JP6784551B2 - Condition monitoring system - Google Patents

Description

The present invention relates to a condition monitoring system capable of appropriately modeling the state of a device even when the state of the device changes.

Patent Document 1 describes, "A simplified model calculation means for expressing a device element with a transfer function model using parameters of a detailed dynamic characteristic model and modeling the input / output of an element block by a transfer function model, a step response of an actual machine. Achievement of calculating values such as transfer function gain and time constant from waveforms Values such as transfer function gain and time constant used in simplified model calculation means and detailed dynamic characteristic model parameters and their values It consists of a means for displaying the relational expression, values such as the gain and time constant of the transfer function calculated by the actual waveform feature extraction means, and the amount of parameter correction for matching the simplified model transfer function with the transfer function calculated by the means. The tuning of the plant simulator dynamic characteristic model parameters can be performed accurately by a logically clear method based on the modeling of physical phenomena, and the time required for adjusting the control system of the plant can be shortened. "

Further, according to Non-Patent Document 1, "By using a physical model for equipment simulation, it is possible to model in detail so as to include the cause of failure of the component parts of the device, and it is possible to perform failure diagnosis of the device and the remainder. It can be used for life prediction. "

Japanese Unexamined Patent Publication No. 9-6407

Tadao Kawai, "Preliminary prediction of troubles occurring in equipment and application of physical model to life expectancy diagnosis", [online], [Search on June 1, 2016], Internet <URL: http://www.jsme .or.jp/conference/joutai/doc/kaisai/20121012/kawai.pdf>

It is difficult to simulate an actual phenomenon with the simulation result of physically modeling the target device, and it takes a lot of labor to match the analysis result with the actual operating state of the device. Patent Document 1 describes tuning parameters by a preconfigured plant simulator. However, it is difficult to incorporate data such as abnormal behavior that may occur when constructing a plant simulator in the same operating conditions as actual operating equipment, and it is necessary to further improve the simulation results.

Further, Non-Patent Document 1 describes that the state of the device can be faithfully reproduced by modeling the component parts of the device including the cause of the failure. However, in reality, the state of the device changes every day, and the model once built does not match the latest state of the device after a lapse of time. Therefore, it is difficult to simulate the same result as the latest equipment only by adjusting the parameters of the model once constructed.

The present invention is an invention for solving the above-mentioned problems, and an object of the present invention is to provide a condition monitoring system capable of appropriately modeling the state of a device even when the state of the device changes. ..

In order to solve the above-mentioned problems, the state monitoring system of the present invention stores a phenomenon component in which the element part caused by the abnormal behavior is pre-modeled on the physical model of the device according to the operating state of the device ( For example, the phenomenon component storage unit 8), the device data acquisition means for acquiring the state data from the sensor attached to the device, the model analysis result acquisition means for acquiring the analysis result simulated from the physical model of the device, and the state data. A comparison means for comparing the analysis results, a parameter adjustment means for adjusting the parameters of the physical model, a phenomenon component selection means for incorporating the phenomenon component into the physical model, and a physical model for the phenomenon component selection means based on the comparison result of the comparison means. It is characterized by having a phenomenon component adjusting means (for example, a phenomenon component controlling means 4) that instructs the parameter adjusting means to adjust the parameters of the selected phenomenon component while instructing the selection of the phenomenon component required for the above. To do. Other aspects of the present invention will be described in embodiments described below.

According to the present invention, even when the state of the device changes, the state of the device can be appropriately modeled.

It is a figure which shows the whole structure of the state monitoring system which concerns on this embodiment. It is a figure which shows the whole structure of the state monitoring system when the phenomenon component is replaced with the phenomenon component which was in the state of the present equipment. It is a figure which shows the device example of the actual machine for demonstrating this embodiment. It is a figure which shows the modeling of the actual machine example of this embodiment. It is a figure which shows the phenomenon component of this embodiment, (a) is a figure which shows the basic phenomenon component, (b)-(d) is the figure which shows the other phenomenon component. It is a flowchart which shows the process of the condition monitoring system which concerns on this embodiment.

Embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
FIG. 1 is a diagram showing an overall configuration of a condition monitoring system according to the present embodiment. FIG. 2 is a diagram showing an overall configuration of a state monitoring system when the phenomenon component is replaced with a phenomenon component that matches the state of the current device. The condition monitoring system 1 may be realized in a local environment such as a general PC (Personal Computer) or via a network such as a cloud.

The target device 10 is composed of a plurality of parts 11, 12, and 13. The physical model 30 for the device includes basic phenomenon components 31, 32, 33, which are physical models corresponding to the components of the device 10. The state monitoring system 1 includes device data acquisition means 2, comparison means 3, phenomenon component control means 4 (phenomenon component adjustment means), model analysis result acquisition means 5, parameter adjustment means 6, phenomenon component selection means 7, and phenomenon component storage unit. It has 8 (database).

Sensors 21 and 22 such as vibration sensors and acceleration sensors are attached to the device 10. The device data acquisition means 2 acquires the sensor data and the device output. The model analysis result acquisition means 5 acquires the analysis value of the phenomenon component constituting the physical model 30 and the output analysis value of the entire physical model 30. The comparison means 3 compares the data acquired by the device data acquisition means 2 and the model analysis result acquisition means 5, and passes the comparison result to the phenomenon component control means 4 (phenomenon component adjustment means). For example, in the case of a device provided with a rotating portion, the bearing serving as the rotating support portion generates vibrations of the housing serving as a housing and various vibrations due to a sliding contact phenomenon peculiar to a bearing. In particular, the sliding portion causes a physical vibration phenomenon due to friction, wear, contact marks, and the like. A physical model that simulates these phenomena is used as a phenomenon component.

The phenomenon component control means 4 determines the comparison result based on threshold conditions such as vibration amplitude and frequency. The phenomenon component selection means 7 selects the phenomenon components 41 and 42 from the phenomenon component storage unit 8 and replaces them with the corresponding basic phenomenon components 31 and 32 on the physical model 30 (see FIG. 2). The parameter adjusting means 6 adjusts the parameters of the phenomenon components 41 and 42.

In the present embodiment, the data acquired by the device data acquisition means 2 and the model analysis result acquisition means 5 are compared by the comparison means 3, and the phenomenon component control means 4 determines the phenomenon component selection means 7 and the parameters based on the results. Using the adjusting means 6, the operation state of the device 10 and the physical model 30 (vibration amplitude, frequency, etc. described above) is repeated until it is determined that the phenomena are substantially the same.

FIG. 3 is a diagram showing an example of an actual device for explaining the present embodiment. The device includes a motor 300, bearings 301a and 301b, a rotor 302, and a rotating shaft 305. A sensor 21 is installed on the bearing 301a.

FIG. 4 is a diagram showing modeling of an actual machine example of the present embodiment. FIG. 4 is a physical model of the device shown in FIG. 3, and includes model parts 400 corresponding to the motor 300, model parts 401a and 401b corresponding to the bearings 301a and 301b (collectively, model parts 401). It is composed of a model part 402 corresponding to the rotor 302 and a model part 405 corresponding to the rotating shaft 305.

5A and 5B are diagrams showing a phenomenon component of the present embodiment, in which FIG. 5A shows a basic phenomenon component, and FIGS. 5B to 5D show other phenomenon components. The phenomenon component shown in FIG. 5 is an example of the phenomenon component held in the phenomenon component storage unit 8. It is desirable to prepare multiple phenomenon components for each phenomenon such as damage that may occur in the parts of the corresponding equipment, and this is constructed by experiments, simulations, and past databases.

The phenomenon component 501a corresponds to the bearings 301a and 301b of the device and models the vibration characteristics of the housing, and has parameters such as elastic modulus (spring constant) and damping rate. Phenomenon components 501b, 501c, 501d are other examples of phenomenon components 501a. The phenomenon components 501a, 501b, 501c, and 501d are collectively referred to as the phenomenon component 501.

The phenomenon component may be a vibration model of these one-degree-of-freedom system or n-degree-of-freedom system, and may be a series configuration, a parallel configuration, or a combination thereof. Further, the vibration model including the sliding characteristics of the bearing may be used. For example, in the case of ball bearings, contact vibration due to scratches or wear marks on the ball portion is assumed, and in the case of sliding bearings, contact vibration due to friction or wear marks is assumed, and a vibration model is constructed. Furthermore, the phenomenon component is equipped with a structural model that incorporates not only the bearing but also the influence of the peripheral structure, such as incorporating the elastic vibration mode of the shaft connected to the bearing and its peripheral structures into the vibration model in advance. You may be.

That is, the phenomenon component may be composed of a vibration model of one degree of freedom system or n degree of freedom system. The phenomenon component may be a series configuration, a parallel configuration, or a combination of vibration models of a one-degree-of-freedom system or an n-degree-of-freedom system. The phenomenon component may be composed of a vibration model having one degree of freedom or n degrees of freedom, and may include a structural model as a peripheral structure thereof.

FIG. 6 is a flowchart showing the processing of the condition monitoring system according to the present embodiment. Refer to FIGS. 1 and 2 as appropriate. The phenomenon component control means 4 determines the result of the comparison means 3 (step S1). If the result does not satisfy the comparison condition (step S1, mismatch), the process proceeds to step S2, and if the result satisfies the comparison condition (step S1, match), the process ends.

In step S2, the phenomenon component control means 4 determines whether or not the phenomenon component is applied to the physical model. When the phenomenon component is not applied (steps S2 and No), the phenomenon component control means 4 instructs the phenomenon component selection means 7 to select the phenomenon component, and the phenomenon component selection means 7 is transmitted from the phenomenon component storage unit 8. Select the phenomenon component (step S5), and incorporate the phenomenon component into the target part on the physical model.

When the phenomenon component is applied (step S2, Yes), the phenomenon component control means 4 determines whether or not the parameter of the phenomenon component has been adjusted (step S3). If the parameters are not adjusted (steps S3 and No), the parameter adjustment means 6 performs parameter adjustment (step S4). After adjusting the parameters, the comparison means 3 executes a comparison of the data acquired by the device data acquisition means 2 and the model analysis result acquisition means 5.

When all the parameter adjustments have been performed in step S3 (steps S3 and Yes), it is determined whether or not there is any omission of application of the phenomenon component (step S6). When there is an application omission (step S6, No), the phenomenon components 41 and 42 are selected by the phenomenon component selection means 7, and the basic phenomenon components 31 and 32 (see FIG. 1) on the physical model are replaced with the phenomenon components 41 and 42 (see FIG. 1). (See FIG. 2). If there is no omission of application (steps S6, Yes), a series of processes is completed. Hereinafter, each means will be described in detail.

(Comparison means)
As shown in FIG. 1, the comparison means 3 compares the sensor data of the sensor 21 attached to the component 11 of the device with the analysis result of the basic phenomenon component 31 corresponding to the sensor data. Further, the sensor data of the sensor 22 of the part 13 and the analysis result of the basic phenomenon component 33 are compared. Further, the output of the device 10 and the analysis output result of the physical model 30 are compared. The comparison may be made not only for each sensor but also for a combination of sensors and outputs. This is because an abnormality in the device affects not only the sensor mounting location but also the sensor in another location. As shown in FIG. 2, when the basic phenomenon component 31 is replaced by the phenomenon component 41, the analysis result of the phenomenon component 41 and the sensor 21 are compared. For comparison, the difference between the sensor data and the analysis output result may be set so as to be within a certain range.

(Phenomenon component control means)
The phenomenon component control means 4 adjusts, for example, the size of scratches on the parameter inner ring and outer ring of the phenomenon component 501. The adjustment method sequentially changes the possible range of the parameter. In addition, if a plurality of phenomenon components are arranged in the physical model, the parameters of each are changed. When the parameters are changed, the analysis result of each phenomenon component and the actual machine data are compared by the comparison means 3, and the comparison is repeated until the convergence condition is reached.

(Phenomenon component selection means)
The phenomenon component selection means 7 preferentially selects the phenomenon component 501a corresponding to the part 301 of the device targeted by the sensor 21 from the phenomenon component storage unit 8 and replaces it with the model part 401. When the phenomenon component control means 4 requests the phenomenon component again, for example, a different phenomenon component 501b is selected and replaced with the phenomenon component 501a. Since abnormalities in equipment do not appear only in one sensor output but often appear in multiple locations, it is better to apply not only the relevant locations but also multiple combinations of phenomenon components. .. In addition, when selecting a phenomenon component, efficiency can be improved by adopting a method of prioritizing and selecting in order of priority according to past cases or a method of prioritizing the phenomenon component related to the phenomenon that can be read from the sensor data. ..

(Parameter adjustment means)
The parameter adjusting means 6 adjusts, for example, the size of scratches on the parameter inner ring and outer ring of the phenomenon component 501. The adjustment method sequentially changes the possible range of the parameter. In addition, if a plurality of phenomenon components are arranged in the physical model, the parameters of each are changed. When the parameters are changed, the analysis result of each phenomenon component and the actual machine data are compared by the comparison means 3, and the comparison is repeated until the convergence condition is reached. The order in which the parameters are changed can be improved in efficiency by prioritizing and applying them according to past experiments and cases.

The present embodiment is not limited to the above-described embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add / delete / replace a part of the configuration of each embodiment with another configuration.

In addition, each of the above configurations may be configured in part or in whole in hardware, or may be configured to be realized by executing a program in a processor. In addition, the control lines and information lines indicate those that are considered necessary for explanation, and do not necessarily indicate all the control lines and information lines in the product. In practice, it can be considered that almost all configurations are interconnected.

In the condition monitoring system 1 of the present embodiment, the element part caused by the abnormal behavior is modeled in advance on the physical model of the device according to the operating state of the device, and the element part model is appropriately modeled based on the actual operation information. It has a function to be incorporated into selection and simulation. As a result, it is possible to provide a system that constantly follows the actual operating state of the device, incorporates a model including abnormal behavior that can occur in the device, and simulates it, and monitors the state of the device based on the result. It becomes.

According to the present embodiment, the physical model of the target device is actually operated by incorporating the phenomenon component corresponding to the actual behavior of the device into the physical model based on each sensor information according to the operating state of the device. A simulation that matches well with is possible. As a result, the condition monitoring of the device can be realized with high accuracy. In addition, since the device model always matches the latest device status, it can be used for equipment failure sign diagnosis by model simulation.

1 Condition monitoring system 2 Equipment data acquisition means 3 Comparison means 4 Phenomenon component control means (Phenomenon component adjustment means)
5 Model analysis result acquisition means 6 Parameter adjustment means 7 Phenomenon component selection means 8 Phenomenon component storage unit (database)
10 Equipment 11, 12, 13 Parts 30 Physical model 31, 32, 33 Basic phenomenon component 41, 42 Phenomenon component 300 Motor 301 Bearing 302 Rotor 305 Rotating shaft 400, 401, 402, 405 Model part 501 Phenomenon component

Claims (5)

A database that stores phenomenon components that pre-model the elements caused by abnormal behavior on the physical model of the device according to the operating state of the device .
A device data acquisition means for acquiring status data from a sensor attached to the device, and
Model analysis result acquisition means for acquiring simulation results from the physical model of the device, and
A comparison means for comparing the state data with the analysis result,
A parameter adjusting means for adjusting the parameters of the physical model and
Phenomenon component selection means for incorporating the phenomenon component into the physical model,
Based on the comparison result of the comparison means, the phenomenon of instructing the phenomenon component selection means to select the phenomenon component required for the physical model and instructing the parameter adjustment means to adjust the parameters of the selected phenomenon component. A state monitoring system characterized by having component adjusting means. The first aspect of claim 1, wherein the phenomenon component adjusting means repeats instructions to the phenomenon component selecting means and the phenomenon component adjusting means until the comparison result of the comparing means reaches a set threshold condition range. Condition monitoring system. The condition monitoring system according to claim 1, wherein the phenomenon component is composed of a vibration model of one degree of freedom system or n degree of freedom system. The condition monitoring system according to claim 1, wherein the phenomenon component has a series configuration, a parallel configuration, or a combination thereof of vibration models of one degree of freedom system or n degree of freedom system. The condition monitoring system according to claim 1, wherein the phenomenon component is composed of a vibration model of one degree of freedom system or an n degree of freedom system and is provided with a structural model serving as a peripheral structure thereof.

Images