A predictive maintenance management system for the entire lifecycle of chemical equipment
By using a predictive maintenance management system for the entire lifecycle of chemical equipment, the health status of equipment is dynamically assessed, solving the problem of difficulty in identifying early potential problems in existing technologies, and enabling precise maintenance decisions and resource optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DALIAN GAOJIA CHEM
- Filing Date
- 2026-04-24
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies make it difficult to identify early potential problems from subtle correlations and anomalies in the operating parameters of chemical equipment, leading to a disconnect between maintenance arrangements and equipment status, which may result in resource waste or sudden failures.
A predictive maintenance management system for the entire lifecycle of chemical equipment is adopted. Through data standardization, coupled analysis, health index calculation, condition assessment and maintenance decision-making modules, the system dynamically assesses the health status of equipment and provides maintenance recommendations.
It enables dynamic and accurate assessment of equipment health status, reduces resource waste and failure risk, and improves the adaptability of the maintenance system.
Smart Images
Figure CN122089291A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of chemical equipment maintenance and management technology, and specifically relates to a predictive maintenance management system for the entire life cycle of chemical equipment. Background Technology
[0002] Chemical equipment operates in complex environments, often under high temperatures and pressures. Its health is crucial to production safety and efficiency, making the completeness of its maintenance management system a primary concern. Current industry practices often involve periodic maintenance plans based on cumulative equipment operating time. These plans also collect operational data such as temperature, flow rate, and pressure, and combine this data with historical failure cases to help assess the equipment's current condition and plan subsequent maintenance schedules.
[0003] However, during the operation of chemical equipment, load adjustments, component aging, and other factors can cause subtle changes in the correlation. Current maintenance methods are unable to identify early potential problems from these subtle changes in parameter correlation, leading to a disconnect between maintenance arrangements and the real-time status of the equipment. This can result in either excessive maintenance and wasted resources, or failure to detect potential problems and sudden malfunctions. Summary of the Invention
[0004] This application provides a predictive maintenance management system for the entire life cycle of chemical equipment, which effectively solves the problem that existing maintenance methods are unable to identify early hidden dangers from subtle changes in parameter correlations. It realizes dynamic and accurate assessment of equipment health status, provides a comprehensive basis for maintenance decisions, and reduces resource waste and failure risks.
[0005] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application provides a predictive maintenance management system for the entire lifecycle of chemical equipment, comprising: Data standardization module: acquires at least three operating parameters of the target chemical equipment within a first time period, performs standardization processing on the operating parameters, and obtains standardized operating parameter data.
[0006] Coupling analysis module: Based on the standardized operating parameter data, analyze the dynamic coupling relationship between multiple operating parameters, and calculate the real-time correlation coefficient between parameter pairs to form a real-time correlation matrix.
[0007] Health indicator calculation module: Calculates initial health indicators based on the standardized operating parameter data; compares and maps the real-time correlation matrix with the pre-stored historical benchmark correlation coefficient to obtain coupled health indicators.
[0008] Status assessment module: The initial health index and the coupled health index are fused and calculated to obtain a comprehensive health index, and the status is classified according to the comprehensive health index to obtain the health status level.
[0009] Maintenance decision module: Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance suggestions are obtained.
[0010] Feedback optimization module: After executing the dynamic maintenance suggestion, update the historical benchmark correlation coefficient and the maintenance urgency correction function based on the maintenance feedback.
[0011] Furthermore, the standardization of the operating parameters includes at least the following: using a minimum-maximum standardization method to normalize the real-time monitoring data to obtain standardized operating parameter data.
[0012] Furthermore, the Pearson correlation coefficients between at least three operating parameters in pairs during the first time period are calculated to form a real-time correlation matrix.
[0013] Furthermore, initial health indicators are calculated based on the standardized operating parameter data, including: By invoking preset single-item scoring rules, each parameter in the standardized operating parameter data is mapped to a preset health score range using linear interpolation, and the initial health score of each operating parameter is calculated.
[0014] The initial health index is obtained by calculating the arithmetic mean of the initial health scores of all operating parameters.
[0015] Furthermore, the real-time correlation matrix is compared and mapped with the pre-stored historical benchmark correlation coefficients to obtain coupled health indicators, including: The historical baseline correlation coefficients between each parameter pair are obtained by reading the historical normal operation data of the device from the historical database. The absolute value of each coefficient in the real-time correlation matrix is compared with the absolute value of the corresponding historical baseline correlation coefficient, and the relative change percentage is calculated to obtain the correlation coefficient deviation of each parameter pair.
[0016] The total dynamic coupling deviation is obtained by summing the deviations of the correlation coefficients for all parameter pairs.
[0017] Based on historical equipment failure cases, a piecewise linear mapping function is pre-calibrated. The total dynamic coupling deviation is input into the piecewise linear mapping function, and the coupling health index is calculated by looking up a table.
[0018] Furthermore, the initial health index and the coupled health index are fused and calculated to obtain a comprehensive health index, specifically: The initial health index and the coupled health index are used as input variables and input into a fuzzy logic rule base established based on the analysis of the device's historical full life cycle data. The comprehensive health index is obtained by calculation through fuzzy inference and centroid method defuzzification.
[0019] Furthermore, based on the comprehensive health indicators, a health status level is obtained by classifying the status, specifically as follows: The values of the comprehensive health indicators are matched and queried with multiple pre-divided continuous value intervals to determine the current health status level of the device.
[0020] Furthermore, based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance recommendations are obtained, including: Query the historical maintenance records of the target equipment from the equipment management information system and extract the cumulative runtime since the most recent preventive maintenance.
[0021] The baseline maintenance urgency value corresponding to the health status level is obtained by querying a preset level and baseline maintenance urgency comparison table.
[0022] The dynamic maintenance urgency value is obtained by calculating the baseline maintenance urgency value and the cumulative running time by inputting a linearly increasing maintenance urgency correction function.
[0023] The dynamic maintenance urgency value is compared with preset emergency maintenance thresholds and planned maintenance thresholds, and dynamic maintenance recommendations containing instructions for immediate maintenance, planned maintenance, or continued observation are generated based on the comparison results.
[0024] Furthermore, the preset level and benchmark maintenance urgency comparison table is established by statistically analyzing the historical operation and fault data of the equipment, and mapping different health status levels to a quantitative benchmark maintenance urgency value.
[0025] Furthermore, after executing the dynamic maintenance recommendation, the historical baseline correlation coefficient and the maintenance urgency correction function are updated based on maintenance feedback, including: After executing the dynamic maintenance recommendations, the maintenance operation records in the maintenance work order system are obtained, and the post-maintenance operation parameter data within the set observation period after the equipment is put back into operation are obtained.
[0026] Update the historical maintenance records in the equipment management information system according to the maintenance operation records; recalculate the correlation coefficients between parameters in the short term based on the post-maintenance operating parameter data, in order to update the historical baseline correlation coefficients.
[0027] Based on the maintenance type and the recovery status of health indicators after maintenance in the maintenance operation record, the growth rate parameter of the maintenance urgency correction function is adjusted using heuristic rules.
[0028] Secondly, this application provides a predictive maintenance management method for the entire life cycle of chemical equipment, including: At least three operating parameters of the target chemical equipment within a first time period are obtained, and the operating parameters are standardized to obtain standardized operating parameter data.
[0029] Based on the standardized operating parameter data, the dynamic coupling relationship between multiple operating parameters is analyzed, and the real-time correlation coefficient between parameter pairs is calculated to form a real-time correlation matrix.
[0030] Initial health indicators are calculated based on the standardized operating parameter data; the real-time correlation matrix is compared and mapped with the pre-stored historical benchmark correlation coefficients to obtain coupled health indicators.
[0031] The initial health index and the coupled health index are fused together to obtain a comprehensive health index, and the health status level is obtained by classifying the status based on the comprehensive health index.
[0032] Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance recommendations are obtained.
[0033] After implementing the dynamic maintenance recommendations, the historical baseline correlation coefficient and the maintenance urgency correction function are updated based on maintenance feedback.
[0034] Thirdly, this application provides a predictive maintenance management device for the entire life cycle of chemical equipment, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the predictive maintenance management method for the entire life cycle of chemical equipment.
[0035] Fourthly, this application provides a readable storage medium storing computer program instructions, which, when read and executed by a processor, perform the steps of a predictive maintenance management method for the entire lifecycle of chemical equipment.
[0036] Fifthly, this application provides a computer program product, including a computer program or instructions, wherein when the computer program or instructions are executed by a processor, steps are taken to implement a predictive maintenance management method for the entire life cycle of chemical equipment.
[0037] The beneficial effects of this application are: This application effectively solves the problem in existing technologies that maintenance methods cannot identify early hidden dangers from subtle changes in parameter correlations. By acquiring and standardizing equipment operating parameters, analyzing the dynamic coupling relationship of parameters to form a correlation matrix, integrating and calculating comprehensive health indicators and classifying them, and combining runtime and correction functions to provide maintenance suggestions and feedback optimization, it achieves dynamic and accurate assessment of equipment health status, provides a comprehensive basis for maintenance decisions, reduces resource waste and failure risks, and improves the adaptability of the maintenance system.
[0038] Other features and advantages of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures pointed out in the description and the accompanying drawings. Attached Figure Description
[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0040] Figure 1 A schematic diagram of a predictive maintenance management system for the entire life cycle of chemical equipment according to this application is shown. Detailed Implementation
[0041] To address the problems raised in the background technology, this paper proposes a method to obtain and standardize equipment operating parameters, analyze the dynamic coupling relationship of parameters to form a correlation matrix, integrate and calculate comprehensive health indicators and classify them, and combine runtime and correction functions to provide maintenance suggestions and feedback optimization, thereby achieving dynamic and accurate assessment of equipment health status.
[0042] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0043] In some embodiments, such as Figure 1 As shown, this application provides a predictive maintenance management system for the entire lifecycle of chemical equipment, including: Data standardization module: Acquire at least three operating parameters of the target chemical equipment within the first time period, standardize the operating parameters, and obtain standardized operating parameter data.
[0044] Coupling Analysis Module: Based on standardized operating parameter data, it analyzes the dynamic coupling relationship between multiple operating parameters and calculates the real-time correlation coefficient between parameter pairs to form a real-time correlation matrix.
[0045] Health indicator calculation module: Initial health indicators are calculated based on standardized operating parameter data; the real-time correlation matrix is compared and mapped with the pre-stored historical benchmark correlation coefficient to obtain coupled health indicators.
[0046] Status assessment module: It integrates the initial health indicators and coupled health indicators to obtain a comprehensive health indicator, and classifies the status according to the comprehensive health indicator to obtain the health status level.
[0047] Maintenance decision module: Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance suggestions are obtained.
[0048] Feedback optimization module: After executing dynamic maintenance suggestions, the historical baseline correlation coefficient and maintenance urgency correction function are updated based on maintenance feedback.
[0049] In some embodiments, sensors deployed on the target chemical equipment collect real-time monitoring data of at least three operating parameters over a continuous time period, such as inlet pressure, outlet pressure, motor current, bearing temperature, and RMS vibration velocity. The real-time monitoring data forms an original sequence at a fixed sampling frequency.
[0050] The real-time monitoring data is preprocessed to obtain standardized operating parameter data. For example, the minimum-maximum standardization method is used to normalize the real-time monitoring data.
[0051] In some embodiments, analyzing the dynamic coupling relationship between multiple operating parameters specifically includes: Calculate the Pearson correlation coefficients between at least three operating parameters in the first time period.
[0052] For m operating parameters, calculate the correlation coefficients pairwise, and obtain the following results. There are several correlation coefficient values. These values are organized into an m×m symmetric matrix R to form the real-time correlation matrix. The element in the i-th row and j-th column of the matrix is... The first element represents the correlation coefficient between the i-th parameter and the j-th parameter. The elements on the main diagonal are the correlation coefficients between the parameter and itself, which are always 1.
[0053] In some embodiments, initial health indicators are calculated based on standardized operating parameter data, including: By invoking preset single-item scoring rules, each parameter in the standardized operating parameter data is mapped to a preset health score range using linear interpolation, and the initial health score of each operating parameter is calculated.
[0054] Specifically, the individual scoring rule can be as follows: a health score range is predefined for each operating parameter, for example, [0, 100], where 0 represents extremely unhealthy and 100 represents perfectly healthy. At the same time, a standard value range for this parameter under ideal operating conditions is defined, and linear interpolation is used to map the current standardized parameter value to the health score range.
[0055] Let the current value of a parameter after standardization be... The lower limit of its standardized value under healthy conditions is The upper limit is .
[0056] like exist If the body is internally healthy, then the body is considered healthy; initial health score. .
[0057] like or Then, based on the degree to which it deviates from the normal range, it is linearly mapped to a lower score in the range [0, 100]. For example, hour, ; hour, .
[0058] For all m operating parameters, calculate their individual health scores, and then calculate the arithmetic mean of these scores to obtain the initial health index. .
[0059] In some embodiments, the real-time correlation matrix is compared and mapped with pre-stored historical benchmark correlation coefficients to obtain coupled health indicators, including: The absolute value of each coefficient in the real-time correlation matrix is compared with the absolute value of the corresponding historical baseline correlation coefficient, and the percentage of relative change is calculated to obtain the correlation coefficient deviation of each parameter pair.
[0060] Retrieve the initial values of the historical baseline correlation coefficient matrix obtained from statistical calculations during the long-term stable operation phase of the device from the historical database. , Each element in This represents the typical correlation coefficient between parameters i and j under a healthy baseline state.
[0061] For each off-diagonal element in the real-time matrix R, calculate its relationship with the corresponding historical baseline. degree of deviation .
[0062] Deviation of all parameter pairs Summing these values yields the Total Dynamic Coupling Deviation (TDCD). The larger the TDCD, the more significant the deviation of the overall coupling mode between the current equipment operating parameters from the healthy baseline state. Based on historical equipment failure cases, a piecewise linear mapping function is pre-calibrated. The total dynamic coupling deviation is input into the piecewise linear mapping function, and the coupling health index is calculated by looking up a table. The lower the value, the more abnormal the cooperative relationship between parameters.
[0063] For example, by analyzing historical data, it was found that: when At that time, the equipment coupling relationship was normal, the health level was high, and the coupling health indicators were normal. Assign a score of 95-100; when At that time, the coupling relationship showed a slight abnormality, the health level was moderate, and the coupling health indicators were... Assign a score of 80-95, when At that time, the coupling relationship was significantly abnormal, the health level was low, and the coupling health indicators were abnormal. A score below 80 is assigned. This achieves the coupling of health indicators with TDCD. The conversion.
[0064] In some embodiments, the initial health indicators and coupled health indicators are fused together to obtain a comprehensive health indicator, specifically: Due to initial health indicators and coupling health indicators The health status is reflected from different perspectives, and the sensitivity of the device to both varies at different stages of its life cycle. Fuzzy logic can be used to fuse these two aspects.
[0065] Based on the device's historical full lifecycle data, a set of fuzzy rules is established by domain experts or through data mining methods to obtain a fuzzy logic rule library.
[0066] For example, rule 1: if the initial health indicators For high, and If the initial health index is high, then the overall health index is excellent; Rule 2: If the initial health index is high... For the middle, and If the initial health index is low, then the overall health index should be noted; Rule 3: If the initial health index is low... It is low, and If the overall health indicators are high, then a warning sign will be issued.
[0067] Initial health indicators and coupling health indicators The membership function is used to convert the membership to high, medium, and low levels of fuzzy linguistic variables. For example, If the membership degree of a high-order member is 0.7 and that of a medium-order member is 0.3, then... It is high.
[0068] Based on all fuzzy rules, calculate the degree of satisfaction of the premises of each rule, and assign the degree of satisfaction to the conclusion of the rule.
[0069] All activated rule conclusions are aggregated to form an output fuzzy set. Finally, the centroid method is used for defuzzification, calculating the abscissa value corresponding to the centroid of the region enclosed by the membership function curve and the abscissa of the fuzzy set. This transforms a fuzzy distribution back into a clear numerical value, yielding a comprehensive health index. The value ranges from 0 to 100.
[0070] In some embodiments, in S4, the health status level is obtained by classifying the status based on comprehensive health indicators, specifically as follows: Comprehensive health indicators The value is matched with multiple pre-divided continuous value intervals to determine the current health status level of the device.
[0071] For example, The health status level is healthy; The health status level is "Caution"; The health status level is warning. The health status level is dangerous.
[0072] In some embodiments, dynamic maintenance recommendations are obtained based on health status levels, combined with cumulative runtime extracted from historical databases, and by calculating maintenance urgency using a maintenance urgency correction function. These recommendations include: Query the historical maintenance records of the target equipment from the equipment management information system to extract the time of the most recent preventive maintenance completion. ,calculate The difference between the current time and the current time yields the cumulative runtime. .
[0073] Based on the health status level, a preset level-to-baseline maintenance urgency lookup table is consulted to obtain the corresponding baseline maintenance urgency value for that level. .
[0074] In some embodiments, a preset level and baseline maintenance urgency mapping table is established by statistically analyzing historical equipment operation and failure data, and mapping different health status levels to a quantified baseline maintenance urgency value.
[0075] Specifically, a long-term time-series data period of the equipment's history, including normal operation and known failure events, is selected, and a comprehensive health index is calculated for each historical moment. .
[0076] Every historical moment The numerical value is labeled as a historical status level, such as healthy, alert, etc.
[0077] For each time point marked as a specific historical status level, check whether a fault occurs within a subsequent preset warning time window, and record the severity level s of the fault, such as level 1-5, with 5 being the most severe.
[0078] For each historical state level L, calculate the failure frequency and average failure severity.
[0079] Normalize the fault occurrence frequency and average fault severity calculated for all levels to obtain the normalized fault occurrence frequency. and average failure severity Then, a weighted fusion is performed to obtain the basic risk score. , This can be achieved by analyzing the correlation between the frequency and severity of failures under different health states in historical data and actual maintenance decisions, and by using statistical methods such as linear regression to determine the weight values that best match historical decisions in the model output. and .
[0080] Basic risk score By linearly mapping to the target range, the baseline maintenance urgency value corresponding to that level is obtained. Baseline maintenance urgency values for all levels. This constitutes a comparison table of maintenance urgency levels and baselines.
[0081] Baseline maintenance urgency value and cumulative runtime The dynamic maintenance urgency value is obtained by calculating the linearly increasing maintenance urgency correction function. Where k represents a preset growth rate parameter, the initial value of which can be obtained by analyzing the historical operation and maintenance records of the equipment and calculating the cumulative operating time at different times. The conditional probability of a failure or the frequency of maintenance intervention is determined by fitting the slope of urgency as running time using linear regression, thus obtaining k.
[0082] Dynamically maintain urgency values With preset emergency maintenance threshold Planned maintenance threshold The system performs a comparison and generates dynamic maintenance recommendations based on the comparison results, including instructions for immediate maintenance, planned maintenance, or continued monitoring.
[0083] For example, if Then, suggestions for immediate maintenance will be generated; if If so, a maintenance plan suggestion will be generated; if If so, a suggestion to continue observation will be generated.
[0084] In some embodiments, after implementing dynamic maintenance recommendations, the historical baseline correlation coefficient and maintenance urgency correction function are updated based on maintenance feedback, including: After implementing dynamic maintenance recommendations, retrieve maintenance operation records from the maintenance work order system, and obtain post-maintenance operating parameter data within the set observation period after the equipment is put back into operation.
[0085] Update historical maintenance records in the equipment management information system based on maintenance operation records; recalculate the correlation coefficients between parameters in the short term based on post-maintenance operating parameter data, and generate a new correlation coefficient matrix. Read the current historical benchmark correlation coefficient matrix from the database. The updated historical baseline correlation coefficient matrix is obtained by applying the exponentially weighted moving average algorithm. And save it back to the database. ;in, This represents a smoothing factor used to control the update magnitude, such as 0.9.
[0086] S63. Based on the maintenance type and the recovery status of health indicators after maintenance in the maintenance operation record, adjust the growth rate parameter of the maintenance urgency correction function using heuristic rules.
[0087] Specifically, the text descriptions in the maintenance operation records are analyzed, and the maintenance type (C) is determined by keyword matching, such as replacement matching major repair, lubrication matching minor maintenance, etc.
[0088] Calculate the recovery rate of health indicators ;in, These represent health indicators before and after maintenance.
[0089] Apply heuristic rules to adjust the growth rate parameter of the maintenance urgency correction function, for example: Rule 1: If maintenance type C is a major repair, and Greater than or equal to a preset high-effect threshold ,like If 80% recovery rate is achieved, the maintenance is considered effective, and the equipment aging process is effectively reset or significantly slowed down. Therefore, the growth rate parameter is adjusted downwards, and the new growth rate parameter is [value missing]. ;in, This represents a discount factor less than 1, such as 0.8. This represents the old growth rate parameter.
[0090] Rule 2: If maintenance type C is minor maintenance, and η is less than or equal to a preset low-effect threshold. ,like Therefore, it is believed that this maintenance failed to effectively improve the underlying health condition, and the potential aging problems of the equipment may still be aggravated. Thus, the growth rate parameter is adjusted upwards, and the new growth rate parameter is... ;in, This represents a magnification factor greater than 1, such as 1.2.
[0091] Rule 3: If neither of the above two conditions is met, then the maintenance effect is considered to be in line with expectations or no special adjustment is required, and the growth rate parameter remains unchanged.
[0092] In some embodiments, this application provides a predictive maintenance management method for the entire lifecycle of chemical equipment, including: Obtain at least three operating parameters of the target chemical equipment within the first time period, and standardize the operating parameters to obtain standardized operating parameter data.
[0093] Based on standardized operating parameter data, the dynamic coupling relationship between multiple operating parameters is analyzed, and the real-time correlation coefficient between parameter pairs is calculated to form a real-time correlation matrix.
[0094] Initial health indicators are calculated based on standardized operating parameter data; coupled health indicators are obtained by comparing and mapping the real-time correlation matrix with the pre-stored historical benchmark correlation coefficients.
[0095] The initial health indicators and coupled health indicators are fused together to obtain a comprehensive health indicator, and the health status level is obtained by classifying the status based on the comprehensive health indicator.
[0096] Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance recommendations are obtained.
[0097] After implementing dynamic maintenance recommendations, the historical baseline correlation coefficient and maintenance urgency correction function are updated based on maintenance feedback.
[0098] In some embodiments, this application provides a predictive maintenance management device for the entire life cycle of chemical equipment, which includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the steps of the predictive maintenance management method for the entire life cycle of chemical equipment.
[0099] In some embodiments, this application provides a readable storage medium storing computer program instructions, which are read and executed by a processor to perform the steps of a predictive maintenance management method for the entire life cycle of chemical equipment.
[0100] In some embodiments, this application provides a computer program product, including a computer program or instructions, wherein when the computer program or instructions are executed by a processor, steps are taken to implement a predictive maintenance management method for the entire life cycle of chemical equipment.
[0101] It should be noted that, in this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.
[0102] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.
[0103] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A predictive maintenance management system for the entire lifecycle of chemical equipment, characterized in that, include: Data standardization module: acquires at least three operating parameters of the target chemical equipment within a first time period, standardizes the operating parameters, and obtains standardized operating parameter data; Coupling analysis module: Based on the standardized operating parameter data, analyze the dynamic coupling relationship between multiple operating parameters, and calculate the real-time correlation coefficient between parameter pairs to form a real-time correlation matrix; Health indicator calculation module: Calculates initial health indicators based on the standardized operating parameter data; The real-time correlation matrix is compared and mapped with the pre-stored historical benchmark correlation coefficient to obtain the coupled health index; Status assessment module: The initial health index and the coupled health index are fused and calculated to obtain a comprehensive health index, and the status is classified according to the comprehensive health index to obtain the health status level; Maintenance decision module: Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance suggestions are obtained; Feedback optimization module: After executing the dynamic maintenance suggestion, update the historical benchmark correlation coefficient and the maintenance urgency correction function based on the maintenance feedback.
2. The system according to claim 1, characterized in that, The standardization of the operating parameters includes at least the following: using a minimum-maximum standardization method to normalize the operating parameters and obtain standardized operating parameter data.
3. The system according to claim 1, characterized in that, Based on the standardized operating parameter data, the Pearson correlation coefficients between at least three operating parameters in the first time period are calculated to form a real-time correlation matrix.
4. The system according to claim 1, characterized in that, Initial health indicators are calculated based on the standardized operating parameter data, including: By invoking preset single-item scoring rules, each parameter in the standardized operating parameter data is mapped to a preset health score range using linear interpolation, and the initial health score of each operating parameter is calculated. The initial health index is obtained by calculating the arithmetic mean of the initial health scores of all operating parameters.
5. The system according to claim 1, characterized in that, The real-time correlation matrix is compared and mapped with the pre-stored historical benchmark correlation coefficients to obtain coupled health indicators, including: The historical baseline correlation coefficients between each parameter pair are obtained by reading the historical normal operation data of the equipment from the historical database. The absolute value of each coefficient in the real-time correlation matrix is compared with the absolute value of the corresponding historical baseline correlation coefficient, and the relative change percentage is calculated to obtain the correlation coefficient deviation of each parameter pair. The total dynamic coupling deviation is obtained by summing the deviations of the correlation coefficients for all parameter pairs. Based on historical equipment failure cases, a piecewise linear mapping function is pre-calibrated. The total dynamic coupling deviation is input into the piecewise linear mapping function, and the coupling health index is calculated by looking up a table.
6. The system according to claim 1, characterized in that, The initial health index and the coupled health index are fused and calculated to obtain a comprehensive health index, specifically: The initial health index and the coupled health index are used as input variables and input into a fuzzy logic rule base established based on the analysis of the device's historical full life cycle data. The comprehensive health index is obtained by calculation through fuzzy inference and centroid method defuzzification.
7. The system according to claim 1, characterized in that, Based on the comprehensive health indicators, a health status level is obtained by classifying the health status, specifically as follows: The values of the comprehensive health indicators are matched and queried with multiple pre-divided continuous value intervals to determine the current health status level of the device.
8. The system according to claim 1, characterized in that, Based on the health status level, combined with the cumulative runtime extracted from the historical database, and using the maintenance urgency correction function to calculate the maintenance urgency, dynamic maintenance recommendations are obtained, including: Query the historical maintenance records of the target equipment from the equipment management information system and extract the cumulative runtime since the most recent preventive maintenance; Based on the health status level, a preset level and baseline maintenance urgency lookup table is queried to obtain the baseline maintenance urgency value corresponding to that level, which is stored in the baseline maintenance urgency lookup table; The baseline maintenance urgency value and the cumulative running time are input into a linearly increasing maintenance urgency correction function to calculate the dynamic maintenance urgency value; The dynamic maintenance urgency value is compared with preset emergency maintenance thresholds and planned maintenance thresholds, and dynamic maintenance recommendations containing instructions for immediate maintenance, planned maintenance, or continued observation are generated based on the comparison results.
9. The system according to claim 8, characterized in that, The corresponding health status and baseline maintenance urgency table is established by statistically analyzing historical equipment operation and failure data, and mapping different health status levels to a quantified baseline maintenance urgency value.
10. The system according to claim 1, characterized in that, After executing the dynamic maintenance recommendation, the historical baseline correlation coefficient and the maintenance urgency correction function are updated based on maintenance feedback, including: After executing the dynamic maintenance recommendations, the maintenance operation records in the maintenance work order system are obtained, and the post-maintenance operation parameter data within the set observation period after the equipment is put back into operation are obtained. Update the historical maintenance records in the equipment management information system according to the maintenance operation records; recalculate the correlation coefficients between parameters in the short term based on the post-maintenance operating parameter data, in order to update the historical baseline correlation coefficients; Based on the maintenance type and the recovery status of health indicators after maintenance in the maintenance operation record, the growth rate parameter of the maintenance urgency correction function is adjusted using heuristic rules.