Equipment Maintenance Management System

Abstract

Provided is an equipment maintenance management system that realizes effective maintenance of equipment, enhances the timeliness of material requirements, and further strongly guarantees the safe and stable operation of equipment. 【Solution means】The equipment maintenance management system includes a data acquisition module for acquiring monitoring records of the operating state of the target equipment, a state analysis module for generating a failure alarm report based on the monitoring records of the operating state of the equipment, an equipment maintenance module for combining the failure alarm records with the equipment inspection and maintenance ledger of the target equipment to generate equipment maintenance records, and a spare parts management module for dynamically adjusting the material purchase plan based on the equipment maintenance records. The failure alarm report generated by performing a state inspection and analysis of the target equipment is combined with the inspection and maintenance ledger of the equipment, and it is ensured that the target maintenance personnel are performing precise maintenance work on the target equipment, generating equipment maintenance records, and dynamically adjusting the material purchase plan.

Description

Technical Field

[0001] The present invention relates to the technical field of power plant equipment repair management, and particularly to an equipment maintenance management system.

Background Art

[0002] In recent years, against the backdrop of the rapid development of the national economy, the power demand has been continuously increasing, and with the rapid expansion of the scale of power plants, power plant production interruption accidents caused by the "chain reaction" caused by defects and failures of power plant equipment have occurred frequently, greatly affecting the orderly development of society. Therefore, accurate management and maintenance of power plant equipment have become one of the important research items. However, the currently known equipment maintenance methods cannot accurately inspect external defects and operating failures of equipment, and cannot deeply analyze defects and accurately locate the defect locations.

[0003] Therefore, the present invention provides an equipment maintenance management system, which effectively associates real-time monitoring of equipment status, in-depth analysis of defect failures, accurate maintenance, and spare parts management, and strongly guarantees the safe and stable operation of equipment.

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present invention combines a fault warning report generated by performing state detection and analysis of target equipment with an equipment inspection and maintenance ledger, implements that the target maintenance personnel perform precise maintenance work on the target equipment, generates equipment maintenance records, and dynamically adjusts the material purchase plan, thereby realizing effective maintenance of equipment, improving the timeliness of material demand, and further providing an equipment maintenance management system that strongly guarantees the safe and stable operation of equipment.

Means for Solving the Problems

[0005] The present invention provides an equipment maintenance management system, including the following.

[0006] A data acquisition module for monitoring the operating status of a target device in real time and generating a monitoring record of the device operating status. A status analysis module for recognizing defects and faults of a target device based on the monitoring record of the device operating status, and generating a fault alarm record of the target device when there are defects and faults in the target device. An equipment maintenance module for combining the fault alarm record with the equipment inspection and maintenance ledger of the target device, implementing that a target maintenance personnel is performing maintenance work on the target device, and generating an equipment maintenance record. A spare parts management module for determining predicted consumption data of spare parts based on the equipment maintenance record and dynamically adjusting the material purchase plan.

[0007] Preferably, the data acquisition module includes the following.

[0008] Using a preset monitoring device attached to a preset monitoring point to monitor the operating status of a target device in real time and obtain first operating status data. A data collection unit that obtains the basic equipment information of the target device from the equipment information database, collates and organizes it with the obtained first operating status data to obtain a monitoring record of the equipment operating status.

[0009] Preferably, the basic equipment information includes equipment name, equipment affiliated department, equipment code, equipment type, shipping date, technical parameters, responsible person, installation location, and spare parts information, etc.

[0010] Preferably, the status analysis module includes the following.

[0011] A data processing unit for performing data preprocessing on the first status data to obtain target image data and target operating status data. A fault inspection unit for performing defect inspection on the current target device using an external inspection block, and performing fault inspection, analysis and prediction on the current target device using an internal inspection block. Perform an external defect inspection on the current target equipment based on the target image data, generate a defect analysis table when a defect exists, and transfer it to the report generation unit, an external inspection block Perform a failure inspection on the current target equipment based on the target operating state data, generate failure analysis data when the presence of a failure is detected, and transfer it to the report generation unit Otherwise, perform a failure prediction on the current target equipment, generate failure prediction data, and transfer it to the report generation unit, an internal inspection block Receive the defect analysis table, failure analysis data, and failure prediction data, organize them together with the basic equipment information of the current target equipment, generate a failure warning report, and transfer it to the equipment repair module, a report generation unit

[0012] Preferably, the external inspection block includes the following

[0013] Construct a training dataset based on a preset amount of past equipment defect reports extracted from the equipment maintenance database, train a neural network to obtain a defect identification model Sharpen the normal image data in the target image data using the CGAN algorithm to obtain a target image Input the target image into the defect identification model to perform defect identification, and determine the first defect site and the first defect type of the current target equipment Predict the first defect risk value of the first defect site based on past maintenance data extracted from the equipment maintenance database Among them, the calculation formula for the first defect risk value is as follows JPEG0007711337000002.jpg9143 Here, Q i is the i-th first defect site X ij is the j-th set risk evaluation amount of the i-th first defect site. Here, the set risk evaluation amount is the ratio of the past maintenance times, the patrol inspection frequency, the average maintenance time, and the risk degree of the defect type δij is the influence weight of the j-th set risk assessment value of the i-th first defect site on the risk assessment of the first defect site Construct a first defect table based on the first defect site, the first defect type, and the first defect risk value Perform defect diagnosis on the thermal image data in the target image data to obtain a second defect site and a defect site identification temperature Based on the difference between the defect site identification temperature and the corresponding set temperature threshold, obtain the second defect risk value of the second defect site Construct a second defect table based on the second defect site, the defect site identification temperature, and the second defect risk value For the defect sites that exist simultaneously in the first defect table and the second defect table, calculate the total value of the corresponding first defect value and the second defect value, and output it as a defect threat value The first defect risk value that exists only in the first defect table and does not exist in the first defect site of the second defect table is used as the defect threat value of the first defect site The second defect risk value that exists only in the second defect table and does not exist in the second defect site of the first defect table is used as the defect threat value of the second defect site Judge the threat levels of all the first defect sites and the second defect sites based on the defect threat values Arrange all the currently detected defect sites in descending order of the defect threat value, combine the obtained defect type, defect site identification temperature, and threat level to create a defect analysis table, and transfer it to the report generation unit

[0014] Preferably, the internal inspection block includes the following

[0015] Perform feature extraction on the target operating state data using a spectral analysis method to obtain important state feature values Use the important state feature values as the input side of a state inspection model constructed based on the random forest algorithm to output the state inspection result of the current target equipment As the state inspection result of the current target equipment, when a fault exists, based on the state threshold range, select the reference operating state quantity whose state value does not belong to the corresponding state threshold range from the target operating state data A failure identification sub-block for combining and analyzing the reference operating state quantity and the setting state - failure related ranking table, obtaining failure related results, and outputting them as failure analysis data. As the current state inspection result for the target equipment, if there is no failure, input the target operating state data into a pre - constructed state prediction model to obtain the prediction result of the state quantity at the next time. If there exists a target operating state quantity whose prediction result of the state quantity does not belong to the corresponding state threshold range, perform prediction correction on the state quantity for that target operating state quantity and update the prediction result of the state quantity. Among them, the calculation formula for the predicted value of the state quantity after correction is as follows: JPEG0007711337000003.jpg24145 Here, JPEG0007711337000004.jpg7142 is the predicted value of the state quantity after correction for the k - th target operating state quantity. E k is the predicted value of the original state quantity at the next time for the k - th target operating state quantity. E kmax is the upper limit of the state threshold for the k - th operating state quantity. E kmin is the lower limit of the state threshold for the k - th operating state quantity. f is a pre - set correction coefficient. A state prediction sub - block for inputting the prediction result of the state quantity into the state prediction model and further obtaining the prediction result of the state quantity at the next time. Using the failure identification sub - block, perform failure diagnosis on the prediction results of the state quantity within the preset time period repeatedly obtained for the state prediction sub - block, and obtain the failure prediction results at each preset time. Based on the failure prediction results, obtain the preset times when failures exist and the corresponding failure items, and construct a failure prediction table in time series. Using the prediction results of the state quantity within the preset time period, construct a state quantity prediction change curve for each target operating state quantity. Analyze the state quantity prediction change curve to obtain the prediction change trend of each target operating state quantity. A change analysis sub - block for outputting the obtained failure prediction table, the state quantity prediction change curves of each target operating state quantity, and the prediction change trends as failure prediction data.

[0016] Preferably, the equipment maintenance module includes the following.

[0017] A personnel screening unit for obtaining selectable maintenance personnel using a list acquisition block, screening the selectable maintenance personnel obtained using a screening block, and obtaining target maintenance personnel. A list acquisition block for obtaining the equipment name and the department to which the equipment belongs of the equipment to be repaired based on the received failure alarm report, and screening corresponding selectable maintenance personnel from the equipment maintenance information database. A screening block for obtaining the work status and defect failure handling records of all selectable maintenance personnel, performing combined analysis with the received defect threat level, failure-related results, and failure prediction table, and screening and obtaining target maintenance personnel. A maintenance unit for obtaining the equipment inspection and maintenance ledger of the target equipment, transferring it to the target maintenance personnel together with the failure alarm report, performing accurate maintenance on the target equipment, and generating equipment maintenance records.

[0018] Preferably, the spare parts management module includes the following.

[0019] Obtaining the consumption data of inspection spare parts and maintenance spare parts of the target equipment based on the equipment maintenance records. A predicted consumption unit for associating the existing spare parts inventory information, unarrived spare parts information, the consumption data of inspection spare parts, and the consumption data of maintenance spare parts to generate predicted consumption data of spare parts. An adjustment unit for dynamically adjusting the procurement plan of materials using the predicted consumption data of the spare parts.

Advantages of the Invention

[0020] Compared with the prior art, the present application has the following advantageous effects.

[0021] By combining the fault warning report generated through the state detection and analysis of the target equipment with the equipment inspection and maintenance ledger, ensuring that the target maintenance personnel perform precise maintenance work on the target equipment, generating equipment maintenance records, and dynamically adjusting the material purchase plan, effective maintenance of the equipment is realized, while enhancing the timeliness of material demand, and further strongly guaranteeing the safe and stable operation of the equipment.

[0022] Other features and advantages of the present invention are described in the specification hereinafter, some of which will become apparent from the description of the specification, or can be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the specification and drawings.

[0023] Hereinafter, the technical solution of the present invention will be described in more detail in conjunction with the drawings and embodiments.

Brief Description of the Drawings

[0024] The drawings are for further understanding of the present invention, form part of the specification, are used in conjunction with the embodiments of the present invention to explain the present invention, and do not limit the present invention.

[0025]

Figure 1

Modes for Carrying Out the Invention

[0026] Hereinafter, preferred embodiments of the present invention will be described in conjunction with the drawings. It should be understood that the preferred embodiments described herein are used to explain and interpret the present invention and do not limit the present invention.

[0027] Embodiments of the present invention provide an equipment maintenance management system, which, as shown in FIG. 1, includes the following.

[0028] A data acquisition module for monitoring the operating state of the target equipment in real time and generating a monitoring record of the equipment operating state. A state analysis module for recognizing defects and faults of target equipment based on the monitoring records of the operation status of the equipment, and generating a fault alarm record of the target equipment when there are defects and faults in the target equipment. An equipment maintenance module for combining the fault alarm record with the equipment inspection and maintenance ledger of the target equipment, implementing that the target maintenance personnel are performing maintenance work on the target equipment, and generating an equipment maintenance record. A spare parts management module for determining the predicted consumption data of spare parts based on the equipment maintenance record and dynamically adjusting the material purchase plan.

[0029] In this embodiment, the target equipment refers to equipment such as generators, fans, and water supply pumps in power plant equipment that monitors the status in real time to ensure production efficiency and production safety. The monitoring record of the equipment operation status is a summary of the basic equipment information of the target equipment and the obtained first operation status data. Among them, the basic equipment information includes equipment name, equipment affiliated department, equipment code, equipment type, shipping date, technical parameters, responsible person, installation location, and spare parts information, etc.

[0030] In this embodiment, the first operation status data is composed of video data detected in real time using a preset inspection device attached to a preset monitoring point and the real-time operation status data of the equipment. Among them, the preset monitoring points are set in advance, and the preset inspection devices include data collection devices, infrared thermal imaging cameras, and ordinary cameras.

[0031] In this embodiment, the fault alarm report is a report generated by summarizing and organizing the defect analysis table, fault analysis data, fault prediction data, and the current basic equipment information of the target equipment. Among them, the defect analysis table is composed of defect location, defect type, defect location identification temperature, and threat level.

[0032] In this embodiment, the fault analysis data refers to the fault-related results, that is, the fault items related to the reference operating state quantity obtained based on the set state - fault-related ranking table. Among them, the set state - fault-related ranking table is established using the degree of correlation between the state quantity and the fault determined by the gray-scale correlation analysis method in advance. The reference operating state quantity means that when it is determined using the state inspection model that there is a fault in the current target equipment, the state numerical value in the target operating state data of the target equipment does not belong to the operating state quantity within the corresponding state threshold range. The fault items are composed of the fault location and the fault type.

[0033] In this embodiment, the fault prediction data includes a fault prediction table, the state quantity prediction change curve and the prediction change trend of each target operating state quantity. Among them, the fault prediction table is composed of arranging the predicted fault occurrence time and the fault name in chronological order, and the prediction change trend includes rising and falling.

[0034] In this embodiment, the state inspection model is a model for inspecting the operating state of the target equipment obtained by using the frequency domain characteristic values of the past operating state data of the equipment as training data and combining random forest algorithm training.

[0035] In this embodiment, the equipment inspection and maintenance ledger is composed of equipment inspection records and technical materials. Among them, the equipment inspection records include the inspection start date and end date, the situation before repair, the inspection content, the situation after inspection, and the maintenance personnel, etc. The technical materials include inspection file packs, inspection process cards, etc. The target maintenance personnel refer to the most suitable maintenance personnel for dealing with the defects and faults of the current target equipment.

[0036] In this embodiment, the equipment maintenance record is composed of equipment basic information and maintenance information. Among them, the equipment basic information includes the equipment name, the department to which the equipment belongs, the equipment code, the equipment type, the shipment date, technical parameters, the responsible person, the installation location, and spare part information, etc. The maintenance information includes the maintenance time, the nature of maintenance, the maintenance content, the maintenance personnel and the maintenance time, the inspection spare part consumption data and the maintenance spare part consumption data. Among them, the nature of maintenance is divided into two types: repair and daily maintenance.

[0037] In this embodiment, the predicted consumption data of spare parts refers to the predicted consumption quantity, predicted consumption spare part name, and predicted consumption spare part specification of spare parts obtained by analyzing the existing spare part inventory information, unarrived spare part information, consumption data of inspection spare parts, and maintenance spare part consumption data in association. The material purchase plan is a demand plan extracted by the material demand department according to the actual situation, and details such as the name, specification, quantity, and price of the required materials are described in detail.

[0038] The above technical solution combines the fault alarm report generated by performing a condition inspection analysis on the target equipment and the equipment inspection and maintenance ledger, so as to implement that the target maintenance personnel are performing precise maintenance work on the target equipment, generate equipment maintenance records, and perform dynamic adjustment on the material purchase plan, thereby realizing effective maintenance of the equipment, improving the timeliness of material demand, and further strongly guaranteeing the safe and stable operation of the equipment.

[0039] The present invention provides an equipment maintenance management system, and the data acquisition module includes the following.

[0040] Using a preset monitoring device attached to a preset monitoring point to monitor the operating status of the target equipment in real time and obtain first operating status data. A data collection unit that obtains the basic equipment information of the target equipment from the equipment information database and collates it with the obtained first operating status data to obtain a monitoring record of the equipment operating status.

[0041] In this embodiment, the target equipment refers to equipment such as generators, fans, and water supply pumps in power plant equipment that needs to be monitored in real time to ensure production efficiency and production safety.

[0042] In this embodiment, the first operating state data is composed of video data obtained by real-time inspection using a preset inspection device attached to a preset monitoring point and real-time operating state data of the equipment. Among them, the preset monitoring point is set in advance, and the preset inspection device includes a data collection device, an infrared thermal imaging camera, and an ordinary camera. The equipment information database is composed of basic equipment information.

[0043] In this embodiment, the monitoring record of the equipment operating state is a record obtained by collating the basic equipment information of the target equipment and the first operating state data obtained. Among them, the basic equipment information includes equipment name, equipment department, equipment code, equipment type, shipping date, technical parameters, responsible person, installation location, spare parts information, etc.

[0044] The above technical solution has the advantageous effect of obtaining the monitoring record of the equipment operating state by collating the first operating state data obtained by real-time monitoring of the operating state of the target equipment using a preset inspection device attached to the preset monitoring point with the basic equipment information of the target equipment, and providing data support for subsequent equipment maintenance.

[0045] The embodiment of the present invention provides an equipment maintenance management system, and the state analysis module includes the following.

[0046] A data processing unit for performing data preprocessing on the first state data to obtain target image data and target operating state data. A fault inspection unit for performing a defect inspection on the current target equipment using an external inspection block and performing a fault inspection, analysis, and prediction on the current target equipment using an internal inspection block. An external inspection block for performing an external defect inspection on the current target equipment based on the target image data and generating a defect analysis table when a defect exists and transferring it to a report generation unit. Perform a fault inspection on the current target equipment based on the target operating state data, and when the presence of a fault is detected, generate fault analysis data and transfer it to the report generation unit. Otherwise, an internal inspection block for performing a fault prediction on the current target equipment, generating fault prediction data, and transferring it to the report generation unit. A report generation unit that receives the defect analysis table, fault analysis data, and fault prediction data, collates and organizes them together with the basic equipment information of the current target equipment, generates a fault warning report, and transfers it to the equipment repair module.

[0047] In this embodiment, the first operating state data is composed of video data detected in real time using a preset inspection device attached to a preset monitoring point and the real-time operating state data of the equipment. Among them, the preset monitoring point is set in advance, and the preset inspection device includes a data collection device, an infrared thermal imaging camera, and an ordinary camera.

[0048] In this embodiment, the target image data is a target image obtained by extracting images from video data collected using an ordinary camera and an infrared thermal imaging camera and performing noise removal processing, and includes an ordinary image and a thermal image. The target operating state data is data obtained by supplementing missing values in the real-time operating state data of the target equipment collected in real time using a data collection device and performing normalization processing.

[0049] In this embodiment, the defect analysis table is composed of a defect site, a defect type, a defect site identification temperature, and a threat level. The fault analysis data refers to the results related to the fault, that is, the fault items related to the reference operating state quantity obtained based on the set state - fault related ranking table. Among them, the set state - fault related ranking table is established using the degree of correlation between the state quantity and the fault determined in advance by the gray-scale degree of correlation analysis method. The reference operating state quantity means that when it is determined that there is a fault in the current target equipment using the state inspection model, the state numerical value in the target operating state data of the target equipment does not belong to the operating state quantity within the corresponding state threshold range.

[0050] In this embodiment, the fault prediction data is composed of a fault prediction table and the predicted change trends of each target operating state quantity. Among them, the fault prediction table is composed of arranging the predicted fault occurrence time and the fault name in time series, and the predicted change trends include rising and falling. The basic equipment information includes equipment name, equipment affiliated department, equipment code, equipment type, shipping date, technical parameters, responsible person, installation location, and spare parts information, etc. The fault alarm report is generated by organizing and collating the defect analysis table, fault analysis data, and fault prediction data with the basic equipment information of the current target equipment.

[0051] The above technical solution uses the video data and operating state data obtained based on the target equipment monitored in real time after data preprocessing to perform defect inspection, fault identification and prediction, generate a fault alarm report, realize a deep analysis of the defect fault, provide important information for equipment maintenance, and thereby has the advantageous effect of improving the maintenance efficiency and accuracy of the equipment.

[0052] The embodiment of the present invention provides an equipment maintenance management system, and the external inspection block includes the following.

[0053] Based on the preset amount of past equipment defect reports extracted from the equipment maintenance database, a training dataset is constructed to train a neural network to obtain a defect identification model. The ordinary image data in the target image data is sharpened by the CGAN algorithm to obtain a target image. The target image is input into the defect identification model for defect identification to determine the first defect site and the first defect type of the current target equipment. Based on the past maintenance data extracted from the equipment maintenance database, the first defect risk value of the first defect site is predicted. Among them, the calculation formula of the first defect risk value is as follows. JPEG0007711337000005.jpg9143 Here, Q i is the i-th first defect site X ij is the j-th set risk evaluation value of the i-th first defect site, where the set risk evaluation value is the ratio of the past maintenance times, the patrol inspection frequency, the average maintenance time, and the risk degree of the defect type δ ij is the influence weight of the j-th set risk evaluation value of the i-th first defect site on the risk degree evaluation of the first defect site Construct a first defect table based on the first defect site, the first defect type, and the first defect risk value, Perform defect diagnosis on the thermal image data in the target image data to obtain a second defect site and a defect site identification temperature, Based on the difference between the defect site identification temperature and the corresponding set temperature threshold, obtain the second defect risk value of the second defect site, Construct a second defect table based on the second defect site, the defect site identification temperature, and the second defect risk value, For the defect sites that exist simultaneously in the first defect table and the second defect table, calculate the total value of the corresponding first defect value and the second defect value, and output it as a defect threat value, Regard the first defect risk value that exists only in the first defect table and does not exist in the first defect site of the second defect table as the defect threat value of the first defect site, Regard the second defect risk value that exists only in the second defect table and does not exist in the second defect site of the first defect table as the defect threat value of the second defect site, Judge the threat levels of all the first defect sites and the second defect sites based on the defect threat values, Arrange all the currently detected defect sites in descending order of the defect threat values, combine the obtained defect type, defect site identification temperature, and threat level to create a defect analysis table, and transfer it to the report generation unit.

[0054] In this embodiment, the defect identification model is a model obtained by training a neural network using a training dataset constructed based on past equipment defect reports that extract preset amounts from the equipment maintenance database, and is used for defect identification of the collected images of the equipment, Among them, the establishment procedure of the training data is as follows.

[0055] 1. Extract past equipment defect image data from the obtained past equipment defect reports, perform preliminary defect category marking, 2. Select the past equipment defect images after category marking, obtain image data with only one type of defect and clear defect characteristics as the first image, 3. Perform expansion processing such as inversion, local shielding, and cutting on the first image to obtain a training dataset.

[0056] In this embodiment, the target image is an image obtained by sharpening a normal image with the CGAN algorithm. The first defect site is the defect site obtained by inputting the sharpened normal image into the defect identification model, such as an insulator or a box door. The first defect type includes damage, deformation, corrosion, etc. The past maintenance data is data extracted from the equipment maintenance database, such as the name of the past maintained equipment, the maintained parts, the maintenance frequency, and the maintenance duration.

[0057] In this embodiment, the first defect risk value is used to characterize the processing urgency of the current first defect site. The first defect table is composed of the first defect site, the first defect type, and the first defect risk value.

[0058] In this embodiment, the thermal image data is an image obtained by extracting images from the video data collected by an infrared thermal imaging camera and performing noise removal processing. The second defect site is the part obtained by decomposing the thermal image, evaluating it based on the HSV color model, and then using the threshold segmentation technique for segmentation. The defect site identification temperature is the temperature at the second defect site, and the set temperature threshold is set in advance.

[0059] In this embodiment, the second defect table is composed of a second defect site, a defect site identification temperature, and a second defect risk value. Among them, the second defect risk value is obtained by calculating the difference between the corresponding defect site identification temperature of the second defect site and the set temperature threshold, and is used to characterize the processing urgency of the current second defect site. The defect threat value is used to characterize the degree of threat of the defect site to the safe operation of the current target equipment. The threat level is divided into four levels: general threat, slightly large threat, large threat, and major threat. The defect analysis table is composed of a defect site, a defect type, a defect site identification temperature, and a threat level.

[0060] The above technical solution can generate a defect analysis table by detecting external defects of equipment for ordinary images and thermal images respectively based on image processing technology, thereby providing accurate recognition of external defects of equipment and highly reliable defect analysis data for future maintenance, which is beneficial to improving the maintenance efficiency and reliability of equipment.

[0061] Embodiments of the present invention provide an equipment maintenance management system, and the internal inspection block includes the following.

[0062] Perform feature extraction on the target operating state data using a spectrum analysis method to obtain important state feature values. Take the important state feature values as the input side of a state inspection model constructed based on the random forest algorithm, and output the state inspection result of the current target equipment. As the state inspection result of the current target equipment, if a fault exists, based on the state threshold range, select a reference operating state quantity whose state value does not belong to the corresponding state threshold range from the target operating state data. A fault identification sub-block for combining and analyzing the reference operating state quantity and the set state-fault related ranking table to obtain a fault related result and output it as fault analysis data. As the state inspection result of the current target equipment, if no fault exists, input the target operating state data into a pre-constructed state prediction model to obtain a prediction result of the state quantity at the next moment. If there exists an object operation state quantity for which the predicted result of the state quantity does not belong to the corresponding state threshold range, perform a prediction correction on the state quantity of the object operation state quantity, update the predicted result of the state quantity, Among them, the calculation formula for the predicted value of the state quantity after correction is as follows: JPEG0007711337000006.jpg24145 Here, JPEG0007711337000007.jpg7142 is the predicted value of the state quantity after correction for the k-th object operation state quantity E k is the predicted value of the original state quantity at the next time for the k-th object operation state quantity E kmax is the upper limit of the state threshold of the k-th operation state quantity E kmin is the lower limit of the state threshold of the k-th operation state quantity f is a preset correction coefficient A state prediction sub-block for inputting the predicted result of the state quantity into the state prediction model and obtaining the predicted result of the state quantity at the next time, Using the fault identification sub-block, perform fault diagnosis on the predicted results of the state quantities within the preset time period repeatedly obtained for the state prediction sub-block, and obtain the fault prediction results at each preset time, Based on the fault prediction results, obtain the preset times when faults exist and the corresponding fault items, and construct a fault prediction table in time series, Using the predicted results of the state quantities within the preset time period, construct a state quantity prediction change curve for each object operation state quantity, Analyze the state quantity prediction change curve to obtain the predicted change trend of each object operation state quantity, A change analysis sub-block for outputting the obtained fault prediction table, the state quantity prediction change curves of each object operation state quantity, and the predicted change trends as fault prediction data.

[0063] In this embodiment, the target operating state data is data obtained by supplementing missing values in the real-time operating state data of the target facility collected in real time using a data collection device and performing normalization processing. The spectrum analysis method is a method for converting the time-domain signal of the target operating state data into the frequency domain. The important state characteristic values are those obtained by extracting state frequency domain characteristics from the target operating state data using the spectrum analysis method.

[0064] In this embodiment, the state inspection model is a model for detecting the operating state of the target facility obtained by combining the frequency domain characteristic values of the past operating state data of the facility as training data and performing random forest algorithm training. The state inspection result has two results: the presence or absence of a fault.

[0065] In this embodiment, the state threshold range is a preset numerical range of the operating state quantity. Among them, the operating state quantity includes voltage, current, power, etc. The reference operating state quantity refers to the operating state quantity whose state value does not belong outside the corresponding state threshold range.

[0066] In this embodiment, the set state-fault correlation ranking table is created using the correlation degree between the state quantity and the fault determined in advance by the gray scale correlation degree analysis method. The fault-related result refers to the fault item related to the reference operating state quantity determined based on the set state-fault correlation ranking table.

[0067] In this embodiment, the state prediction model is a model used for predicting the state quantity at a future time obtained by training an LSTM neural model with the past important state data of a preset quantity extracted from the equipment operation database as input values.

[0068] In this embodiment, the prediction result of the state quantity is the predicted state quantity at the next time obtained by inputting the current operating state data into the state prediction model and outputting it. The fault prediction result is the fault item obtained by performing fault identification based on the prediction result of the state quantity. The preset time is the time when a fault occurs within the preset time period.

[0069] In this embodiment, the fault prediction table is a table created in chronological order with the preset times when faults exist and the corresponding fault items. Among them, the fault items are composed of the fault name and the fault type. The state quantity prediction change curve is created using the prediction results of the state quantity within the preset time period, and the prediction change trend includes two trends: rising and falling. The fault prediction data includes the fault prediction table, the state quantity prediction change curves of the state quantities of each target operating state, and the prediction change trend.

[0070] The above technical solution combines the spectrum analysis method and the random forest algorithm to perform fault detection on the operating state of the current target equipment. When there is no fault in the current target equipment, a state prediction model is constructed to analyze the change trend of the operating state of the target equipment, so as to realize effective monitoring and control of the equipment state, and further strongly guarantee the safe and stable operation of the equipment.

[0071] The embodiment of the present invention provides an equipment maintenance management system, and the equipment maintenance module includes the following.

[0072] A personnel selection unit for obtaining selectable maintenance personnel using a list acquisition block, and screening the obtained selectable maintenance personnel using a screening block to obtain target maintenance personnel. A list acquisition block for obtaining the equipment name and the department to which the equipment belongs of the equipment to be repaired based on the received fault alarm report, and screening the corresponding selectable maintenance personnel from the equipment maintenance information database. A screening block for obtaining the working status and defect fault handling records of all selectable maintenance personnel, combining and analyzing the received defect threat level, fault-related results, and fault prediction table, and screening and obtaining target maintenance personnel. It includes a maintenance unit for obtaining the equipment inspection and maintenance ledger of the target equipment, transferring it to the target maintenance personnel together with the fault alarm report, performing accurate maintenance on the target equipment, and generating equipment maintenance records.

[0073] In this embodiment, the failure alarm report is generated by collating and organizing the defect analysis table, failure analysis data, failure prediction data, and the basic equipment information of the current target equipment. The equipment to be repaired refers to the target equipment that generates a failure alarm record and sends it to the equipment repair module. The equipment maintenance database is composed of equipment maintenance data and includes equipment maintenance parameter information such as the name of the maintained equipment, the department to which the equipment belongs, the maintenance location, the maintenance frequency, the maintenance staff number, and the defect failure handling record.

[0074] In this embodiment, the selectable maintenance staff are selected from the equipment maintenance database according to the name of the equipment to be repaired and the department to which the equipment belongs. As working states, there are two types: not working and working.

[0075] In this embodiment, the equipment inspection and maintenance ledger is composed of equipment inspection records and technical documents. Among them, the equipment inspection records include the inspection start date and end date, the situation before repair, the inspection content, the situation after repair, and the maintenance staff, etc. The technical documents include inspection file packs, inspection work procedure cards, etc. The target maintenance staff refers to the most suitable maintenance staff for handling the defects and failures of the current target equipment.

[0076] By combining and analyzing the above technical solutions, that is, by combining the failure alarm report and the equipment inspection and maintenance ledger, the maintenance staff are selected to accurately inspect and repair the defects and failures existing in the current target equipment, realizing effective maintenance of the equipment, and having the advantageous effect of helping the equipment to operate safely and stably.

[0077] The embodiment of the present invention provides an equipment maintenance management system, and the spare parts management module includes the following.

[0078] Based on the equipment maintenance records, obtain the consumption data of inspection spare parts and the consumption data of maintenance spare parts of the target equipment. An estimated consumption unit that associates the existing spare parts inventory information, the information of unarrived spare parts, the consumption data of inspection spare parts, and the consumption data of maintenance spare parts to generate the predicted consumption data of spare parts. An adjustment unit that dynamically adjusts the procurement plan of materials by using the predicted consumption data of the spare parts.

[0079] In this embodiment, the equipment maintenance record is composed of the basic information and maintenance information of the equipment. Among them, the equipment basic information includes the equipment name, the department to which the equipment belongs, the equipment code, the equipment type, the shipping date, the technical parameters, the person in charge, the installation location, and the spare parts information, etc. The maintenance information includes the maintenance time, the nature of maintenance, the content of maintenance, the maintenance personnel, the maintenance time, the inspection spare parts consumption data, and the maintenance spare parts consumption data. Among them, the nature of maintenance has two types: repair and daily maintenance.

[0080] In this embodiment, the inspection spare parts consumption data is the number, specification, and name of the spare parts used for the repair of the equipment. The maintenance spare parts consumption data is the number, specification, and name of the spare parts used for the maintenance of the equipment. The unarrived spare parts information is the number, specification, and name of the unstocked spare parts purchased.

[0081] In this embodiment, the predicted consumption data of the spare parts is the predicted spare parts consumption quantity, the predicted consumption spare parts name, and the specification of the predicted consumption spare parts obtained by associating and analyzing the existing spare parts inventory information, the unarrived spare parts information, the inspection spare parts consumption data, and the maintenance spare parts consumption data. The material purchase plan is the demand plan extracted by the material demand department according to the actual situation, and details such as the name, specification, quantity, and price of the required materials are described in detail.

[0082] The above technical solution has the advantageous effect that by determining the predicted consumption data of the spare parts based on the equipment maintenance record and dynamically adjusting the material purchase plan, the timeliness of the material demand can be guaranteed, the flexibility in the management of the equipment spare parts can be enhanced, and the equipment maintenance can be strongly guaranteed.

[0083] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these changes and modifications of the present invention are within the claims of the present invention and its equivalent technical scope, the present invention also intends to include these changes and modifications.

Claims

1. An equipment maintenance management system, comprising: A data acquisition module for monitoring the operating status of the target equipment in real time and generating a monitoring record of the equipment operating status; A status analysis module for recognizing defects and failures of the target equipment based on the monitoring record of the equipment operating status, and generating a failure warning record of the target equipment when there are defects and failures in the target equipment; An equipment maintenance module for combining the failure warning record with the equipment inspection and maintenance ledger of the target equipment, implementing that the target maintenance personnel perform maintenance work on the target equipment, and generating an equipment maintenance record; A spare parts management module for determining the predicted consumption data of spare parts based on the equipment maintenance record and dynamically adjusting the material purchase plan; Among them, the status analysis module includes the following: A data processing unit for performing data preprocessing on the first status data to obtain target image data and target operating status data; A failure inspection unit for performing defect inspection on the current target equipment using an external inspection block, and performing failure inspection, analysis and prediction on the current target equipment using an internal inspection block; An external inspection block for performing external defect inspection on the current target equipment based on the target image data, generating a defect analysis table when there are defects, and transferring it to the report generation unit; Performing a failure inspection on the current target equipment based on the target operating status data, generating failure analysis data when the presence of a failure is detected, and transferring it to the report generation unit; Otherwise, performing a failure prediction on the current target equipment, generating failure prediction data, and transferring it to the report generation unit using an internal inspection block; A report generation unit for receiving the defect analysis table, failure analysis data and failure prediction data, collating and organizing them with the basic equipment information of the current target equipment, generating a failure warning report, and transferring it to the equipment repair module; Among them, the external inspection block includes the following: Constructing a training data set based on a preset amount of past equipment defect reports extracted from the equipment maintenance database, training a neural network to obtain a defect identification model; Sharpening the normal image data in the target image data using the CGAN algorithm to obtain a target image; Input the target image into the defect identification model to perform defect identification, determine the first defect site and the first defect type of the current target equipment, Predict the first defect risk value of the first defect site based on the past maintenance data extracted from the equipment maintenance database, Among them, the calculation formula of the first defect risk value is as follows, Here, Q ｉ is the i-th first defect site X ij is the j-th set risk assessment value of the i-th first defect site, where the set risk assessment value is the ratio of the number of past maintenance times, the frequency of patrol inspection, the average maintenance time, and the risk level of the defect type δij is the influence weight of the j-th set risk evaluation amount of the i-th first defect site on the risk degree evaluation of the first defect site Construct a first defect table based on the first defect site, the first defect type, and the first defect risk value, Perform defect diagnosis on the thermal image data in the target image data, and obtain the second defect site and the defect site identification temperature, Obtain the second defect risk value of the second defect site based on the difference between the defect site identification temperature and the corresponding set temperature threshold, Construct a second defect table based on the second defect site, the defect site identification temperature, and the second defect risk value, For the defect sites that exist simultaneously in the first defect table and the second defect table, calculate the total value of the corresponding first defect value and the second defect value, and output it as the defect threat value, Regard the first defect risk value that exists only in the first defect table and does not exist in the first defect site of the second defect table as the defect threat value of the first defect site, Regard the second defect risk value that exists only in the second defect table and does not exist in the second defect site of the first defect table as the defect threat value of the second defect site, Judge the threat levels of all the first defect sites and the second defect sites based on the defect threat value, Sort all the currently detected defect sites in descending order of the defect threat value, combine the obtained defect type, defect site identification temperature, and threat level to create a defect analysis table, and transfer it to the report generation unit, Among them, the internal inspection block includes the following, Perform feature extraction on the target operating state data using the spectrum analysis method to obtain important state feature values, As the input side of the state inspection model constructed based on the random forest algorithm for the important state feature values, output the state inspection result of the current target equipment, As the state inspection result of the current target equipment, when a fault exists, based on the state threshold range, select the reference operating state quantity whose state value does not belong to the corresponding state threshold range from the target operating state data, A fault identification sub-block for combining and analyzing the reference operating state quantity and the set state-fault related ranking table to obtain a fault related result and output it as fault analysis data, As the current status inspection result of the target equipment, if there is no fault, input the target operating status data into a pre-constructed status prediction model to obtain the prediction result of the status quantity at the next time. If there exists a target operating quantity where the prediction result of the status quantity does not belong to the corresponding status threshold range, perform a prediction correction on the status quantity for that target operating status quantity and update the prediction result of the status quantity. Among them, the calculation formula for the predicted value of the corrected status quantity is as follows. Here, is the predicted value of the corrected status quantity of the k-th target operating status quantity E k is the predicted value of the original state quantity at the next time of the k-th target operating state quantity E kmax is the upper limit of the state threshold of the k-th operating state quantity E kmin is the lower limit of the state threshold of the k-th operating state quantity f is a preset correction coefficient A status prediction sub-block for inputting the prediction result of the status quantity into the status prediction model and further obtaining the prediction result of the status quantity at the next time, Using a fault identification sub-block, perform fault diagnosis on the prediction results of the status quantities within a preset time period repeatedly obtained for the status prediction sub-block, and obtain the fault prediction results at each preset time. Based on the fault prediction results, obtain the preset time when a fault exists and the corresponding fault items, and construct a fault prediction table in time series. Using the prediction results of the status quantities within a preset time period, construct a status quantity prediction change curve for each target operating status quantity. Analyze the status quantity prediction change curve to obtain the prediction change trend of each target operating status quantity. A change analysis sub-block for outputting the obtained fault prediction table, the status quantity prediction change curves of each target operating status quantity, and the prediction change trends as fault prediction data, characterized by the equipment maintenance management system described above.

2. The data acquisition module includes the following. Using a preset monitoring device attached to a preset monitoring point, monitor the operating status of the target equipment in real time to obtain first operating status data, A data collection unit that obtains the basic equipment information of the target equipment from the equipment information database and organizes and collates it with the obtained first operating status data to obtain a monitoring record of the equipment operating status, characterized by the equipment maintenance management system described in Claim 1.

3. The basic equipment information includes equipment name, equipment belonging department, equipment code, equipment type, shipping date, technical parameters, responsible person, installation location, and spare parts information, etc., characterized by the equipment maintenance management system described in Claim 2.

4. The equipment maintenance module includes the following. A personnel screening unit for obtaining selectable maintenance personnel using a list acquisition block, screening the selectable maintenance personnel obtained using a screening block, and obtaining target maintenance personnel. A list acquisition block for obtaining the equipment name and equipment department of the equipment to be repaired based on the received failure alarm report, and screening corresponding selectable maintenance personnel from the equipment maintenance information database. A screening block for obtaining the work status and defect failure handling records of all selectable maintenance personnel, combining and analyzing them with the received defect threat level, failure-related results, and failure prediction table, and screening and obtaining target maintenance personnel. The equipment maintenance management system according to claim 1, characterized by including a maintenance unit for obtaining the equipment inspection and maintenance ledger of the target equipment, transferring it to the target maintenance personnel together with the failure alarm report, performing accurate maintenance on the target equipment, and generating equipment maintenance records.

5. The spare parts management module includes the following. Obtaining the consumption data of inspection spare parts and maintenance spare parts of the target equipment based on the equipment maintenance records. A predicted consumption unit for associating the existing spare parts inventory information, unarrived spare parts information with the consumption data of inspection spare parts and maintenance spare parts to generate predicted consumption data of spare parts. The equipment maintenance management system according to claim 1, characterized by including an adjustment unit for dynamically adjusting the procurement plan of materials using the predicted consumption data of the spare parts.

Images