficult to formulate; therefore, the established models are unvalidated, and no equipment health rating criteria have been established.

[0005] Traditional early fault detection technologies for industrial applications rely on complex fault mechanism analysis, fault diagnosis, signal analysis and processing, and external expert prior knowledge. They cannot automatically detect early faults without the need for external expert experience or prior knowledge. Traditional alarm methods based on fixed thresholds suffer from numerous false alarms, missed alarms, and repeated alarms for the same event, easily leading to alarm fatigue among operators or maintenance personnel.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] One of the objectives of this invention is to provide an intelligent maintenance decision-making method, system, electronic device, and storage medium for refining and chemical equipment, thereby overcoming the problem that existing technologies rely on external expert experience or prior knowledge to identify equipment faults and risks and make maintenance decisions.

[0008] Another objective of this invention is to provide a method, system, electronic device, and storage medium for intelligent maintenance decision-making of refining and chemical equipment, thereby improving the problems of missed alarms and false alarms in equipment maintenance decision-making.

[0009] To achieve the above objectives, according to a first aspect of the present invention, the present invention provides an intelligent maintenance decision-making method for refining and chemical equipment, comprising the following steps:

[0010] S110 acquires real-time monitoring data from various measuring points of the refining and chemical equipment;

[0011] S120 inputs the acquired real-time monitoring data into the early warning model to determine whether the device has an early fault warning. If yes, proceed to step S130; otherwise, return to step S110.

[0012] S130 obtains the equipment health assessment level and the equipment dynamic risk assessment level based on the health assessment model and the dynamic risk assessment model, respectively; and

[0013] S140 makes maintenance decisions based on the equipment health assessment level and the equipment dynamic risk assessment level.

[0014] Furthermore, in the above technical solution, step S140 includes:

[0015] If at least one of the equipment health assessment level and the equipment dynamic risk assessment level corresponds to a shutdown operation, a maintenance alarm will be triggered; and

[0016] Otherwise, return to step S110.

[0017] Furthermore, in the above technical solution, the real-time monitoring data of refining and chemical equipment includes corrosion-sensitive characteristic parameters and process operating parameters of static equipment, original waveform data of vibration monitoring of dynamic equipment, and static RBI and static RCM evaluation results.

[0018] Furthermore, in the above technical solution, the corrosion-sensitive characteristic parameters of the static equipment include operating temperature, sulfur content of distillate oil, acid value of distillate oil, material, cyanide concentration, pH value, and hydrogen sulfide content in water.

[0019] Furthermore, in the above technical solutions, the early warning models include the WPD+DKPCA intelligent early warning model, the XJH-1 model, and the XJH-2 model.

[0020] Furthermore, in the above technical solution, the rule for determining whether the equipment has an early fault warning is as follows: perform model calculation on the real-time monitoring data of each measuring point to obtain the prediction result of the current early warning model for each measuring point; if N consecutive warning records appear under any early warning model, it is determined that the measuring point has an early fault warning; and if any measuring point has an early fault warning, it is determined that the equipment has an early fault warning.

[0021] Furthermore, in the above technical solution, the calculation of the WPD+DKPCA intelligent early warning model for the real-time monitoring data of the measuring point includes: training the model, which pre-trains the reference signal data to obtain the model feature values ​​of the device; calculating the model, which performs wavelet packet decomposition and dynamic kernel principal component analysis on the real-time raw vibration signal data of the measuring point to obtain the real-time feature values ​​T2 and SPE; and comparing the real-time feature values ​​with the model feature values. When the real-time feature values ​​T2 and SPE exceed the preset model feature values, an early warning record is made.

[0022] Furthermore, in the above technical solution, the XJH-1 model calculation of the real-time monitoring data of the measuring point includes: training a model, which extracts early fault sensitive feature values ​​from the original vibration signal data that meets the conditions and processes them using an improved l1 trend filter, and then inputs them into the SVDD to obtain the radius and center of the hypersphere; and calculating a model, which extracts early fault sensitive feature values ​​from the real-time original vibration signal data and real-time online monitoring data and processes them using an improved l1 trend filter, and then inputs them into the trained SVDD model. When the real-time model feature value exceeds the preset model training value, an early warning record is made.

[0023] Furthermore, in the above technical solution, the XJH-2 model calculation for the real-time monitoring data of the measuring points includes: training a model, which extracts early fault sensitive feature values ​​and processes them using an improved l1 trend filter, and then uses a self-learning method to obtain the alarm threshold line; and calculating a model, which extracts early fault sensitive feature values ​​from the real-time raw vibration signal data and processes them using an improved l1 trend filter, and when the real-time model feature value exceeds the alarm threshold line, it is recorded as an early warning record.

[0024] Furthermore, in the above technical solution, the health evaluation model includes a cross-correlation function, a cohesion function, a spectral distance function, and a corrosion model.

[0025] Furthermore, in the above technical solution, the corrosion model includes one or more of the following: high-temperature sulfur / naphthenic acid corrosion model, spheroidization model, graphitization model, and ammonium hydrosulfide corrosion model.

[0026] Furthermore, in the above technical solution, obtaining the equipment health evaluation level based on the health evaluation model includes: calculating the corrosion model based on the real-time corrosion characteristic parameter data of the measuring points of the static equipment, obtaining the failure probability influence coefficient k, and obtaining the current corrosion model health evaluation level based on the health function H = 1 / k and the range of H values ​​for each health level. The failure probability influence coefficient k represents the ratio of the key influencing parameter that causes the change in the equipment failure probability under different corrosion models to its design value.

[0027] Furthermore, in the above technical solutions, when using the high-temperature sulfur / naphthenic acid corrosion model, the real-time corrosion rate is calculated based on the operating temperature, oil sulfur content, acid value, and gaseous H2S content. Combined with the design corrosion rate, the failure probability influence coefficient k is obtained. When using the spheroidization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, the failure probability influence coefficient k is obtained. When using the graphitization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, the failure probability influence coefficient k is obtained. And when using the ammonium hydrosulfide corrosion model, the real-time corrosion rate is calculated based on the cyanide concentration. Combined with the design corrosion rate, the failure probability influence coefficient k is obtained.

[0028] Furthermore, in the above technical solution, obtaining the equipment health evaluation level according to the health evaluation model includes: calculating the cross-correlation function, cohesion function, and spectral distance function for the real-time raw vibration signal data of the moving equipment's measuring points, and obtaining the calculation results of each function; comparing the calculation results of each function with the health evaluation matrix to determine the evaluation results of each function; and combining a preset two-out-of-three voting model or conservative principle to obtain the health evaluation level of each measuring point of the moving equipment, thereby obtaining the health evaluation level of the moving equipment.

[0029] Furthermore, in the above technical solution, the dynamic risk assessment model includes a corrosion model and a dynamic risk assessment matrix.

[0030] Furthermore, in the above technical solution, obtaining the dynamic risk level of the equipment based on the dynamic risk assessment model includes: obtaining the dynamic risk level of the equipment based on the real-time health and importance level of the equipment and the dynamic risk assessment matrix.

[0031] Furthermore, in the above technical solution, the dynamic risk level of the equipment obtained according to the dynamic risk assessment model includes: if one of the high-temperature sulfur / naphthenic acid corrosion model, spheroidization corrosion model, graphitization corrosion model, and ammonium hydrosulfide corrosion model is used to calculate the real-time corrosion sensitivity parameter data of the measuring points of the static equipment, the real-time failure probability level is obtained, and combined with the failure consequence level in the evaluation system, the dynamic risk level of the measuring point under the current model is obtained; if the wet hydrogen sulfide corrosion model is used, the dynamic risk level of the measuring point under the current model is obtained based on the real-time pH value of the measuring point and the hydrogen sulfide content in the water, according to the wet hydrogen sulfide corrosion model matrix; and the dynamic risk level of the static equipment is the highest dynamic risk level among all measuring points.

[0032] Furthermore, in the above technical solution, when using the high-temperature sulfur / naphthenic acid corrosion model, the real-time corrosion rate is calculated based on the operating temperature, oil sulfur content, acid value, and gaseous H2S content. Combined with the design corrosion rate, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. The real-time failure probability level is obtained based on the range of values ​​for the failure probability level. When using the spheroidization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. The real-time failure probability level is obtained based on the range of values ​​for the failure probability level. The range of values ​​for the failure probability level is used to obtain the real-time failure probability level. When using the graphitization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. Based on the range of values ​​for the failure probability level, the real-time failure probability level is obtained. Similarly, when using the ammonium hydrosulfide corrosion model, the real-time corrosion rate is calculated based on the cyanide concentration. Combined with the design corrosion rate, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. Based on the range of values ​​for the failure probability level, the real-time failure probability level is obtained.

[0033] According to a second aspect of the present invention, the present invention provides an intelligent maintenance decision-making system for refining and chemical equipment, comprising: a data acquisition unit for acquiring real-time monitoring data of various measuring points of the refining and chemical equipment; an early fault warning unit for inputting the acquired real-time monitoring data into an early warning model to determine whether the equipment has an early fault warning; a model evaluation unit for obtaining the equipment health evaluation level and the equipment dynamic risk evaluation level respectively based on the judgment result of the early fault warning unit, according to a health evaluation model and a dynamic risk evaluation model; and a maintenance decision-making unit for making maintenance decisions based on the equipment health evaluation level and the equipment dynamic risk evaluation level.

[0034] According to a third aspect of the present invention, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform the intelligent maintenance decision-making method for refining and chemical equipment as described in any of the above technical solutions.

[0035] According to a fourth aspect of the present invention, the present invention provides a non-transitory computer-readable storage medium storing computer-executable instructions for causing a computer to execute a refining and chemical equipment intelligent maintenance decision-making method as described in any of the above technical solutions.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The intelligent maintenance decision-making method for refining and chemical equipment of the present invention organically combines an early warning model for equipment driven by real-time monitoring data, a health evaluation model driven by failure probability influencing factors and data, and a dynamic risk evaluation model based on real-time health evaluation and failure consequences. This enables accurate maintenance decisions, avoids missed alarms, false alarms, and alarm fatigue of operators or maintenance personnel.

[0038] 2. This invention enables data-driven automated maintenance decision-making, eliminating the need to rely on external expert experience. Potential risks of equipment can be controlled in a timely manner, providing timely guidance for enterprises to formulate maintenance strategies and carry out equipment maintenance and repair work, thereby achieving the goal of early detection, early warning, and early handling of equipment failures.

[0039] 3. This invention enables equipment risk management to shift from static analysis to dynamic perception, and from post-event emergency response to pre-event prevention. It can improve the inherent safety level of enterprise production, effectively avoid production losses, casualties, and social and environmental impacts caused by equipment accidents, and ensure long-term operation of the equipment.

[0040] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, and to make the above and other objects, technical features and advantages of the present invention easier to understand, one or more preferred embodiments are listed below and described in detail with reference to the accompanying drawings. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating an intelligent maintenance decision-making method for refining and chemical equipment according to an embodiment of the present invention.

[0042] Figure 2 This is a schematic diagram of an intelligent maintenance decision-making system for refining and chemical equipment according to an embodiment of the present invention.

[0043] Figure 3 A schematic diagram of the hardware structure of an electronic device for implementing an intelligent maintenance decision-making method for refining and chemical equipment according to an embodiment of the present invention. Detailed Implementation

[0044] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0045] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0046] In this document, for ease of description, spatial relative terms such as “below,” “under,” “down,” “above,” “above,” “up,” etc., are used to describe the relationship of one element or feature to another element or feature in the accompanying drawings. It should be understood that spatial relative terms are intended to encompass different orientations of an object in use or operation, in addition to those depicted in the figures. For example, if an object in the figure is flipped, an element described as “below” or “under” another element or feature would be oriented “above” that element or feature. Thus, the exemplary term “below” can encompass both the downward and upward orientations. An object may also have other orientations (rotated 90 degrees or other orientations), and the spatial relative terms used herein should be interpreted accordingly.

[0047] In this document, the terms "first," "second," etc., are used to distinguish two different elements or parts, and are not used to define specific positions or relative relationships. In other words, in some embodiments, the terms "first," "second," etc., can also be used interchangeably.

[0048] like Figure 1 As shown, the intelligent maintenance decision-making method for refining and chemical equipment according to a specific embodiment of the present invention includes the following steps:

[0049] S110 acquires real-time monitoring data from various measuring points of the refining and chemical equipment.

[0050] Furthermore, in one or more exemplary embodiments of the present invention, the real-time monitoring data of the refining and chemical equipment includes corrosion-sensitive characteristic parameters and process operating parameters of static equipment, vibration monitoring raw waveform data of dynamic equipment, and static RBI and static RCM evaluation results, etc.

[0051] Furthermore, in one or more exemplary embodiments of the present invention, the corrosion-sensitive characteristic parameters of the static equipment include temperature (operation, design), sulfur content of distillate oil (operation, design), acid value of distillate oil (operation, design), material, cyanide concentration (operation, design), pH value, and hydrogen sulfide content in water, etc.

[0052] S120 inputs the acquired real-time monitoring data into the early warning model to determine whether the device has an early fault warning. If yes, proceed to step S130; otherwise, return to step S110.

[0053] Furthermore, in one or more exemplary embodiments of the present invention, the early warning model includes the WPD+DKPCA intelligent early warning model, the XJH-1 model, and the XJH-2 model.

[0054] Furthermore, in one or more exemplary embodiments of the present invention, the rule for determining whether the device has an early fault warning is as follows: perform model calculation on the real-time monitoring data of each measuring point to obtain the current early warning model prediction result of each measuring point; if N warning records occur consecutively under any early warning model, it is determined that the measuring point has an early fault warning; and if any measuring point has an early fault warning, it is determined that the device has an early fault warning.

[0055] Furthermore, in one or more exemplary embodiments of the present invention, the calculation of the WPD+DKPCA intelligent early warning model on the real-time monitoring data of the measuring point includes: training the model, which pre-trains the reference signal data to obtain the model feature values ​​of the device; calculating the model, which performs wavelet packet decomposition and dynamic kernel principal component analysis on the real-time raw vibration signal data of the measuring point to obtain the real-time feature values ​​T2 and SPE; and comparing the real-time feature values ​​with the model feature values, and when the real-time feature values ​​T2 and SPE exceed the preset model feature values, an early warning record is made.

[0056] Furthermore, in one or more exemplary embodiments of the present invention, the XJH-1 model calculation of the real-time monitoring data of the measuring point includes: training a model, which extracts early fault sensitive feature values ​​from the original vibration signal data that meets the conditions and processes them using an improved l1 trend filter, and then inputs them into the SVDD to obtain the radius and center of the hypersphere; and calculating a model, which extracts early fault sensitive feature values ​​from the real-time original vibration signal data and real-time online monitoring data and processes them using an improved l1 trend filter, and then inputs them into the trained SVDD model. When the real-time model feature value exceeds the preset model training value, an early warning record is made.

[0057] Furthermore, in one or more exemplary embodiments of the present invention, the XJH-2 model calculation of the real-time monitoring data of the measuring point includes: training a model, which obtains an alarm threshold line by extracting early fault sensitive feature values ​​and applying an improved l1 trend filter; and calculating a model, which extracts early fault sensitive feature values ​​from the real-time raw vibration signal data and applies an improved l1 trend filter, and records an early warning record when the real-time model feature value exceeds the alarm threshold line.

[0058] Based on the health assessment model and the dynamic risk assessment model, S130 obtains the equipment health assessment level and the equipment dynamic risk assessment level, respectively.

[0059] Furthermore, in one or more exemplary embodiments of the present invention, the health assessment model includes a cross-correlation function, a cohesion function, a spectral distance function, and a corrosion model.

[0060] Furthermore, in one or more exemplary embodiments of the present invention, the corrosion model includes one or more of the following: a high-temperature sulfur / naphthenic acid corrosion model, a spheroidization model, a graphitization model, and an ammonium hydrosulfide corrosion model.

[0061] Furthermore, in one or more exemplary embodiments of the present invention, obtaining the equipment health evaluation level according to the health evaluation model includes: performing corrosion model calculations on the real-time corrosion characteristic parameter data of the measuring points of the static equipment to obtain the failure probability influence coefficient k; and obtaining the health evaluation level of the current corrosion model based on the health function H = 1 / k and the range of H values ​​for each health level, wherein the failure probability influence coefficient k represents the ratio of the real-time monitoring value of the key influencing parameter that causes the change in the equipment failure probability under different corrosion models to its design value.

[0062] For example, corrosion rate, operating hours, and remaining service life directly reflect the corrosion status of the equipment, defined as influencing factor f. Changes in each monitoring parameter are directly related to changes in the influencing factor. When calculating the magnitude of the influencing factor by acquiring multiple monitoring parameters, α1, α2, α3, ..., αn represent the real-time monitoring values ​​of each parameter, and the influencing factor is calculated as f(α1, α2, α3, ..., αn), where α10, α20, α30, ..., αn0 represent the design values ​​of each parameter. The equipment failure probability influence coefficient k is then expressed as:

[0063]

[0064] When multiple monitoring parameters are difficult to obtain, or when there are monitoring parameters that have a major impact on corrosion, a single monitoring parameter is selected to fit the relationship between the parameter and the influencing factors. Assuming that monitoring parameters including operating temperature α1, pH value α2, and operating flow rate α3 all affect the corrosion rate, and that temperature has the most significant impact on the corrosion rate, then operating temperature is determined as the key influencing parameter. The relationship between operating temperature and the influencing factors is fitted as f(α1), where α10 represents the design temperature. The failure probability influence coefficient k is then expressed as:

[0065]

[0066] Furthermore, in one or more exemplary embodiments of the present invention, when a high-temperature sulfur / naphthenic acid corrosion model is used, the real-time corrosion rate is calculated based on the operating temperature, oil sulfur content, acid value, and gaseous H2S content, and the failure probability influence coefficient k is obtained by combining it with the design corrosion rate; when a spheroidization model is used, the real-time service life is calculated based on the operating temperature, and the failure probability influence coefficient k is obtained by combining it with the design service life; when a graphitization model is used, the real-time service life is calculated based on the operating temperature, and the failure probability influence coefficient k is obtained by combining it with the design service life; and when an ammonium hydrosulfide corrosion model is used, the real-time corrosion rate is calculated based on the cyanide concentration, and the failure probability influence coefficient k is obtained by combining it with the design corrosion rate.

[0067] Furthermore, in one or more exemplary embodiments of the present invention, obtaining the equipment health evaluation level according to the health evaluation model includes: calculating the cross-correlation function, cohesion function, and spectral distance function respectively on the real-time raw vibration signal data of the measuring points of the moving equipment, and obtaining the calculation results of each function; comparing the calculation results of each function with the health evaluation matrix to determine the evaluation result of each function; and combining a preset two-out-of-three voting model or conservative principle to obtain the health evaluation level of each measuring point of the moving equipment, thereby obtaining the health evaluation level of the moving equipment.

[0068] Furthermore, in one or more exemplary embodiments of the present invention, the dynamic risk assessment model includes a corrosion model and a dynamic risk assessment matrix.

[0069] Furthermore, in one or more exemplary embodiments of the present invention, obtaining the dynamic risk level of a device according to the dynamic risk assessment model includes: obtaining the dynamic risk level of the device based on the real-time health and importance level of the device and the dynamic risk assessment matrix.

[0070] Furthermore, in one or more exemplary embodiments of the present invention, the data-driven real-time dynamic risk assessment of equipment is based on static RBI and RCM assessments. The static RBI assessment database consists of corrosion-sensitive characteristic parameter design data, static RBI assessment failure probability level discrimination criteria, and static RBI failure consequence level composition, while the static RCM assessment database consists of vibration monitoring raw waveform data and static RCM assessment failure probability, failure consequence, and risk level, also known as the equipment real-time dynamic risk assessment knowledge base.

[0071] Furthermore, in one or more exemplary embodiments of the present invention, obtaining the dynamic risk level of equipment according to the dynamic risk assessment model includes: if one of the high-temperature sulfur / naphthenic acid corrosion model, spheroidization corrosion model, graphitization corrosion model, and ammonium hydrosulfide corrosion model is used to perform model calculations on the real-time corrosion sensitivity characteristic parameter data of the measuring points of the static equipment, the real-time failure probability level is obtained, and combined with the failure consequence level in the evaluation system, the dynamic risk level of the measuring point under the current model is obtained; if the wet hydrogen sulfide corrosion model is used, the dynamic risk level of the measuring point under the current model is obtained based on the real-time pH value of the measuring point and the hydrogen sulfide content in the water, according to the wet hydrogen sulfide corrosion model matrix; and the dynamic risk level of the static equipment is the highest dynamic risk level among all measuring points.

[0072] Furthermore, in one or more exemplary embodiments of the present invention, when using the high-temperature sulfur / naphthenic acid corrosion model, the real-time corrosion rate is calculated based on the operating temperature, oil sulfur content, acid value, and gaseous H2S content. Combined with the design corrosion rate, a failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, a dynamic total damage factor is calculated. Based on the range of failure probability levels, the real-time failure probability level is obtained. When using the spheroidization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, a failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, a dynamic total damage factor is calculated. Based on the failure probability level range, a real-time failure probability level is obtained. The range of values ​​for the failure probability level is used to obtain the real-time failure probability level. When using the graphitization model, the real-time service life is calculated based on the operating temperature. Combined with the design service life, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. Based on the range of values ​​for the failure probability level, the real-time failure probability level is obtained. Similarly, when using the ammonium hydrosulfide corrosion model, the real-time corrosion rate is calculated based on the cyanide concentration. Combined with the design corrosion rate, the failure probability influence coefficient k is obtained. Based on the failure probability influence coefficient k and the total damage factor, the dynamic total damage factor is calculated. Based on the range of values ​​for the failure probability level, the real-time failure probability level is obtained.

[0073] For example, the damage factor is determined based on the analyzed damage mechanism (uniform or local thinning, cracking, creep, etc.), which is related to the material, operating conditions, service status, and inspection techniques for quantifying damage.

[0074] S140 makes maintenance decisions based on the equipment health assessment level and the equipment dynamic risk assessment level.

[0075] Further, in one or more exemplary embodiments of the present invention, step S140 includes:

[0076] If at least one of the equipment health assessment level and the equipment dynamic risk assessment level corresponds to a shutdown operation, a maintenance alarm will be triggered; and

[0077] Otherwise, return to step S110.

[0078] The intelligent maintenance decision-making method, system, electronic equipment and storage medium of the present invention are described in more detail below by way of specific embodiments. It should be understood that the embodiments are merely exemplary and the present invention is not limited thereto.

[0079] Example 1

[0080] This embodiment uses the intelligent maintenance decision-making method for refining and chemical equipment of the present invention to make maintenance decisions for a certain refining and chemical equipment.

[0081] From February 23 to April 9, the refining equipment operated normally, with all monitoring points showing normal data and no early fault warnings. On April 10, an early fault warning appeared at one of the monitoring points. Based on the health assessment model and the dynamic risk assessment model, the equipment's health assessment level was determined to be "recommended shutdown," and its dynamic risk assessment level was determined to be "high risk" (corresponding to shutdown). In this embodiment, monitoring continued until April 12 after the maintenance alarm, during which time the maintenance alarm persisted. Subsequently, a shutdown inspection was conducted, and the equipment problem was promptly identified.

[0082] In this embodiment, based on historical data, the health evaluation level of this device at this measuring point was mostly in "healthy state" before April 10, with occasional "sub-healthy state". From April 10 to April 12, the health evaluation level was always in "suggested shutdown".

[0083] Example 2

[0084] Combination Figure 2As shown, in this embodiment, the intelligent maintenance decision-making system for refining and chemical equipment according to the present invention includes: a data acquisition unit 10, which is used to acquire real-time monitoring data of each measuring point of the refining and chemical equipment; an early fault warning unit 20, which is used to input the acquired real-time monitoring data into an early warning model to determine whether the equipment has an early fault warning; a model evaluation unit 30, which is used to obtain the equipment health evaluation level and the equipment dynamic risk evaluation level according to the judgment result of the early fault warning unit, based on the health evaluation model and the dynamic risk evaluation model; and a maintenance decision-making unit 40, which is used to make maintenance decisions based on the equipment health evaluation level and the equipment dynamic risk evaluation level.

[0085] Example 3

[0086] This embodiment provides a non-transitory (non-volatile) computer storage medium that stores computer-executable instructions that can execute the methods in any of the above method embodiments and achieve the same technical effect.

[0087] Example 4

[0088] This embodiment provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, cause the computer to perform the methods described above and achieve the same technical effects.

[0089] Example 5

[0090] Figure 3 This is a schematic diagram of the hardware structure of the electronic device for implementing the intelligent maintenance decision-making method for refining and chemical equipment according to this embodiment. The device includes one or more processors 610 and a memory 620. Taking one processor 610 as an example, the device may also include an input device 630 and an output device 640.

[0091] The processor 610, memory 620, input device 630, and output device 640 can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0092] The memory 620, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 610 executes various functional applications and data processing of the electronic device by running the non-transitory software programs, instructions, and modules stored in the memory 620, thereby implementing the processing method of the above-described method embodiments.

[0093] The memory 620 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; the data storage area may store data, etc. Furthermore, the memory 620 may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 620 may optionally include memory remotely located relative to the processor 610, and these remote memories may be connected to the processing device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0094] Input device 630 can receive input digital or character information and generate signal input. Output device 640 may include display devices such as a display screen.

[0095] One or more modules are stored in memory 620 and, when executed by one or more processors 610, execute:

[0096] S110 acquires real-time monitoring data from various measuring points of the refining and chemical equipment;

[0097] S120 inputs the acquired real-time monitoring data into the early warning model to determine whether the device has an early fault warning. If yes, proceed to step S130; otherwise, return to step S110.

[0098] S130 obtains the equipment health assessment level and the equipment dynamic risk assessment level based on the health assessment model and the dynamic risk assessment model, respectively; and

[0099] S140 makes maintenance decisions based on the equipment health assessment level and the equipment dynamic risk assessment level.

[0100] The above-described product can execute the methods provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the methods. Technical details not described in detail in this embodiment can be found in the methods provided in other embodiments of the present invention.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general-purpose hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0103] The foregoing description of specific exemplary embodiments of the present invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. Any simple modifications, equivalent changes, and alterations made to the foregoing exemplary embodiments should fall within the scope of protection of the present invention.

Claims

1. A method for intelligent maintenance decision of refining and chemical equipment, characterized in that, The method comprises the following steps: S110 obtaining real-time monitoring data of each measuring point of the refining equipment; S120 inputting the obtained real-time monitoring data into an early warning model to determine whether an early failure warning of the equipment occurs, if yes, entering step S130, if not, returning to step S110; S130 obtaining an equipment health degree evaluation grade and an equipment dynamic risk evaluation grade respectively according to a health degree evaluation model and a dynamic risk evaluation model; The health degree evaluation model comprises a cross-correlation function, a cohesion function, a spectral distance function and a corrosion model; the corrosion model comprises one or more of a high-temperature sulfur / naphthenic acid corrosion model, a spheroidization model, a graphitization model and an ammonium bisulfide corrosion model; obtaining the equipment health degree evaluation grade according to the health degree evaluation model comprises: calculating the real-time corrosion characteristic parameter data of the measuring point of the static equipment by using the corrosion model to obtain a failure possibility influence coefficient k, and obtaining the health degree evaluation grade of the current corrosion model according to a health degree function H=1 / k and the value range of H of each health degree grade, wherein the failure possibility influence coefficient k represents a ratio of a key influence parameter causing a change in the failure possibility of the equipment under different corrosion models to a design value; calculating the real-time original vibration signal data of the measuring point of the dynamic equipment by using the cross-correlation function, the cohesion function and the spectral distance function respectively to obtain calculation results of the functions; comparing the calculation results of the functions with a health degree evaluation matrix to determine evaluation results of the functions; and obtaining the health degree evaluation grade of each measuring point of the dynamic equipment and further obtaining the health degree evaluation grade of the dynamic equipment by combining a pre-set three-out-of-two voting model or a conservative principle; The dynamic risk evaluation model comprises a corrosion model and a dynamic risk evaluation matrix; obtaining the equipment dynamic risk grade according to the dynamic risk evaluation model comprises: obtaining a dynamic risk grade of the dynamic equipment according to the real-time health degree and the importance degree grade of the dynamic equipment according to the dynamic risk evaluation matrix; and obtaining the equipment dynamic risk grade according to the dynamic risk evaluation model comprises: if one of the high-temperature sulfur / naphthenic acid corrosion model, the spheroidization corrosion model, the graphitization corrosion model and the ammonium bisulfide corrosion model is used to calculate the real-time corrosion sensitive characteristic parameter data of the measuring point of the static equipment, a real-time failure possibility grade is obtained, and a dynamic risk grade of the measuring point under the current model is obtained by combining a failure consequence grade in the evaluation system; if the wet hydrogen sulfide corrosion model is used, the dynamic risk grade of the measuring point under the current model is obtained according to the real-time PH value and the hydrogen sulfide content in water according to the wet hydrogen sulfide corrosion model matrix; and the dynamic risk grade of the static equipment is the highest dynamic risk grade among the measuring points; S140 making a maintenance decision according to the equipment health degree evaluation grade and the equipment dynamic risk evaluation grade.

2. The refinery intelligent maintenance decision-making method of claim 1, wherein, Step S140 comprises: if at least one corresponding operation in the equipment health degree evaluation grade and the equipment dynamic risk evaluation grade is shutdown, triggering a maintenance alarm; and otherwise, returning to step S110. 3.The refinery intelligent maintenance decision-making method according to claim 1, characterized in that, The real-time monitoring data of the refining equipment comprises corrosion sensitive characteristic parameters and process operation parameters of the static equipment, vibration monitoring original waveform data of the dynamic equipment, and static RBI and static RCM evaluation results.

4. The refinery intelligent maintenance decision making method of claim 1, wherein, The corrosion sensitive characteristic parameters of the static equipment include operating temperature, sulfur content of the distillate oil, acid value of the distillate oil, material, cyanogen concentration, PH value and hydrogen sulfide content in water.

5. The refinery intelligent maintenance decision-making method according to claim 1, characterized in that, The early warning models include a WPD+DKPCA intelligent early warning model, an XJH-1 model and an XJH-2 model.

6. The refinery intelligent maintenance decision-making method according to claim 5, characterized in that, The rule for determining whether the early failure warning of the equipment occurs is that: The model calculation is performed on the real-time monitoring data of each measuring point to obtain the prediction result of the current early warning model of each measuring point; If the early warning record occurs continuously for N times under any early warning model, it is determined that the early failure warning of the measuring point occurs; and If the early failure warning of any measuring point occurs, it is determined that the early failure warning of the equipment occurs.

7. The refinery intelligent maintenance decision making method of claim 5, wherein, The WPD+DKPCA intelligent early warning model calculation on the real-time monitoring data of the measuring point includes: The model training is performed on the reference signal data to obtain the model characteristic value of the equipment; The model calculation is performed on the real-time original vibration signal data of the measuring point through wavelet packet decomposition and dynamic kernel principal component analysis to obtain the real-time characteristic values T2 and SPE; and The real-time characteristic values are compared with the model characteristic values, and when the real-time characteristic values T2 and SPE exceed the preset model characteristic values, one early warning record is counted. 8.The refinery intelligent maintenance decision-making method according to claim 5, characterized in that, The XJH-1 model calculation on the real-time monitoring data of the measuring point includes: The model training is performed on the original vibration signal data meeting the conditions to extract the early failure sensitive characteristic values and input the SVDD after the improved l1 trend filtering processing to obtain the radius and center of the hypersphere; and The model calculation is performed on the real-time original vibration signal data to extract the early failure sensitive characteristic values and input the trained SVDD model after the improved l1 trend filtering processing, and when the real-time model characteristic value exceeds the preset model training value, one early warning record is counted. 9.The refinery intelligent maintenance decision-making method of claim 5, wherein, The XJH-2 model calculation on the real-time monitoring data of the measuring point includes: The model training is performed through the extraction of the early failure sensitive characteristic values and the improved l1 trend filtering processing to obtain the alarm threshold line by using the self-learning method; and The model calculation is performed on the real-time original vibration signal data to extract the characteristic values of the early failure sensitive characteristic values and the improved l1 trend filtering processing, and when the real-time model characteristic value exceeds the alarm threshold line, one early warning record is counted.

10. The intelligent maintenance decision-making method for the refining equipment according to claim 1, characterized in that when the high-temperature sulfur / naphthenic acid corrosion model is used, the real-time corrosion rate is calculated according to the operating temperature, the sulfur content of the oil product, the acid value, the gas-phase H2S content, and the failure possibility influence coefficient k is obtained in combination with the design corrosion rate; when the spheroidization model is used, the real-time service life is calculated according to the operating temperature, and the failure possibility influence coefficient k is obtained in combination with the design service life; when the graphitization model is used, the real-time service life is calculated according to the operating temperature, and failure possibility influence coefficient k is obtained in combination with the design service life; and when the ammonium bisulfide corrosion model is used, the real-time corrosion rate is calculated according to the cyanogen concentration, and the failure possibility influence coefficient k is obtained in combination with the design corrosion rate.

11. The intelligent maintenance decision-making method for the refining equipment according to claim 1, characterized in ​ ​ When the high-temperature sulfur / naphthenic acid corrosion model is used, the real-time corrosion rate is calculated according to the operating temperature, the sulfur content of the oil product, the acid value and the gas-phase H2S content, the failure possibility influence coefficient k is obtained in combination with the design corrosion rate, the dynamic total damage factor is calculated according to the failure possibility influence coefficient k and the total damage factor, and the real-time failure possibility level is obtained according to the value range of the failure possibility level; When the spheroidization model is used, the real-time service life is calculated according to the operating temperature, the failure possibility influence coefficient k is obtained in combination with the design service life, the dynamic total damage factor is calculated according to the failure possibility influence coefficient k and the total damage factor, and the real-time failure possibility level is obtained according to the value range of the failure possibility level; When the graphitization model is used, the real-time service life is calculated according to the operating temperature, the failure possibility influence coefficient k is obtained in combination with the design service life, the dynamic total damage factor is calculated according to the failure possibility influence coefficient k and the total damage factor, and the real-time failure possibility level is obtained according to the value range of the failure possibility level; and When the ammonium sulfide corrosion model is used, the real-time corrosion rate is calculated according to the cyanide concentration, the failure possibility influence coefficient k is obtained in combination with the design corrosion rate, the dynamic total damage factor is calculated according to the failure possibility influence coefficient k and the total damage factor, and the real-time failure possibility level is obtained according to the value range of the failure possibility level.

12. A refinery intelligent maintenance decision system characterized by, The method comprises: a data acquisition unit configured to acquire real-time monitoring data of each measuring point of the refining equipment; an early fault warning unit configured to input the acquired real-time monitoring data into an early warning model to determine whether an early fault warning occurs in the equipment; a model evaluation unit configured to obtain an equipment health degree evaluation level and an equipment dynamic risk evaluation level according to the health degree evaluation model and the dynamic risk evaluation model based on the determination result of the early fault warning unit; and a maintenance decision unit configured to make a maintenance decision according to the equipment health degree evaluation level and the equipment dynamic risk evaluation level.

13. An electronic device, comprising: comprise: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the at least one processor to perform the intelligent maintenance decision method for refining equipment according to any one of claims 1-11.

14. A non-transitory computer-readable storage medium, comprising: The non-transitory computer-readable storage medium stores computer-executable instructions for causing the computer to perform the intelligent maintenance decision method for refining equipment according to any one of claims 1-11.

Images