Condition Monitoring System

Abstract

To provide a state monitoring system that can be adjusted on an edge side and ensure real-time processing performance.SOLUTION: A state monitoring system for monitoring a state of a facility includes: a sensor attached to the facility; a data measuring device that receives a detection signal of the sensor and acquires measurement data from the detection signal in accordance with a predetermined measurement condition; and a data diagnostic device that receives the measurement data from the data measuring device and executes diagnostic processing to diagnose the state of the facility on the basis of the measurement data. In the state monitoring system, the data diagnostic device has an edge application and an industrial IoT platform, the edge application has a data collecting / analyzing section, and the data collecting / analyzing section calculates a feature quantity of the measurement data from the data measuring device and delivers the feature quantity to the industrial IoT platform.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a status monitoring system that monitors the status of equipment. [Background technology]

[0002] 2. Description of the Related Art Condition monitoring systems are known that perform processing such as calculation of effective values ​​and frequency analysis on measurement data collected using sensors installed in industrial or other facilities, and monitor and diagnose the condition of the facilities based on the results of the processing.

[0003] As a conventional technology, a data system for a production environment is known that uses IoT (Internet of Things) technology to adjust various parameters such as the sampling period and scaling value of a sensor in response to the input of multiple sensor data, taking into account the input value of the sensor and the network throughput (Patent Document 1). In addition, in a system for continuously monitoring multiple machines, in which multiple devices are capable of operating bidirectionally, a technology is known in which multiple measured process parameters are acquired in real time, a derived amount is determined from the process parameters, and a change is recommended for the operation of the device based on the derived amount or the process parameters (Patent Document 2). Furthermore, in a status monitoring system that collects and analyzes measurement data using sensors installed in equipment and monitors the equipment status based on the analysis results, a technology is known in which the measurement conditions and calculation parameters of the vibration analysis are adjusted based on the time required for vibration analysis and the time required for communication in order to ensure real-time performance, and the calculation time is adjusted (Patent Application No. 2020-163942). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2020-530159 [Patent Document 2] Patent No. 5295482 Summary of the Invention [Problem to be solved by the invention]

[0005] There are cases where a real-time state monitoring performance is required for a condition monitoring system that monitors the state of equipment. For example, in a production site or the like, when performing analysis processing such as frequency analysis on equipment measurement data (vibration data, etc.) and diagnosing the equipment based on the results of the analysis processing, if the diagnosis results are not responsive, the detection of anomalies in the equipment and the response to the anomalies will be delayed. In such cases, real-time performance is required for the analysis processing of the measurement data and the diagnosis processing based on the analysis results.

[0006] In recent years, as a form of performing real-time processing at the production site, there is a system form using an industrial IoT platform that provides functions such as analysis processing of the above measurement data and diagnosis processing based on the analysis results only on the edge side (terminals, terminal networks, and points that send data collected in their vicinity to the cloud in the field of IoT technology, etc.). However, if the data collection and distribution function in this industrial IoT platform provided only on the edge side is a function for data with a low sampling frequency (several Hz) compared to vibration acceleration data, such as temperature and pressure, it cannot be said that it is suitable for collecting vibration acceleration data with a high sampling frequency (tens of thousands of Hz or more). In addition, the data collection software included in the industrial IoT platform is generally tailored to the communication standards of the production equipment side, and is not designed with the type of data collected from the edge side sensors, the target of status monitoring, or the method of detecting anomalies in mind.

[0007] The present invention has been made to solve the above problems, and has as its object to provide a status monitoring system that can be adjusted on the edge side and ensures real-time processing. [Means for solving the problem]

[0008] In order to achieve the above object, the status monitoring system according to the present invention comprises: A status monitoring system for monitoring a status of equipment, comprising: A sensor attached to the facility; a data measurement device that receives a detection signal from the sensor and acquires measurement data from the detection signal in accordance with a predetermined measurement condition; a data diagnosis device that receives the measurement data from the data measurement device and executes a diagnosis process to diagnose a state of the equipment based on the measurement data, The data diagnostic device has an edge application and an industrial IoT platform, This edge application has a data collection and analysis unit that calculates features of the measurement data from the data measurement device and delivers the features to an industrial IoT platform. In the above configuration, the sensor may include at least one of a vibration sensor, a temperature sensor, a pressure sensor, a strain sensor, a load sensor, and an AE (Acoustic Emission) sensor.

[0009] According to the above configuration, the condition monitoring system of the present invention includes the sensor on the edge side, the data measurement device, and the data diagnosis device, and may also include a network between each device to which an industrial IoT platform can be applied. Among them, the data diagnosis device includes an industrial IoT platform and an edge application (which may include, for example, a data collection function such as vibration data, an analysis processing function, a diagnosis function, etc.). The data collection and analysis unit included in the edge application is incorporated into the processing between the data measurement device and the industrial IoT platform, and even if data with a high sampling frequency such as vibration acceleration data is input, it calculates the feature amount required for diagnosis and delivers the feature amount to the industrial IoT platform. This makes it possible to significantly reduce the amount of data collected and delivered by the industrial IoT platform, and it is possible to adjust it on the edge side and ensure real-time processing.

[0010] In the above configuration, the edge application may include the data collection and analysis unit, a data diagnosis unit, a management and control unit, and a data display unit. Also, in the above configuration, the data collection and analysis unit of the edge application may calculate the feature amount according to the type of the sensor. This makes it possible to adjust parameters, etc. on the edge side.

[0011] In the above configuration, the data collection and analysis unit of the edge application may calculate a single scalar quantity for each measurement for each sensor. This makes it possible to adjust the amount of data delivered to the industrial IoT platform when different types of sensors are installed, regardless of the sampling frequency of each sensor.

[0012] In the above configuration, when the type of the sensor is a vibration sensor, the data collection and analysis unit of the edge application may calculate at least one of an effective value, an overall value, a peak value, a crest factor, a kurtosis, and a skewness, thereby enabling easy and accurate diagnosis.

[0013] In the above configuration, the data collection and analysis unit of the edge application may calculate at least one of an effective value, a peak value, a cumulative peak number, and an energy equivalent value when the type of the sensor is an AE sensor. This allows for easy and highly accurate diagnosis when an AE sensor is used. Effect of the Invention

[0014] The status monitoring system according to the present invention can be adjusted on the edge side, making it possible to ensure real-time processing. [Brief description of the drawings]

[0015] [Figure 1] 1 is a block diagram showing a configuration of a status monitoring system according to an embodiment of the present invention; [Diagram 2] FIG. 4 is a flow chart illustrating the operation of the condition monitoring system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are designated by the same reference numerals and detailed description thereof will be omitted.

[0017] <Embodiment 1> FIG. 1 shows a configuration diagram illustrating an overview of a state monitoring system according to a first embodiment. The state monitoring system according to this embodiment is a system for monitoring the state of equipment, and includes an industrial IoT platform and an edge application that operates on the industrial IoT platform, and can be used for this edge application. When performing analysis (e.g., vibration analysis) on the industrial IoT platform, this edge application collects, analyzes, and outputs features of vibration acceleration data and the like before inputting the data to the industrial IoT platform. In this embodiment, the edge application includes a data collection and analysis unit that performs output of the features, and a data diagnosis unit that performs diagnosis based on data delivered from the industrial IoT platform, as described later, and includes a management and control unit that manages and controls these processes to enable real-time diagnosis (diagnosis with high responsiveness) such as bearing abnormality detection when vibration acceleration data with a high sampling frequency (e.g., tens of thousands of Hz or more) is used as input data.

[0018] With this configuration, the condition monitoring system of this embodiment can be adjusted simply by customizing the edge application without modifying the industrial IoT platform, and when vibration acceleration data with a high sampling frequency is input, it is possible to perform responsive real-time vibration analysis. In addition, by calculating the feature amount required for diagnosis by the edge application on the edge side, it is possible to ensure real-time processing by eliminating unnecessary processing and unnecessary file transfer.

[0019] A condition monitoring system 100 shown in FIG. 1 includes a sensor 10, a data measuring device 20, and a data diagnostic device (hereinafter also simply referred to as a diagnostic device) DA.

[0020] The sensor 10 includes various sensors such as a vibration sensor attached to equipment or the like. The sensors included in the sensor 10 are not limited to vibration sensors, and may be temperature sensors, pressure sensors, strain sensors, load sensors, AE (Acoustic Emission) sensors, or the like. The sensor 10 is connected to a data measuring device 20. If the sensor 10 is a vibration sensor, it transmits an analog signal to the data measuring device 20. If the sensor 10 is a sensor that outputs digital values, such as a load sensor or an angle sensor, it transmits a digital signal to the data measuring device 20. In the following, the sensor 10 will be described as including a vibration sensor and performing abnormality detection (vibration detection, etc.) of a bearing that rotationally supports a shaft of a rotating machine in industrial equipment.

[0021] The data measuring device 20 receives a detection signal (analog signal or digital signal) from the sensor 10. The data measuring device 20 is, for example, a data logger or a PLC (Programmable Logic Controller) installed at a production site. The data measuring device 20 acquires measurement data from the detection signal of the sensor according to predetermined measurement conditions that are individually set. The measurement conditions are, for example, the measurement interval, measurement time, and sampling frequency of the measurement data.

[0022] For example, the measurement interval corresponds to the transmission interval of measurement data from the data measuring device 20 to the diagnostic device DA (specifically, for example, the data collection and analysis unit 43 of the edge application 40 described later). In the diagnostic device DA, for example, an analysis process is performed on a group of measurement data for each measurement interval, and a diagnosis process is performed using the results of the analysis process. The measurement time is the time during which the measurement is actually performed within the measurement interval in the data measuring device 20, and there are cases where the measurement interval = the measurement time. The sampling frequency is the frequency at which the detection signal (analog signal) from the sensor 10 is sampled. In the case of a sensor that outputs a digital signal, it is the data output rate.

[0023] The data diagnostic device DA includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), etc. (none of which are shown in the figure). The CPU loads a program stored in the ROM into the RAM etc. and executes it. The program stored in the ROM is a program in which the processing procedures of the data diagnostic device DA are written. The data diagnostic device DA is operated on a management system of, for example, a production site or the like.

[0024] The data diagnostic device DA includes the above-mentioned industrial IoT platform 60 and an edge application 40 that operates in cooperation with the industrial IoT platform 60. The industrial IoT platform 60 is software that is generally installed in an industrial computer included in a management system such as a production site. This platform refers to a preset monitoring directory in which data of detection signals from the sensor 10 is stored, and collects data (for example, feature data of an edge application described later). This platform processes the collected data according to preset processing conditions. This platform distributes the processed data to a data diagnostic unit of the edge application 40 described later according to preset distribution conditions.

[0025] The edge application 40 is software that is generally installed on an industrial computer, similar to the industrial IoT platform 60. However, the edge application 40 is separate and independent software from the industrial IoT platform 60, and shares data with the industrial IoT platform 60 by transferring files.

[0026] 1 further illustrates in detail the configuration of the data diagnosis device DA included in the condition monitoring system 100. According to FIG. 1, the edge application 40 of the data diagnosis device DA includes a data collection and analysis unit (front end) 43, a data display unit 45, a management and control unit 47, and a data diagnosis unit (back end) 49. The data collection and analysis unit 43 may be divided into a data collection unit and a data analysis unit. The data display unit 45 and the data diagnosis unit 49 may be installed in another computer that exists on an external system different from the industrial IoT platform 60 or on a network such as a WAN, a LAN, or the Internet.

[0027] The data collection and analysis unit 43 (for example, the data collection unit) receives measurement data acquired in the data measurement device 20 according to the measurement conditions from the data measurement device 20. The data collection and analysis unit 43 (for example, the data analysis unit) also performs analysis processing on the measurement data. The analysis processing includes, for example, a process of calculating the effective value (RMS (Root Mean Square)) of the measurement data, a process of performing frequency analysis by performing a fast Fourier transform (FFT (Fast Fourier Transform)) on the measurement data, a process of calculating an overall value, etc. Note that before performing the FFT, the input data may be subjected to a low-pass filter and a high-pass filter.

[0028] Furthermore, the data collection and analysis unit 43 (for example, the above-mentioned data analysis unit) outputs and stores the result of the analysis process (hereinafter, sometimes referred to as "feature amount") for the detection signal (measurement data) in a specified area (such as the above-mentioned monitoring directory) monitored by the industrial IoT platform 60. Preferably, when the measurement data is vibration acceleration data, the vibration acceleration data obtained by one measurement is calculated as a single scalar amount such as an effective value, an overall value, a peak value, a crest factor, a kurtosis, or a skewness as the feature amount. Also, when the measurement data is data from an AE (Acoustic Emission) sensor, for example, a single scalar amount such as an effective value, a peak value, a peak detection frequency, a cumulative number of peaks, or an energy equivalent value is similarly calculated as the feature amount. Note that the feature amount depends on the type of sensor. In this embodiment, the average value of the most recent N pieces of data may be used as the evaluation value for this feature amount. In this way, when the evaluation value with a significantly reduced amount of data is stored in the above-mentioned monitoring directory on the industrial IoT platform 60 side, the amount of input data reduction of the industrial IoT platform 60 is the reciprocal of "data sampling frequency [Hz] x measurement interval [sec]".

[0029] The industrial IoT platform 60 includes a data collection unit, a data processing unit, and a data distribution unit 61. As described above, the data collection unit of the industrial IoT platform 60 refers to the above-mentioned preset monitoring directory and collects feature data of the edge application. The data processing unit processes the feature data of the collected data according to preset processing conditions. The data distribution unit distributes the collected feature data to the data diagnosis unit of the edge application 40 only when the collected feature data has a value within a preset range. Preferably, the processing conditions are such that only the collected feature data of the edge application 40 within a valid range is distributed to the data diagnosis unit 49 of the edge application 40. The set range is set in advance in the industrial IoT platform 60.

[0030] The data diagnostic unit 49 is connected to the industrial IoT platform 60, and performs diagnosis using the processed feature amount delivered from the industrial IoT platform 60 as input data. The edge application 40 has the functions of a "learning mode" and a "diagnosis mode" as operation modes. When the operation mode is the "learning mode", the feature amount is stored a specified number of times from the start of measurement in order to obtain a reference value used to calculate the threshold value referred to in the above diagnosis. When the specified number of times is completed, the statistical value of the feature amount (for example, average value, etc.) is set as a reference value and stored. From the feature amount acquired in the learning mode, a value obtained by multiplying a value (for example, standard deviation, etc.) that quantifies the magnitude of variation by a coefficient is added to the reference value to obtain a threshold value. In this case, for example, if three coefficients are prepared, three threshold values, threshold value 1, threshold value 2, and threshold value 3, are obtained. Note that the number of threshold values ​​is not limited to three, and may be one or more. For example, the coefficient refers to a value input and set by the management and control unit 47, and the reference value and threshold value are passed to the management and control unit 47 described later and stored.

[0031] The above diagnosis in the data diagnosis unit 49 is performed when the operation mode is the "diagnosis mode". The diagnosis is performed, for example, according to the following 1) to 3). 1) The average value of the most recent N times of processed data delivered from the industrial IoT platform 60 is set as an evaluation value. 2) This evaluation value is compared with each threshold value stored in the management and control unit 47 to determine the corresponding category (level). 3) The corresponding category and evaluation value are set as the diagnosis result. This diagnosis result is passed to the management and control unit 47, and the diagnosis result is stored in a directory specified on the industrial IoT platform 60 side. The diagnosis settings (average score, etc.) are set by the management and control unit 47. If the data diagnosis unit 49 is not arranged in the same housing as the industrial IoT platform 60, the above steps are executed via a communication network such as the Internet or a LAN or an external system.

[0032] The management and control unit 47 performs management and control within the edge application 40. It passes display data to the data display unit 45 within the edge application 40, and acquires data input at the display unit 45. The management and control unit 47 passes analysis settings input at the display unit 45 to the data collection and analysis unit 43, and also receives and stores the reference values ​​and threshold values ​​output by the data collection and analysis unit 43. The management and control unit 47 passes diagnosis settings input at the display unit 45 to the data diagnosis unit 49, and also receives diagnosis results output from the data diagnosis unit 49, and passes them to the display unit 45.

[0033] In this embodiment, the display unit 45 is an input / output user interface, and is connected to the management / control unit 47 to receive and display display data. The display unit 45 also has an input means, and passes input information to the management / control unit 47.

[0034] Next, the flow up to diagnosis in the diagnostic device DA will be described with reference to the flow diagram of Fig. 2. Note that steps S101 to S109 in the diagram are executed by the data collection and analysis unit 43, the subsequent steps S201 to S207 are executed by the industrial IoT platform 60, and the subsequent steps S301 to S329 (except S313) are executed by the data diagnosis unit 49 (S313 is executed by the management and control unit 47).

[0035] When this flow is executed (START), the data collection and analysis unit 43 reads vibration (acceleration) data DT in this embodiment from the data measurement device 20 (S101), calculates a feature amount (S103), and stores the most recent data of the feature amount (S105). Next, an evaluation value is calculated and [analyzed] (S107), and stored (S109).

[0036] After that, when the industrial IoT platform 60 reads the evaluation values ​​from the data collection and analysis unit 43 from the monitoring directory and collects a predetermined amount (S201), it determines whether the processing conditions are met (S203). If the processing conditions are not met, the process of the same flow is terminated (END), and if the processing conditions are met, the evaluation values ​​are processed (S205) and the processed evaluation values ​​(processed data) are saved (S207).

[0037] Next, the data diagnosis unit 49 reads the evaluation value (processed data) from the industrial IoT platform 60 (S301), and determines whether the current mode is the learning mode or not (S303). If it is the learning mode, the process proceeds to S305, and if it is not the learning mode (if it is the diagnosis mode), the process proceeds to S317. The current mode is appropriately set by the management and control unit 47 to either the learning mode or the diagnosis mode.

[0038] In S305, learning data is collected and stored (S307). In S309, it is determined whether a predetermined amount of learning data has been collected. If it is determined that the learning data has been collected the designated number of times, the threshold value is calculated using the coefficient and reference value of the management and control unit 47 as described above (S311). If it is not determined that the predetermined amount of learning data has been collected, the processing of this flow is terminated (END).

[0039] The calculated threshold value and the above-mentioned reference value are stored in the management / control unit 47 (S313).

[0040] Next, in S315, the thresholds (first threshold to third threshold) stored in the management / control unit 47 are read and set, and the mode is changed to the diagnosis mode. In the diagnosis mode, the above-mentioned corresponding category (level) is determined, and the diagnosis result is determined and stored as described above. In S317, for example, if the first threshold to third threshold are such that the first threshold < the second threshold < the third threshold, the above-mentioned read evaluation value (average value of the processed data, etc.) is compared with the first threshold, and if the evaluation value is smaller than the first threshold, it is determined to be level 0 (S319), and if the evaluation value is not smaller than the first threshold (not smaller than the first threshold), the process proceeds to S321.

[0041] In S321, the loaded evaluation value (processed data) is compared with a second threshold value, and if the evaluation value is smaller than the second threshold value (i.e., the evaluation value is equal to or larger than the first threshold value and smaller than the second threshold value), it is determined to be level 1 (S323), and if the evaluation value is not smaller than the second threshold value (equal to or larger than the second threshold value), the process proceeds to S325. In S325, the loaded evaluation value (processed data) is compared with a third threshold value, and if the evaluation value is smaller than the third threshold value (i.e., the evaluation value is equal to or larger than the second threshold value and smaller than the third threshold value), it is determined to be level 2 (S327), and if the evaluation value is not smaller than the third threshold value (equal to or larger than the third threshold value), it is determined to be level 3 (S329), and the flow process ends.

[0042] According to each of the above embodiments, the condition monitoring system of the present embodiment can be adjusted simply by customizing the edge application without modifying the industrial IoT platform, and enables responsive real-time vibration analysis using vibration acceleration data with a high sampling frequency (for example, tens of thousands of Hz or more) as input. In addition, by calculating the feature amount required for diagnosis by the edge application on the edge side, it is possible to ensure real-time processing by eliminating unnecessary processing and unnecessary file transfer volume.

[0043] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present invention is defined by the claims, not the above description, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0044] 10 Sensors 20 Data measurement device 40 Edge Applications 43 Data Collection and Analysis Department 45 Data display section 47 Management and Control Section 49 Data Diagnostics Department 60 Industrial IoT Platform 100 Condition Monitoring System DA Data Diagnostic Device

Claims

1. A status monitoring system for monitoring a status of equipment, comprising: A sensor attached to the facility; a data measurement device that receives a detection signal from the sensor and acquires measurement data from the detection signal in accordance with a predetermined measurement condition; a data diagnosis device that receives the measurement data from the data measurement device and executes a diagnosis process to diagnose a state of the equipment based on the measurement data, The data diagnostic device has an edge application and an industrial IoT platform, The edge application includes a data collection and analysis unit that calculates a feature of the measurement data from the data measurement device and delivers the feature to the industrial IoT platform; The edge application runs on the industrial IoT platform; The industrial IoT platform collects the feature data, processes the collected data, and distributes the processed data, Further, a data diagnosis unit that performs diagnosis based on the processed data distributed from the industrial IoT platform. Condition monitoring system.

2. The sensor includes at least one of a vibration sensor, a temperature sensor, a pressure sensor, a strain sensor, a load sensor, and an AE sensor. The condition monitoring system according to claim 1 .

3. The edge application has the data collection and analysis unit, a data diagnosis unit, a management and control unit, and a data display unit.

3. A condition monitoring system according to claim 1 or 2.

4. The data collection and analysis unit of the edge application calculates the feature amount according to the type of the sensor. A condition monitoring system according to any one of claims 1 to 3.

5. The data collection and analysis unit of the edge application calculates a single scalar quantity for each sensor and each measurement. A condition monitoring system according to any one of claims 1 to 4.

6. When the type of the sensor is a vibration sensor, the data collection and analysis unit of the edge application calculates at least one of an effective value, an overall value, a peak value, a crest factor, a kurtosis, and a skewness. A condition monitoring system according to any one of claims 1 to 5.

7. When the type of the sensor is an AE sensor, the data collection and analysis unit of the edge application calculates at least one of an effective value, a peak value, a cumulative number of peaks, and an energy equivalent value. A condition monitoring system according to any one of claims 2 to 5.

Images