KR102344852B1 - Digital Twin-based Prediction and Diagnostic Device for Pump Bearing Systems - Google Patents

Abstract

The present invention relates to a digital twin-based predictive diagnosis device for a pump bearing system, and in particular, it is possible to track the cause of a defect by comprehensively analyzing various operating variables and monitoring variables related to the monitoring target facilities constituting the pump bearing system. , a digital twin-based predictive diagnosis of a pump bearing system that enables early detection of fault occurrence using various diagnostic models, and enables users to intuitively identify the cause and situation of a fault by using and visualizing a digital twin model. It's about the device.
The constituent means constituting the digital twin-based predictive diagnosis device of the pump bearing system according to the present invention includes a data collection module that collects monitoring/operation data related to facilities to be monitored of at least one pump bearing system, and the data collection module collected by the data collection module. A data processing module that stores and manages data and performs data communication with a linked server, and predicts failure of a pump bearing system through an analysis and simulation process that applies the predictive diagnosis basic information data input from the data processing module to at least one model And a prediction and diagnosis module that performs diagnosis and prediction of optimal operating conditions, and a digital twin model processing module that links and processes the predictive diagnosis result information data input from the prediction and diagnosis module to a digital twin 3D model and provides it through visualization It is characterized in that it is composed.

Description

Digital Twin-based Prediction and Diagnostic Device for Pump Bearing Systems

The present invention relates to a digital twin-based predictive diagnosis device of a pump bearing system, and in particular, it is possible to track the cause of a defect by comprehensively analyzing various operating variables and monitoring variables related to the monitoring target facilities constituting the pump bearing system. , a digital twin-based predictive diagnosis of a pump bearing system that enables early detection of fault occurrence using various diagnostic models, and uses and visualizes a digital twin model to enable users to intuitively identify the cause and situation of a fault It's about the device.

As shown in Figure 1, the large pump bearing system (1) has a bearing set (3) that mostly supports a rotating body, a lubricant supply device (5) for supplying lubricant to the bearing set, and a cooling water supply device ( 7) is included.

Lubricating oil supplied to the bearing set (3) is very important to minimize friction between the bearings (Journal and Thrust bearings) constituting the bearing set and the rotating body and to remove heat generated by this. play a role

If the lubrication supply is not smooth due to a problem such as a failure of the lubricant supply device 5 or clogging of the filter in the intermediate line, the temperature of the bearings rises and damage occurs, and the rotating shaft cannot function properly to support the rotating body. cause secondary damage.

The cooling water supply device 7 is a device that cools the lubricating oil supplied from the lubricating oil supply device 5 to the bearing set 3, and cools the temperature of the lubricating oil uniformly. The coolant supply problem is also one of the major causes of damage to the bearings by raising the temperature of the lubricating oil by failing to maintain the temperature.

As such, unlike the small pump, the large pump bearing system 1 has a bearing set 30, a lubricating oil supply unit (Hydraulic Power Unit) 5 and a cooling water supply unit (Cooling Unit), as shown in FIG. ) (7) and so on.

Since the pump bearing system 1 supports the axial and radial loads generated by the dynamic behavior of the rotating body while rotating components such as the shaft, coupling, and impeller, the pump It is directly related to the safety and drivability of Accordingly, the industry includes the monitoring of the pump bearing system as a very important factor among the operation status monitoring factors of large pumps.

However, since most monitoring technologies for these pump bearing systems simply judge whether or not the operation limit values set by the user have been exceeded, they only provide information on whether an abnormality has occurred and the root cause is the related auxiliary device (lubricating oil supply). It is not possible to accurately provide information on whether it is a problem with the device, cooling water supply device) or a problem with the rotating body.

In addition, when the operation limit values are set conservatively, abnormal alarms occur frequently, making it difficult to manage the monitoring system. Therefore, most of them should be set non-conservatively high to perform serious maintenance at the replacement level in case of an abnormality. In addition, quick measures are needed to prevent the spread of damage to the rotating shaft and couplings, but structurally, the components constituting the pump bearing system are very complex, so it takes a lot of time to analyze the cause.

On the other hand, digital twin refers to a technology for obtaining accurate information about the characteristics of real assets by creating twins of real objects on a computer and simulating situations that may occur in real life with a computer. Since it is possible to know the various states, productivity, and operation scenarios of real assets, it is attracting attention recently because it can improve the overall efficiency of production and services in various industries.

A digital twin is similar to a digital mockup in that it creates a clone. However, unlike mockups, which become obsolete once product design is complete and production and sales occur, digital twins persist long after the product is complete and the state data it collects can be provided to users of the product. have

Due to these advantages of digital twins, various processes and related technologies using digital twins are being developed throughout the industry. However, research on digital twins is still in the initial idea stage, and technology development or application for practical application to industry is insufficient. In addition, predictive diagnosis or monitoring technology for a pump bearing system to which such a digital twin is applied has not yet been proposed.

Republic of Korea Patent Registration No. 10-1988776 (Announcement Date: June 12, 2019, Title of Invention: Digital twinship system based on two-dimensional modeling)

The present invention was created to solve the problems of the prior art as described above, and it is possible to trace the cause of the defect by comprehensively analyzing various operating variables and monitoring variables related to the monitoring target facilities constituting the pump bearing system, and , a digital twin-based predictive diagnosis of a pump bearing system that enables early detection of fault occurrence using various diagnostic models, and enables users to intuitively identify the cause and situation of a fault by using and visualizing a digital twin model. The purpose is to provide a device.

The constituent means constituting the digital twin-based predictive diagnosis device of the pump bearing system according to the present invention proposed to solve the above problems is data collection that collects monitoring/operation data related to the monitoring target facilities of at least one pump bearing system. module, a data processing module that stores and manages the data collected by the data collection module and performs data communication with a linked server, an analysis that applies the predictive diagnosis basic information data received from the data processing module to at least one model; Prediction and diagnosis module that predicts and diagnoses faults of the pump bearing system and predicts optimal operating conditions through the simulation process, and the prediction and diagnosis result information data input from the prediction and diagnosis module is linked to the digital twin 3D model and processed through visualization It is characterized in that it is configured to include a digital twin model processing module to provide.

Here, the monitoring target equipment is a bearing set, a lubricant supply device, and a coolant supply device.

In addition, the data processing module is a database for storing and managing predictive diagnosis basic information data including monitoring/operation data about the monitoring target facilities collected by the data collection module, related device operation information data, and existing failure history information data wherein the prediction and diagnosis module applies the predictive diagnosis basic information data to at least one model and analyzes it to perform an operation of detecting the occurrence and type of a defect in the pump bearing system; the predictive diagnosis basic information data is applied to at least one model and simulated to predict the state change of the pump bearing system. It is characterized in that it comprises an integrated prediction and diagnosis unit that performs diagnosis, fault occurrence location identification, fault occurrence prediction, and optimal operating condition prediction of equipment.

According to the digital twin-based predictive diagnosis device of the pump bearing system of the present invention having the above problems and solutions, a defect occurs because various operating variables and monitoring variables related to the monitoring target facilities constituting the pump bearing system are comprehensively analyzed. The cause can be traced quickly, accurately and easily, and the occurrence of a fault can be detected early because of the use of various diagnostic models, and the cause and situation of the fault can be intuitively identified by the user because the digital twin model is used and visualized. There are advantages to being able to

In addition, according to the present invention, since the prediction and diagnosis module applies at least one analysis model and at least one simulation model, it is possible to solve the problem of difficulty in early detection of defects or failures of the existing monitoring system, and and immediate response action, which has the effect of enabling recovery through simple minor maintenance rather than serious maintenance such as replacement of parts.

In addition, according to the present invention, since the digital twin three-dimensional model information for the pump bearing system is provided to enable operation and control, the operator can easily recognize the structure and operation state information of the equipment to be monitored, and the abnormal state In case of occurrence, an advantage arises in that it is possible to minimize the time to identify and analyze the phenomenon.

1 is a schematic cross-sectional view of a general vertical type large pump bearing system.
2 is a block diagram of a digital twin-based predictive diagnosis apparatus of a pump bearing system according to an embodiment of the present invention.
3 is a block diagram of a prediction and diagnosis module constituting a digital twin-based predictive diagnosis apparatus of a pump bearing system according to an embodiment of the present invention.
4 is a model configuration diagram respectively included in the analysis unit and the simulation unit of the prediction and diagnosis module constituting the digital twin-based predictive diagnosis apparatus of the pump bearing system according to the embodiment of the present invention.
5 is an exemplary diagram of analysis through a data-based diagnosis model included in the analysis unit of the prediction and diagnosis module constituting the digital twin-based predictive diagnosis apparatus of the pump bearing system according to the embodiment of the present invention.
6 is an exemplary diagram of analysis through a machine learning diagnostic model included in the analysis unit of the prediction and diagnosis module constituting the digital twin-based predictive diagnosis apparatus of the pump bearing system according to the embodiment of the present invention.
7 is an exemplary analysis diagram through a rule-based diagnosis model included in the analysis unit of the prediction and diagnosis module constituting the digital twin-based predictive diagnosis apparatus of the pump bearing system according to the embodiment of the present invention.
8 is a block diagram of a digital twin model processing module constituting a digital twin-based predictive diagnosis apparatus of a pump bearing system according to an embodiment of the present invention.

Hereinafter, a preferred embodiment of a digital twin-based predictive diagnosis apparatus of a pump bearing system according to the present invention having the above problems, solutions and effects will be described in detail with reference to the accompanying drawings.

Since the present invention can apply various transformations and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description. Effects and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various forms.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, and when described with reference to the drawings, the same or corresponding components are given the same reference numerals, and the overlapping description thereof will be omitted. .

In the following embodiments, terms such as include or have means that the features or components described in the specification are present, and the possibility that one or more other features or components will be added is not excluded in advance.

In the drawings, the size of the components may be exaggerated or reduced for convenience of description. For example, since the size and thickness of each component shown in the drawings are arbitrarily indicated for convenience of description, the present invention is not necessarily limited to the illustrated bar.

In cases where certain embodiments may be implemented differently, a specific order of operations may be performed differently from the described order. For example, two operations described in succession may be performed substantially simultaneously, or may be performed in an order opposite to the described order.

2 is a block diagram of a digital twin-based predictive diagnosis apparatus of a pump bearing system according to an embodiment of the present invention.

As shown in FIG. 2 , the digital twin-based predictive diagnosis apparatus 100 of the pump bearing system according to the embodiment of the present invention includes a data collection module 10 that collects monitoring/operation data, stores and manages the collected data, and The data processing module 30 for performing data communication with various linked servers, the prediction and diagnosis module 50 for performing prediction and diagnosis related to the pump bearing system using the predictive diagnosis basic information data and at least one model, and the predictive diagnosis It is configured to include a digital twin model processing module 70 that links and processes the resulting information data to the digital twin 3D model and provides it through visualization processing.

The data collection module 10 performs an operation of collecting monitoring/operation data related to monitoring target facilities of at least one pump bearing system. The data collection module 10 may collect data related to only one pump bearing system, but collects data related to a plurality of pump bearing systems. Therefore, it is preferable that the monitoring/operation data input to the data collection module 10 is input together with the specific information of the corresponding pump bearing system or the monitoring target facilities of the pump bearing system.

Preferably, the monitored equipment does not include all components of the pump bearing system, but rather a bearing set, a lubricant supply device and a coolant supply device that closely affect bearing defects or damage. A plurality of various sensors for acquiring the monitoring/operation data are disposed or mounted on the bearing set, the lubricant supply device and the coolant supply device.

The sensors acquire monitoring/operation data related to the bearing set, the lubricant supply device and the coolant supply device, perform data conversion/processing as necessary, and then transmit the data to the data collection module 10 . Of course, the monitoring/operation data acquired from the sensors is transmitted to a separate sensing data collection module (not shown) and integrated, and then this sensing data collection module is integrated into the bearing set, lubricant supply device, and coolant supply device. It may be configured to transmit related monitoring/operation data to the data collection module 10 over a network.

It is preferable that the monitoring/operation data about the bearing set collected by the data collection module 10 include upper journal bearing temperature, lower journal bearing temperature, thrust bearing temperature, rotating body rotation speed, and rotating body vibration data. . Accordingly, corresponding sensors for sensing the corresponding monitoring/operation data are disposed or mounted around the journal bearing and thrust bearing constituting the bearing set and around the rotating body supported by the bearing set.

The monitoring/operation data about the lubricating oil supply device collected by the data collection module 10 preferably includes lubricating oil temperature, lubricating oil supply pressure, lubricating oil supply flow rate, lubricating oil storage tank level, and filter differential pressure. Accordingly, corresponding sensors for sensing the corresponding monitoring/operation data are disposed or mounted around the parts constituting the lubricant supply device.

The monitoring/operation data about the cooling water supply device collected by the data collection module 10 preferably includes a cooling water temperature, a cooling water supply pressure, and a cooling water supply flow rate. Accordingly, corresponding sensors for sensing the corresponding monitoring/operation data are disposed or mounted around the parts constituting the cooling water supply device.

In addition to the operation of receiving and collecting monitoring/operation data related to the bearing set, the lubricant supply device and the coolant supply device in real time, the data collection module 10 includes a digital twin-based predictive diagnosis device for the pump bearing system according to the present invention and Separately, it is possible to perform an operation of receiving and collecting data related to the pump bearing system from a previously built monitoring device.

Data collected from the data collection module 10, specifically monitoring/operation data related to the bearing set, lubricant supply device, and coolant supply device collected in real time, and furthermore, data related to the pump bearing system received from the existing monitoring device are transmitted to the data processing module 30 .

The data processing module 30 stores and manages the data collected by the data collection module 10 and performs data communication with a related linked server. The data processing module 30 includes a database 31 (indicated by reference numeral 31 in FIG. 3 ) for storing and managing the data collected by the data collection module 10 .

The data processing module 30 may perform data communication with a nearby associated server (not shown) to receive necessary information, for example, various information data related to the pump bearing system, and store and manage it in the database 31 . Also, it is possible to control the operation of the related linked server (not shown) as necessary. The link server may include a big data management server that stores and manages big data information data related to the pump bearing system and peripheral devices.

The database 31 transmits data additionally transmitted from the data collection module 10 as well as monitoring/operation data about the monitoring target facilities collected in real time, that is, from an existing monitoring device. Including related device operation information data and existing failure history information data included in the received pump bearing system related data and/or various information data related to the pump bearing system received by the data processing module 30 from the link server It stores and manages predictive diagnosis basic information data.

That is, the data processing module 30 is the predictive diagnosis basic information including the monitoring/operation data about the monitoring target facilities collected by the data collection module 10, the related device operation information data, and the existing failure history information data. A database 31 for storing and managing data is provided.

Predictive diagnosis basic information data including monitoring/operation data about the monitoring target facilities, related device operation information data, and existing failure history information data collected by the data collection module 10 stored and managed in the database 31 is recorded by the prediction and diagnosis module 50 and is used as basic information data for analysis and simulation.

The related device operation information data means operating information data regarding other components constituting the pump bearing system and/or other related devices disposed adjacent to the pump bearing system. In addition, the existing failure history information data means information data on the existing failure history related to the monitoring target facilities of the pump bearing system. These data are requested in order for the prediction and diagnosis module 50 to perform prediction and diagnosis on various faults of the pump bearing system through various models.

The prediction and diagnosis module 50 applies the predictive diagnosis basic information data input from the data processing module 30 to at least one model through an analysis and simulation process to predict and diagnose defects of the pump bearing system and optimal operating conditions. Performs an operation to perform a prediction.

The prediction and diagnosis module 50 predicts various defects of the pump bearing system by analyzing and simulating the predictive diagnosis basic information data composed of the various information data using at least one model built in advance, stored and managed, and At the same time as diagnosing, it performs an operation to predict the optimal operating conditions.

For this purpose, the prediction and diagnosis module 50 includes an analysis unit 51 that analyzes the predictive diagnosis basic information data input from the database 31 through at least one model, as shown in FIG. 3 , The simulation unit 53 for simulating the predictive diagnosis basic information data input from the database 31 through at least one model, and the operation of predicting and diagnosing the failure of the pump bearing system integrally through the analysis and simulation results It is configured to include an integrated prediction and diagnosis unit 55 that performs

Specifically, the prediction and diagnosis module 50 includes an analysis unit 51 that applies and analyzes the predictive diagnosis basic information data to at least one model to detect the occurrence and type of a defect in the pump bearing system, the The simulation unit 53, which performs an operation of predicting the state change of the pump bearing system by applying and simulating the predictive diagnosis basic information data to at least one model, and the results of the analysis unit 51 and the simulation unit 53 It comprises an integrated prediction and diagnosis unit 55 that performs fault occurrence diagnosis, fault type diagnosis, fault occurrence cause diagnosis, fault occurrence location identification, fault occurrence prediction, and optimal operating condition prediction by comprehensive analysis.

As shown in FIG. 4 , the models applied by the analysis unit 51 include a data-based diagnostic model, a machine learning diagnostic model, a rule-based diagnostic model, and a statistical analysis model. The analysis unit 51 applies at least one of these models to analyze the predictive diagnosis basic information data to detect whether a defect has occurred regarding whether a defect in the pump bearing system has occurred, and furthermore, the type of the generated defect, e.g. For example, it performs an operation to detect the type of defect such as rotation axis misalignment or friction.

Among the models applied by the analysis unit 51, the Data-driven Diagnosis Model causes a problem in the pump bearing system (bearing set and related auxiliary devices (lubricating oil supply device and coolant supply device)) It is a diagnostic model for early detection of whether or not Accordingly, the analysis unit 51 can detect the occurrence of a defect in the pump bearing system early through this data-based diagnostic model.

The data-based diagnostic model is a variable or genetic algorithm with high correlation (Correlation Method, correlation value of 0.6 to 0.7 or more) with respect to various sensor measurement data monitoring the bearing set constituting the pump bearing system and related auxiliary devices. It is created by grouping the optimized variables through (Genetic Algorithm). As shown in FIG. 5 , the generated data-based diagnostic model can form a dynamic band, and the occurrence of a defect can be detected early by applying and analyzing the predictive diagnostic basic information data. .

Specifically, the analysis unit 51 evaluates the residual by comparing the actual sensor measurement value during operation of the pump bearing system and the predicted value output through the data-based diagnostic model using the data-based diagnostic model, and , it is possible to detect the occurrence of defects early by detecting the monitoring variables and the occurrence timing of which the residual amount deviates from the dynamic band (before the threshold is reached).

In addition, among the models applied by the analysis unit 51, the machine learning diagnostic model (Machine Learning Diagnosis Model) is from a vibration signal indicating the dynamic behavior of the rotating body, bearing wear due to problems such as misalignment or installation failure of the shaft. It is a model for detecting whether a fault has occurred. Accordingly, the analysis unit 51 can detect the failure type as well as the occurrence of a failure in the pump bearing system at an early stage through this machine learning diagnostic model.

The machine learning diagnostic model is generated using SVM and identifiers such as k-NN and LDA after calculating statistical feature variables representing signal characteristics from vibration signals, selecting feature information that is easy to identify defects, and the like. As shown in FIG. 6 , the analysis unit 51 using the machine learning diagnostic model generated as described above calculates a feature amount using the value measured by the vibration sensor during pump operation, and then the generated machine learning diagnosis Defect occurrence and defect types can be detected by inputting into the model and evaluating whether the feature quantity belongs to the steady-state region or the defective state region.

In addition, among the models applied by the analysis unit 51, the Rule-based Diagnosis Model of the pump bearing system is based on the vibration state of the rotating body including the bearing and the operation state information of related auxiliary devices. It is generated by forming rule-sets for fault diagnosis according to the result of the failure cause analysis (RCA) (see FIG. 7 ). Accordingly, the analysis unit 51 may detect the failure type as well as the occurrence of a failure in the pump bearing system at an early stage through this rule-based diagnostic model, and may further detect the cause of the failure.

As shown in FIG. 7 , the analysis unit 51 sets a plurality of fault diagnosis rule sets of the generated rule-based diagnosis model after receiving the vibration signal state monitoring variables and operation state information of related auxiliary devices during pump operation. Among them, it is possible to evaluate the occurrence of a situation that satisfies the relevant requirements to detect the occurrence and type of the defect and the cause of the defect.

In addition, among the models applied by the analysis unit 51 , the statistical analysis model corresponds to a statistical chart data model that has been previously formed and constructed. The analysis unit 51 receives the vibration signal state monitoring variables and operation state information of related auxiliary devices during pump operation, and compares and analyzes it with a statistical chart corresponding to the statistical analysis model to detect abnormalities and defects in the pump bearing system. can be detected early.

Meanwhile, as shown in FIG. 5 , the models applied by the simulation unit 53 include a data model, a physical model, and a hybrid model. The simulation unit 53 applies at least one of these models to analyze the predictive diagnosis basic information data to track and predict the state change of the pump bearing system. Through this, it is possible to predict the occurrence of defects in the pump bearing system or the monitoring target facilities constituting the same, and furthermore, it is possible to track and predict the optimal operating conditions of the monitoring target facilities.

Among the models applied by the simulation unit 53, the data model corresponds to a model generated based on correlation information of monitoring and operation variables of the pump bearing system or a model generated based on a hydrodynamic relational expression of a piping system, and , the simulation unit 53 may predict the state change of the output variable according to the change in the condition of the input variable to predict the operating state of the pump bearing system or track or predict the optimal operating condition of the equipment to be monitored.

Among the models applied by the simulation unit 53, the physical model is a model generated based on a mechanical relationship representing the structure and relationship of the pump bearing system, and predicts the state change of the output variable according to the condition of the input variable to determine the pump bearing system. It is possible to predict the operating state of the bearing system or to track or predict the optimal operating conditions of the equipment to be monitored. To verify the generated physical model, a statistical hypothesis test or a Bayesian method is applied.

Among the models applied by the simulation unit 53, the hybrid model is a method of using a mixture of the physical model and the data model, and is generated by compensating for disadvantages such as limitations in the operating range of the data model and difficulties in developing the physical model. corresponds to the model

The results of the analysis performed through the at least one model of the analysis unit 51 and the results of the simulation performed through the at least one model of the simulation unit 53 are transmitted to the integrated prediction and diagnosis unit 55 , and the integrated prediction and the diagnosis unit 55 comprehensively analyzes the performance results to perform various diagnoses and predictions related to the final defect.

That is, the integrated prediction and diagnosis unit 55 comprehensively analyzes the results of the analysis unit 51 and the simulation unit 53 to diagnose the occurrence of a defect, diagnose the type of the defect, diagnose the cause of the defect, identify the location of the defect, Prediction of occurrence of defects and prediction of optimal operating conditions of facilities.

The fault occurrence diagnosis is a diagnosis of whether a fault has occurred in the pump bearing system or the equipment to be monitored constituting them, and the fault type diagnosis is a diagnosis of whether the generated fault is mass unbalance, misalignment, shaft crack ( Shaft crack), contact wear (rubbing), and looseness (looseness) is a diagnosis for which type, and the diagnosis of the cause of a defect is a diagnosis of what is the main cause of the defect (e.g., an increase in lubricating oil temperature). Diagnosis, fault location identification is identification of the location where a fault has occurred, fault occurrence prediction is a prediction of whether a fault will occur soon and when it will occur, and prediction of optimal operating conditions of a facility is It means prediction or tracking of driving conditions that can be operated normally.

The integrated prediction and diagnosis unit 55 may generate and apply an integrated diagnostic model. The integrated diagnostic model corresponds to a comprehensive diagnostic model for diagnosing the occurrence and type of a defect early by comprehensively analyzing the performance results of the various models described above, and finally identifying the cause and location of the defect. As the integrated diagnostic model, any one of various models may be applied. In the present invention, an ensemble technique is adopted and applied. The ensemble technique is an excellent technique for making a final decision by synthesizing the results of a plurality of diagnostic models. Of course, other similar models are applicable.

The integrated prediction and diagnosis unit 55 comprehensively analyzes the results of the analysis unit 51 and the simulation unit 53, and the defect occurrence diagnosis, defect type diagnosis, defect occurrence cause diagnosis, and defect corresponding to the results performed. Identification of occurrence location, prediction of occurrence of defects, and prediction of optimal operating conditions of equipment, that is, predictive diagnosis result information data are transmitted to the digital twin model processing module 70 Then, the digital twin model processing module 70 identifies these data It performs an action that is applied to the digital twin 3D model and visualized to provide it to users or administrators.

That is, the digital twin model processing module 70 processes the predictive diagnosis result information data input from the prediction and diagnosis module 50 in connection with the digital twin 3D model and provides the data through visualization.

The digital twin model processing module 70 selects any one of a plurality of digital twin 3D models, creates a digital object by linking the predictive diagnosis result information data, and provides visualization as a digital twin 3D model. do. Information data that is visualized and provided with the digital twin 3D model includes not only the predictive diagnosis result information data but also data collected in real time, that is, monitoring/operation data on the monitoring target facilities. Accordingly, the prediction and diagnosis module 60 transmits the predictive diagnosis result information data and the monitoring/operation data regarding the monitoring target facilities together to the digital twin model processing module 70, and furthermore, to these data It transmits the unique information data of the related pump bearing system or the equipment to be monitored constituting them together.

The digital twin 3D model is a digital object that can diagnose the current state of the system by combining data and information representing the structure and operation of the physical system of the pump bearing system, and predict the occurrence of faults and the optimal operating state of the equipment.

As shown in FIG. 8, the digital twin model processing module 70 includes a digital twin control unit 71, a model/content storage unit 73, a digital object generation unit 75, and an HMI/visualization processing unit 77. is comprised of

The digital twin control unit 71 receives the predictive diagnosis result information data and monitoring/operation data regarding the monitoring target facilities, and furthermore, the pump bearing system related to these data or the unique monitoring target facilities constituting these data. Receive information data.

Then, the digital twin control unit 71 selects at least one of the plurality of digital twin 3D models stored and managed in the model/content storage unit 73 and at least one of the plurality of visualization contents based on the received information data. It is transmitted to the digital object generating unit 75 . In addition, the digital twin control unit 71 transmits the received predictive diagnosis result information data and monitoring/operation data regarding the facilities to be monitored to the digital object generating unit 75 .

The model/content storage unit 73 stores various digital twin 3D models and various visualization contents so that the predictive diagnosis result information data and monitoring/operation data related to the monitoring target facilities can be visualized through the 3D model. store and manage The model/content storage unit 73 may store and manage at least one digital twin 3D model corresponding to each pump bearing system or a monitoring target facility constituting them, and the predictive diagnosis result information data and the monitoring target It is possible to store and manage at least one visualization content corresponding to each of the monitoring/operation data regarding the facilities.

The digital twin control unit 71 may analyze data including the unique information data transmitted from the prediction and diagnosis module 50 to select a corresponding digital twin 3D model and at least one visualization content. The digital twin control unit 71 controls the selected digital twin 3D model and at least one visualization content to be transmitted from the model/content storage unit 73 to the digital object generation unit 75 .

The digital object generating unit 75 includes the data transmitted from the digital twin control unit 71 to the received digital twin 3D model and at least one visualization content, that is, the predictive diagnosis result information data and the monitoring target facilities. related monitoring/operation data are linked and provided to the HMI/visualization processing unit 77 .

The HMI/visualization processing unit 77 performs an operation of providing at least one visualization content provided from the digital twin 3D model to the user. The HMI/visualization processing unit 77 visualizes and provides 3D model-based operation and fault early diagnosis information of the pump bearing system, visualizes and provides 3D model-based simulation information of the pump bearing system, and provides 3D model-based operation and The analysis of the structure of the pump bearing system through control is visualized and provided, and furthermore, by providing the user signal analysis screen, it is possible to provide a function to track the cause of the failure.

The digital twin model processing module 70 can dramatically improve the limitation of information provision of the existing monitoring system that provides only simple graphs and numerical information. In particular, by providing three-dimensional model information about the pump bearing system to enable operation and control, the operator can easily recognize the structure and operating state information of the equipment to be monitored, and it reduces the time to identify and analyze phenomena when abnormal conditions occur. It has the advantage of being able to minimize it.

Although the embodiments according to the present invention have been described above, these are merely exemplary, and those of ordinary skill in the art will understand that various modifications and equivalent ranges of embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be defined by the following claims.

1: Pump bearing system 3: Bearing set
5: Lubricating oil supply device 7: Cooling water supply device
10: data collection module 30: data processing module
31: database 50: predictive and diagnostic module
51: analysis unit 53: simulation unit
55: integrated prediction and diagnosis unit 70: digital twin model processing module
71: digital twin control unit 73: model/content storage unit
75: digital object creation unit 77: HMI/visualization processing unit
100: Digital twin-based predictive diagnosis device of pump bearing system

Claims (3)

a data collection module that collects monitoring/operation data related to monitoring target facilities of at least one pump bearing system; a data processing module that stores and manages the data collected by the data collection module and performs data communication with an associated server; a prediction and diagnosis module for predicting and diagnosing defects of the pump bearing system and predicting optimal operating conditions through an analysis and simulation process of applying the predictive diagnosis basic information data received from the data processing module to at least one model; It comprises a digital twin model processing module that provides through visualization by linking and processing the predictive diagnosis result information data input from the prediction and diagnosis module to the digital twin 3D model,
The equipment to be monitored is a bearing set, a lubricating oil supply device and a coolant supply device, which closely affect the bearing failure or damage of the pump bearing system;
The monitoring/operation data about the bearing set collected by the data collection module includes upper journal bearing temperature, lower journal bearing temperature, thrust bearing temperature, rotating body rotation speed and rotating body vibration data, and the data collecting module is Monitoring/operation data related to the lubricating oil supply device that is collected includes lubricating oil temperature, lubricating oil supply pressure, lubricating oil supply flow rate, lubricating oil storage tank level and filter differential pressure, and monitoring/operation of the cooling water supply device collected by the data collection module The data include coolant temperature, coolant supply pressure, coolant supply flow rate,
The data processing module has a database for storing and managing predictive diagnosis basic information data including monitoring/operation data on the monitoring target facilities collected by the data collection module, related device operation information data, and existing failure history information data do,
The prediction and diagnosis module includes an analysis unit that applies and analyzes the predictive diagnosis basic information data to at least one model to detect the occurrence and type of a defect in the pump bearing system, and collects the predictive diagnosis basic information data into at least one A simulation unit that performs an operation to predict the state change of the pump bearing system by applying it to the model and comprehensively analyzing the results of the analysis unit and the simulation unit to diagnose fault occurrence, fault type diagnosis, fault cause diagnosis, fault occurrence It includes an integrated prediction and diagnosis unit that identifies the location, predicts the occurrence of defects, and predicts the optimal operating conditions of the facility,
The models applied in the simulation unit include a data model, a physical model, and a hybrid model, and the simulation unit analyzes the predictive diagnosis basic information data through at least one model among these models to analyze the pump bearing system or a monitoring target constituting the same. Predict the occurrence of defects in facilities and track and predict the optimal operating conditions of the facilities to be monitored,
The diagnosis of the cause of the defect is a diagnosis of what is the main cause of the occurrence of the defect, the identification of the location of the defect is the identification of the location where the defect has occurred, and the prediction of the occurrence of the defect is the diagnosis of whether the defect will occur soon and when it will occur. Prediction of optimal operating conditions of facilities means prediction or tracking of operating conditions under which monitoring target facilities can operate normally without failure,
The digital twin model processing module is configured to include a digital twin control unit, a model/content storage unit, a digital object generation unit, and an HMI/visualization processing unit,
The digital twin control unit selects at least one of a plurality of digital twin 3D models stored and managed in the model/content storage unit and at least one of a plurality of visualization contents based on the received information data and transmits it to the digital object generation unit, , transmits the received predictive diagnosis result information data and monitoring/operation data on the monitoring target facilities to the digital object generator,
The model/content storage unit stores and manages at least one digital twin 3D model corresponding to each pump bearing system or a monitoring target facility constituting them, and monitors/manages the predictive diagnosis result information data and the monitoring target facilities Storing and managing at least one visualization content corresponding to each of the driving data,
The digital object generating unit links the received digital twin 3D model and at least one visualization content with the predictive diagnosis result information data transmitted from the digital twin control unit and the monitoring/operation data regarding the facilities to be monitored, An operation is performed to provide this to the HMI/visualization processing unit,
The HMI/visualization processing unit visualizes and provides 3D model-based operation and fault early diagnosis information of the pump bearing system, visualizes and provides 3D model-based simulation information of the pump bearing system, and provides 3D model-based operation and control A digital twin-based predictive diagnostic device for a pump bearing system, characterized in that it provides visualization of the structure analysis of the pump bearing system.
The method according to claim 1,
The monitoring target equipment is a digital twin-based predictive diagnosis device for a pump bearing system, characterized in that it is a bearing set, a lubricant supply device, and a coolant supply device.
delete

Images