KR20240039877A - Equipment failure prediction diagnostic device and method therefor - Google Patents

Abstract

The present invention relates to a device for predicting and diagnosing equipment failure and a method thereof, and includes a facility data collection unit equipped with at least one individual board corresponding to the type of facility to be diagnosed and having a plurality of input channels through which data of the facility to be diagnosed is input. ; and a facility data processing unit including a common board that supplies power to the facility data collection unit and performs real-time data determination on data on diagnostic target equipment input from at least one of a plurality of input channels using a diagnostic model corresponding to the channel; Including, the equipment data collection unit and the equipment data processing unit are connected through a serial communication interface, and an integrated equipment failure prediction and diagnosis device is introduced.

Description

Equipment failure prediction diagnosis device and method {EQUIPMENT FAILURE PREDICTION DIAGNOSTIC DEVICE AND METHOD THEREFOR}

The present invention relates to a facility failure prediction and diagnosis device and method for miniaturizing the device by integrating a facility data collection unit and a facility data processing unit and ensuring scalability and maintainability of line application.

In the equipment field, Prognostics and Health Management (PHM) is being applied to diagnose the current condition of the equipment and predict failure. Predictive failure management not only reduces costs, but also allows the system to operate stably, and is rapidly growing to realize cost-effective, downtime-free maintenance.

In addition, unlike cloud computing's data center that manages data centrally in a physically distant location, computing is supported at the edge of the network close to devices (things), allowing each device to An edge system is used that allows data to be analyzed and utilized.

Meanwhile, in a typical equipment failure prediction and diagnosis device, the systems for performing edge computing are all separate, and in order to perform the edge computing function, there are separate devices for collecting data from the equipment being diagnosed and a device for processing the collected data. , The connection between two devices generally uses wired communication such as USB (Universal Serial Bus) or Ethernet. However, the USB connection is slow, and the Ethernet method also has problems that may result in data loss. Ultimately, the connection method between individual devices has problems with lower maintainability and scalability than when configured as a single system.

Accordingly, a method is needed to miniaturize a system for performing edge computing into a single device and to prevent data loss and noise while transmitting data at high speed.

The matters described as background technology above are only for the purpose of improving understanding of the background of the present invention, and should not be taken as acknowledgment that they correspond to prior art already known to those skilled in the art.

KR 10-2019-0091868 A

The present invention reduces costs by miniaturizing the device through a structure in which the equipment data collection unit and the equipment data processing unit are integrated, enables predictive maintenance by diagnosing the condition of the equipment in real time, and provides expandability and maintainability of line applications. The purpose is to provide a facility failure prediction diagnostic device and method to ensure.

As a means to solve the above technical problem, the present invention includes: a facility data collection unit equipped with at least one individual board corresponding to the type of facility to be diagnosed and having a plurality of input channels through which data of the facility to be diagnosed is input; and a facility data processing unit including a common board that supplies power to the facility data collection unit and performs real-time data determination on data on diagnostic target equipment input from at least one of a plurality of input channels using a diagnostic model corresponding to the channel; Including, the equipment data collection unit and the equipment data processing unit are connected through a serial communication interface and may include a facility failure prediction diagnosis device integrated into the device.

For example, the equipment data processing unit transmits the collected data to an external server, and the external server diagnoses the condition of the equipment based on the processed data, predicts equipment failure or lifespan, and calculates the replacement time of the equipment. It may include a Prognostics and Health Management (PHM) server that determines data in a manner that does this.

For example, the external server stores a plurality of diagnostic models corresponding to at least one of the equipment status diagnosis and life prediction algorithms, and the equipment data processing unit stores the external diagnostic model corresponding to the type of equipment recognized on the common board. It can be provided from the server.

For example, the equipment data collection unit may convert data input through at least one of a plurality of input channels into a digital signal, parse the converted data, and store it in a buffer.

For example, the buffer may include: a first buffer storing parsed data parsed at a first sampling rate; and a second buffer storing parsed data down-sampled at a second sampling rate lower than the first sampling rate.

For example, the equipment data processing unit may transmit data in the first buffer to an external server and store data in the second buffer in a database.

For example, data input from the diagnostic target equipment may include at least one of speed data, vibration data, current data, and temperature data from the equipment.

For example, at least one individual board may be provided to be detachable.

For example, if the accumulated data stored in the buffer is larger than a preset size, the equipment data collection unit may insert the data into a data queue to be thread-processed by the equipment data processing unit.

For example, the facility data processing unit may determine whether the size of the data queue is 0 or more, and extract data from the data queue if the size of the data queue is 0 or more.

For example, serial communication may include SPI communication.

As a method for solving the above technical problem, the present invention provides a facility failure prediction system in which a facility data collection unit and a facility data processing unit equipped with at least one individual board having a plurality of input channels are connected through a serial communication interface, and are integrated into an integrated facility. A method of predicting and diagnosing equipment failure for controlling a diagnostic device, comprising: inputting data of equipment to be diagnosed; and

It may include performing real-time data determination on data of a diagnostic target equipment input from at least one of a plurality of input channels using a diagnostic model corresponding to the channel.

For example, it may further include determining the data by performing a diagnosis of the condition of the equipment based on the processed data, predicting failure or lifespan of the equipment, and calculating a replacement time for the equipment.

For example, the method may further include converting data input through at least one of a plurality of input channels into a digital signal, parsing the converted data, and storing the converted data in a buffer.

For example, if the accumulated data stored in the buffer is larger than the preset size, the step of inserting the data into a data queue to be thread-processed in the equipment data processing unit may be further included.

For example, the method may further include determining whether the size of the data queue is 0 or more, and extracting data from the data queue if the size of the data queue is 0 or more.

According to the equipment failure prediction and diagnosis device and method of the present invention, the equipment data collection unit and the equipment data processing unit are integrated into a structure that reduces costs by miniaturizing the device and enables predictive maintenance by diagnosing the condition of the equipment in real time. This makes it possible to secure the scalability and maintainability of line application. In addition, the integrated structure reduces design costs compared to external connecting structures, and allows it to be configured adjacent to equipment through miniaturization.

The effects that can be obtained from the present invention are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below. will be.

1 is an external perspective view showing an equipment failure prediction and diagnosis device according to an embodiment of the present invention.
Figure 2 is a diagram showing the configuration of the equipment data collection unit and equipment data processing unit that constitute the equipment failure prediction and diagnosis device.
Figure 3 is a table showing the reliability verification results of data input through an input channel.
Figure 4 is a functional block diagram of the entire system.
Figure 5 is a diagram showing a process in which the facility data collection unit parses data converted into digital signals and stores them in the first buffer and the second buffer.
Figure 6 is a flowchart showing a method of collecting and storing data in a facility failure prediction and diagnosis device.

Specific structural and functional descriptions of the embodiments of the present invention disclosed in the present specification or application are merely illustrative for the purpose of explaining the embodiments according to the present invention, and the embodiments according to the present invention may be implemented in various forms. and should not be construed as limited to the embodiments described in this specification or application.

Since the embodiments according to the present invention can make various changes and have various forms, specific embodiments will be illustrated in the drawings and described in detail in the specification or application. However, this is not intended to limit the embodiments according to the concept of the present invention to a specific disclosed form, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and unless clearly defined in this specification, should not be interpreted as having an ideal or excessively formal meaning. .

Hereinafter, the present invention will be described in detail by explaining preferred embodiments of the present invention with reference to the accompanying drawings. The same reference numerals in each drawing indicate the same member.

Before explaining the equipment failure prediction and diagnosis method according to the embodiments of the present invention, the configuration of the equipment failure prediction and diagnosis device 1000 applicable to the embodiments will first be described.

1 and 2 are respectively a perspective view of the exterior of the equipment failure prediction and diagnosis device 1000 according to an embodiment of the present invention, and the equipment data collection unit 400 and the equipment data processing unit ( The functional diagram of 500) is shown.

Figure 1 is an external perspective view of a facility failure prediction and diagnosis device 1000. The equipment failure prediction and diagnosis device 1000 is also called a CMS (Condition Monitoring System) module. The facility failure prediction and diagnosis device 1000 may include a housing case 100, a cooling fin 200, a facility data collection unit 400, a facility data processing unit 500, and an external connection port. First, the housing case 100 surrounds the outer surface of the equipment failure prediction and diagnosis device 1000 and is made of aluminum, making it possible to lighten the equipment failure prediction and diagnosis device 1000. In addition, since the inside and outside of the equipment failure prediction and diagnosis device 1000 can be connected through a bolting structure, it is easy to attach and detach the individual boards 410 and 420, which will be described later. Additionally, a plurality of cooling fins 200 are provided on the outside of the housing case 100 to discharge heat generated from the CPU 513A and GPU 513B.

In addition, the equipment failure prediction and diagnosis device 1000 can be manufactured in a monolithic structure where one service is one large structure compared to a system configuration capable of edge computing. The shape of the equipment failure prediction and diagnosis device 1000 is low in height and is formed so that the width is longer than the front-to-back length, thereby lowering the center of gravity and thus ensuring stability.

Meanwhile, the equipment failure prediction and diagnosis device 1000 may be provided with an external connection port 300 for connection to the outside. Types of external connection ports 300 include BNC (Bayonet Neill-Concelman) connector 411 8ch, which can collect data by connecting a sensor cable, and VCC (Voltage of Common Collector)/GND (Ground) ports. , which may be provided on the front of the equipment failure prediction and diagnosis device 1000. Additionally, a power connector, an HDMI (High-Definition Multimedia Interface) connection port, a USB (Universal Serial Bus) connection port, and a LAN (Local Area Network) connection port may be provided on the rear of the equipment failure prediction and diagnosis device 1000. Additionally, an LED light capable of displaying the status of each input channel and an LED light capable of displaying the power supply status may be provided on the front of the equipment failure prediction and diagnosis device 1000. The external connection port 300 described so far is preferably provided on the equipment data collection unit 400 side as shown in FIG. 1 in terms of data input and output.

Referring to FIG. 2, a functional block diagram of the equipment data collection unit 400 and the equipment data processing unit 500, which constitute the equipment failure prediction and diagnosis device 1000, can be seen. The equipment data collection unit 400 may be equipped with at least one individual board 410 or 420 that corresponds to the type of equipment being diagnosed and has a plurality of input channels. At this time, each individual board (410, 420) is configured to be detachable, so that various data can be collected according to customization of the individual boards (410, 420).

In addition, the package of the module that directly collects and controls data can be designed as each individual board (410, 420), and the circuit is configured with a plurality of channels, for example, 8 channels, so that data from the equipment to be diagnosed can be input. Data can be input from the diagnostic target equipment through at least one of a plurality of channels provided in the BNC connector 411 provided in the individual board 410, and the acquired data is converted to analog through an ADC (Analog to Digital Converter, 412). A signal can be converted into a digital signal.

Additionally, the data converted into a digital signal is transmitted to the attenuator 413 and attenuated according to the device management voltage, and the attenuated data can be collected in an MCU (Micro Controller Unit, 414). The collected data may be parsed and stored in the buffer of the controller (MCU, 414). Here, data parsing refers to the task of assembling data and refining it into desired data.

Meanwhile, diagnostic target equipment data input to a plurality of input channels may include at least one of speed data, vibration data, current data, and temperature data obtained from sensors provided for each equipment. The types of data described above are illustrative and are not necessarily limited thereto, and are not limited to the types of data that can be collected from equipment.

This explains a system for ensuring the reliability of diagnostic target equipment data input through an input channel. The system for reliability confirmation can be configured in a single ended GND form by connecting VCC and GND to the BNC connector 411 and terminal block of the equipment failure prediction diagnosis device 1000 through a power supply. When applying a voltage from 1V to 24V in 1V increments with a power supply, the value output from the ADC (412) can be stored, and the average value of each 8ch of data received 10 times can be calculated.

The results of verifying reliability through this configuration are shown in Figure 3.

During the verification process, the bit resolution of the facility failure prediction diagnosis device was set to 24 bits so that the 2 ²³ value was expressed as the ADC (412) value when 24 V was applied.

Specifically, the base voltage was 0.475V, and in order to check whether the high-speed sampling rate could be reproduced, the sampling rate was divided into 8kS/s and 16kS/s, and data collected for 10 seconds each was checked. First, the average/standard deviation of the output values of the ADC (412), which is expressed as a plot when there is only the equipment failure prediction and diagnosis device (1000), was checked, and the ADC (412) when the vibration sensor was attached to the equipment failure prediction and diagnosis device (1000) was checked. The average/standard deviation of the output values were checked and compared.

As shown in FIG. 3, the floating value comes in almost constantly, confirming that the noise of the equipment failure prediction and diagnosis device 1000 itself is at a negligible level, and that there is no significant difference in the sampling rate change. In the form of the facility failure prediction diagnosis device (1000) combined with a vibration sensor, the ratio of the average to the standard deviation was 0.19% at 8kS/s, and the ratio of the average to the standard deviation was 0.14% at 16kS/s, which confirmed that there was almost no deviation. You can. This confirms that reliability has been secured for the data collected by connecting the vibration sensor.

As one of the individual boards 410 and 420, an Ethernet module 420 may be provided. The Ethernet module's controller (MCU, 414) can also receive power through POE (511B) and exchange data with Giga Ethernet (512H). Additionally, the controller (MCU, 414) can be connected to a plurality of Ethernet connectors via an Ethernet hub chipset. External devices can be connected to each Ethernet connector through an Ethernet cable.

Returning to FIG. 2, the equipment data processing unit 500 includes a common board 510 that performs real-time data determination on the data of the diagnostic target equipment input from at least one of a plurality of input channels using a diagnostic model corresponding to the channel. It can be provided. Here, the common board 510 may be composed of a power stage 511, an interface stage 512, and a processing device stage 513.

Specifically, by connecting the RJ-45 standard connector port (511A) to the power terminal 511, the power generated through POE (Power Of Ethernet, 511B) is transmitted to the individual boards 410 and 420 of the equipment data collection unit 400. You can move to . In addition, the power stage 511 supplies power to the equipment data collection unit 400 and can also supply power to the processing device stage 513 in the form of 1.1V, 1.5V, 1.8V, and 3.3V. At this time, the power supply form is not limited to the above-described form. Additionally, the interface stage 512 may be connected to the processing unit 513 through LPDDR4 (512A) and DDR 32bit interface. In addition, the interface stage 512 is composed of a flash memory and controller integrated package and can be connected to the processing unit 513 through an eMMC (Embedded Multi-Media Controller, 512B) and MMC interface.

Status LED (512C) for power or channel data collection status alarm can be connected through GPIO (General-Purpose Input/Output), and debugger/downloader (512D) for firmware can be connected to USB and UART (Universal Asynchronous Receiver/Output). It can be connected to the processing unit 513 through a transmitter.

In addition, the Audio AMP (512E) for providing a defective alarm after data determination can be connected to the processing unit 513 in the I2S (Integrated Interchip Sound) method. The connection between the controller (MCU, 414) and the CPU (513A) in the facility data collection unit 400 can be accomplished through SPI communication (512F), which will be described later, and can be interfaced using RGMII (Reduced Gigabit Media Independent Interface). Meanwhile, the wireless method supports Bluetooth and Wifi (512G), and Giga Ethernet (512H) can be configured for the Ethernet module.

In addition, the CPU (513A) and GPU (513B) constitute the processing unit 513, and the CPU (513A) and GPU (513B) command data collection and preprocessing in real time and are equipped with an AI driving algorithm to determine real-time data. It is possible to perform edge computing through .

Additionally, the equipment data processing unit 500 may be connected to the equipment data collection unit 400 through a serial communication interface. In order to transmit data at high speed and process it without data loss or noise, it is desirable to use a serial communication interface, which is short-distance communication. As an example, serial communication includes SPI (Serial Peripheral Interface) communication 512F. A method such as Ethernet may be considered for connection between the facility data processing unit 500 and the facility data collection unit 400, but the Ethernet method is not preferable because there is a risk of data loss during data transmission. However, SPI communication (512F) has the advantage of being able to transmit data in real time at high speed when transmitting data over a short distance and there is no data loss. Therefore, using SPI communication (512F), high-speed data processing is possible, and data can be judged in real time in the facility data processing unit 500 through an AI driving algorithm. Therefore, the SPI communication (512F) method can be said to be a communication method suitable for the facility failure prediction and diagnosis device 1000, which is configured as a single device by combining the facility data collection unit 400 and the facility data processing unit 500.

Figure 4 is a functional block diagram 600 of the entire system.

Referring to FIG. 4, the functional block diagram 600 of the CPU 513A may include a Linux-based operating system 610, a communication protocol 620, and an IoT framework 630, which is embedded SW. The communication protocol 620 includes HTTP (HyperText Transfer Protocol, 621), a protocol for hypertext transmission, TCP/IP (Transmission Control Protocol/Internet Protocol, 622), an Internet information transmission protocol, and MQTT (Message Queuing Telemetry), a message queuing protocol. Transport, 623) exists, and this is an example and is not necessarily limited to this.

Meanwhile, the IoT framework 630 includes a result monitoring part 631, a data collection/transmission part 632, a device discovery part 633, a resource management part 634, a data storage part 635, and a management part 636. ) can be composed of.

First, the device search part 633 can search for external devices connected to the facility failure prediction diagnosis device 1000. The resource management part 634 can manage resources stored in the CPU 513A, and the data storage part 635 can store data collected from equipment. Additionally, the management part 636 provides information on which diagnostic model to install. In addition, it is possible to perform a function of managing resources by providing a place to store some data in the device and checking the occupancy rate of the CPU (513A) and GPU (513B) in the device. In the data collection/transmission part (183B), data collected from the equipment end through the BNC connector 411 is received through each channel, passes through the internal buffer of the controller (MCU, 414) of the equipment data collection unit 400, and is transmitted to the CPU ( 513A) Can be collected and transmitted in real time through an internal buffer. In the diagnostic model loading and determination part 650, the diagnostic model corresponding to the type of equipment recognized on the common board 510 provided from the external server 2000 is mounted on the virtual machine of the AI framework 640, and It can perform the function of providing different diagnostic models. This makes it possible to make real-time data judgments on the transmitted data, and to perform edge computing functions while visualizing the results monitored in the result monitoring part (183D) along with the data.

In addition, the external server 2000 may include a Prognostics and Health Management (PHM) server that can diagnose the condition of the equipment based on the processed data, predict the failure or life of the equipment, and calculate the replacement time of the equipment. You can. The external server 2000 may perform predictive maintenance of equipment based on data transmitted from the equipment data processing unit 500. At this time, the external server 2000 may store a plurality of diagnostic models corresponding to at least one of the equipment status diagnosis and life prediction algorithms and provide them to the equipment data processing unit 500. Accordingly, the equipment data processing unit 500 is provided with a diagnostic model corresponding to the type of equipment recognized by the common board 510 among the stored diagnostic models and can determine the data in real time.

Figure 5 is a diagram showing a process in which the equipment data collection unit 400 parses data converted into digital signals and stores them in a first buffer and a second buffer.

Referring to the left drawing of FIG. 5, you can see the process in which the data of the equipment data collection unit 400 is divided into the first buffer and the second buffer and stored. The facility failure prediction diagnosis device 1000 capable of driving a high-speed sampling rate can receive and parse up to 16,000 samples per second from the BNC connector 411. In order to receive all parsed data without missing anything, it is necessary to store it primarily with an intermediate buffer and control data transmission through the message queue method. At this time, the data may be divided into two buffers and stored to be transmitted separately to the data transmission thread and the data monitoring thread. To be transmitted to the data transmission thread, parsed data parsed at a first sampling rate, which is a high-speed sampling rate, may be stored in a first buffer. Additionally, in order to be transmitted to the data monitoring thread, the data may be down-sampled to a second sampling rate lower than the first sampling rate and stored in a second buffer so that there is no delay in determining real-time data.

At this time, in order to utilize the shared resources of the CPU (513A) called threads, a method of accessing data by synchronizing with time control can be used. Mutex is used as a synchronization mechanism provided by Pthread, and the amount of data to be inserted into the data queue can be controlled to stop or transfer through the mutex_lock command. The data queue is implemented as temporary storage through a ring buffer, and data can be transmitted using First In First Out (FIFO).

Referring to the left drawing of FIG. 5, the equipment data processing unit 500 transmits the data in the first buffer to the external server 2000 and confirms the process of transmitting the data in the second buffer to the database so that it is stored in the database. You can. More specifically, the data temporarily stored in the first buffer and the second buffer are first-in-first-out in the data queue using the queue method in order to be transmitted to each thread, and the mutex_lock command is used to control access to the shared resource called the thread. This makes it possible to control data blocking and data transmission. Each thread can also be controlled with the mutex_lock command to automatically adjust the resources required for data transmission and extraction to prevent data loss. Accordingly, the data extracted from the thread can be managed by being transmitted to an external server (2000) or a database (DB, 3000) through the TCP/IP protocol.

Based on the configuration of the equipment failure prediction and diagnosis device 1000 described above, a facility failure prediction and diagnosis method (S700) according to an embodiment will be described with reference to FIG. 6.

FIG. 6 shows equipment failure prediction and diagnosis in which the equipment data collection unit 400 and the equipment data processing unit 500, which are equipped with at least one individual board 410 and 420 having a plurality of input channels according to FIG. 1, are connected through a serial communication interface. This is a flowchart (S700) showing a method of collecting and storing data in the device 1000.

First, the data collected from the BNC connector 411 is converted into a digital signal through the ADC 412, temporarily stored in the controller (MCU, 414), and then transmitted to the CPU (513A) of the platform module through internal SPI communication (512F). Can be transmitted (S701). Thereafter, a large amount of data parsed at high speed (16 kS/s) in the CPU 513A is parsed and stored in a first buffer storing the parsed data parsed at a first sampling rate and a second sampling rate lower than the first sampling rate. It is stored and accumulated in a second buffer that stores the down-sampled parsed data (S702). Data can be stored divided into two buffers to be split and transmitted to the data transmission thread and data monitoring thread.

The equipment data collection unit 400 may determine whether the accumulated data stored in the first buffer is larger than the preset size (N) to check whether the level can be stored in the buffer (S703). This is to check the amount of data accumulated in the buffer and check whether it can be stored in the buffer. If it is determined that the accumulated data stored in the first buffer is not larger than the preset size (N), additional data is received (N in S703). However, if it is determined that the accumulated data stored in the first buffer is larger than the preset size (N) (Y in S703), the data can be inserted into the data queue to be thread-processed in the facility data processing unit 500 (S704) ). When all data is accumulated in the data queue, the first buffer is emptied to receive the next data (S705). The process described so far can be performed by the equipment data collection unit 400 (S701-S705).

After the buffer is emptied, the process by which the equipment data processing unit 500 transmits data to the external server 2000 and the database (DB, 3000) will be described.

The facility data processing unit 500 may check the current size of the data queue and prepare for data collection in the CPU 513A (S710). It is possible to determine whether to extract data from the data queue by determining whether the size of the data queue is 0 or more (S711). If the size of the data queue is less than 0, continue to check the size of the data queue (N in S711). However, if the size of the data queue is 0 or more (Y in S711), data can be extracted from the data queue (S712). At this time, if there is data in the data queue, the data can be extracted by performing time synchronization access control from the data queue by simultaneously using the mutex_lock command.

Afterwards, the extracted data is sent to the server through the TCP/IP protocol (S713). The facility data processing unit 500 may receive the data reception result from the external server 2000 and prepare again to extract new data from the data queue (S714). According to the above data processing process, the equipment data processing unit 500 can smoothly perform real-time data judgment on the data of the diagnostic target equipment using a diagnostic model corresponding to the corresponding channel.

Ultimately, costs are reduced by miniaturizing the device through a structure in which the facility data collection unit and the facility data processing unit are integrated into one unit, and the condition of the facility is diagnosed in real time through an AI algorithm, enabling predictive maintenance of the facility and line application. Scalability and maintainability can be secured.

Meanwhile, the above-described present invention can be implemented as computer-readable code on a program-recorded medium. Computer-readable media includes all types of recording devices that store data that can be read by a computer system. Examples of computer-readable media include HDD (Hard Disk Drive), SSD (Solid State Disk), SDD (Silicon Disk Drive), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. There is. Accordingly, the above detailed description should not be construed as restrictive in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

100: housing case 200: cooling fin
300: External connection port 400: Equipment data collection unit
500: Equipment data processing unit 410, 420: Individual board
411: BNC connector 412: ADC
413: Attenuator 414: Controller (MCU)
510: common board 511: power stage
511A: Connector Port 511B: POE
512: Interface 512A: LPDDR4
512B: eMMC 512C: Status LED
512D: Debugger/Downloader 512E: Audio AMP
512F: SPI communication 512G: Bluetooth, Wifi
512H: Giga Ethernet 513: Processing device stage
513A: CPU 513B: GPU
610: Operating system 620: Communication protocol
621: HTTP 622: TCP/IP
623: MQTT 630: IOT framework
631: Result monitoring part 632: Data collection/transmission part
633: Device discovery part 634: Resource management part
635: Data storage part 636: Management part
640: AI framework 650: Diagnostic model loading and judgment part
1000: Equipment failure prediction and diagnosis device 2000: External server
3000: Database (DB)

Claims (16)

A facility data collection unit equipped with at least one individual board corresponding to the type of facility to be diagnosed and having a plurality of input channels through which data of the facility to be diagnosed is input; and
A facility data processing unit that supplies power to the facility data collection unit and includes a common board that performs real-time data determination on data on diagnostic target equipment input from at least one of a plurality of input channels using a diagnostic model corresponding to the channel; Including,
The equipment data collection department and the equipment data processing department,
It is connected through a serial communication interface and is an integrated equipment failure prediction diagnostic device. In claim 1,
The facility data processing department transmits the collected data to an external server,
The external server is
Characterized by including a PHM (Prognostics and Health Management) server that determines the data by performing a diagnosis of the condition of the equipment based on the processed data, predicting the failure or life of the equipment, and calculating the replacement time of the equipment. Equipment failure prediction diagnostic device. In claim 2,
The external server is
A plurality of diagnostic models corresponding to at least one of the equipment condition diagnosis and life prediction algorithms are stored,
Equipment data processing department,
A facility failure prediction diagnostic device characterized in that a diagnostic model corresponding to the type of facility recognized on a common board is provided from an external server. In claim 2,
Equipment data collection department,
An equipment failure prediction and diagnosis device that converts data input through at least one of a plurality of input channels into a digital signal, parses the converted data, and stores it in a buffer. In claim 4,
The buffer is,
a first buffer storing parsed data parsed at a first sampling rate; and
A facility failure prediction and diagnosis device comprising a second buffer that stores parsed data down-sampled at a second sampling rate lower than the first sampling rate. In claim 5,
Equipment data processing department,
A facility failure prediction and diagnosis device characterized in that the data in the first buffer is transmitted to an external server, and the data in the second buffer is stored in a database. In claim 1,
The data input from the equipment subject to diagnosis is:
A facility failure prediction and diagnosis device comprising at least one of speed data, vibration data, current data, and temperature data from the facility. In claim 1,
At least one individual board:
A facility failure prediction diagnostic device characterized in that it is provided in a detachable manner. In claim 4,
Equipment data collection department,
If the accumulated data stored in the buffer is larger than the preset size, the equipment failure prediction and diagnosis device is characterized in that the data is inserted into a data queue to be threaded in the equipment data processing unit. In claim 9,
Equipment data processing department,
A facility failure prediction diagnosis device that determines whether the size of the data queue is 0 or more, and extracts data from the data queue if the size of the data queue is 0 or more. In claim 1,
An equipment failure prediction and diagnosis device wherein serial communication includes SPI communication. In the equipment failure prediction and diagnosis method, wherein the equipment data collection unit and the equipment data processing unit, which are equipped with at least one individual board having a plurality of input channels, are connected through a serial communication interface and control an integrated equipment failure prediction and diagnosis device,
Entering data on equipment to be diagnosed; and
A facility failure prediction diagnosis method comprising: performing real-time data determination on data of a diagnostic target facility input from at least one of a plurality of input channels using a diagnostic model corresponding to the corresponding channel. In claim 12,
Diagnosing the condition of the equipment based on the processed data, predicting the failure or life of the equipment, and determining the data in a way to calculate the replacement time of the equipment. A facility failure prediction diagnosis method further comprising: . In claim 12,
A method for predicting and diagnosing equipment failure, further comprising converting data input through at least one of a plurality of input channels into a digital signal, parsing the converted data, and storing the converted data in a buffer. In claim 14,
If the accumulated data stored in the buffer is larger than the preset size, inserting the data into a data queue to be thread-processed by the equipment data processing unit. Equipment failure prediction diagnosis method further comprising: In claim 15,
A method for predicting and diagnosing equipment failure, further comprising determining whether the size of the data queue is 0 or more, and extracting data from the data queue if the size of the data queue is 0 or more.

Images