KR20190091700A - Predictive maintenance method and system of facilities by studying type of electricity consumption data using facilities control characteristic - Google Patents

Abstract

According to the present invention, a predictive maintenance method of facilities of a power consumption data learning type using facility control characteristics comprises the following steps: repeatedly executing an operation over a plurality of times for the same control facility by using a control signal; detecting each of power consumption data in accordance with execution of the plurality of times; generating a reference power consumption data model for the control facility by using the detected power consumption data; detecting a degree of an error of individual power consumption data in accordance with control of the control facility by using the generated reference power consumption data model; and determining whether abnormality of the individual power consumption data exists in accordance with the degree of the error.

Description

Predictive maintenance method and system of facilities by studying type of electricity consumption data using facilities control characteristic}

The present invention relates to an automated production system, and more particularly, to a facility predictive maintenance technology for an automated production system.

With the fourth industrial revolution, smartization of manufacturing industry is becoming more important. Information such as power vibration of each facility in the factory can be collected and predicted and maintained in advance. In other words, the economic cost can be reduced through precautionary maintenance.

Facility management is the overall activity to maximize the function of the facility by increasing the productivity, reducing the opportunity loss, and planning, maintaining, and improving the device according to the company's policy to improve profitability. In other words, facility management can be referred to as keeping the entire process intact and improving production efficiency and quality, and reducing maintenance costs, all the way from the beginning until the plant or plant is lost.

Mechanization and automation are applied to modern facilities (equipment or systems), and the proportion of equipment is increasing in the management of companies. Product quality, cost, and output depend on the newness and reliability of equipment. It is becoming more frequent.

In addition, the environment around the company is changing into a viable form only when it meets the needs of consumers who demand urgency and diversity at the same time. In other words, how flexible and reliable it is in terms of equipment has emerged as a source of competitiveness of the company, and the maintenance and improvement of the equipment occupy an important position in the manufacturing production process.

However, in the current data collection method, the measurement equipment receives data at regular time intervals through the measurement equipment. In this step, there is a problem in that a difference in scan rate occurs depending on the performance and communication environment of the measurement equipment.

The problem to be solved by the present invention is to compare the power consumption data newly measured for each facility with the power consumption data model for each reference operation, the facility to determine the abnormality of the equipment state by comparing with the power consumption data model through statistical techniques The present invention relates to a power consumption data learning type predictive maintenance method and system using control characteristics.

In order to solve the above problems, the power consumption data learning type predictive maintenance method using the facility control characteristics comprises the steps of repeatedly performing the operation for the same control facility for the same control facility using a control signal; Detecting respective power consumption data according to the execution of the plurality of times; Generating a reference power consumption data model for the control facility using the detected power consumption data; Detecting an error degree of individual power consumption data according to the control of the control facility by using the generated reference power consumption data model; And determining whether the individual power consumption data is abnormal according to the detected degree of error.

The control signal may detect the respective power consumption data based on the start point and the end point of the PLC operation signal.

The reference power consumption data model may be generated by using a statistical technique on a cumulative result of the detected power consumption data.

In order to solve the above problems, the power consumption data learning type predictive maintenance system using facility control characteristics includes an operation control module which repeatedly executes an operation over a plurality of times for the same control facility using the same control signal; A communication module for transmitting the control signal to the control facility and receiving a parameter value according to the control signal provided from the control facility; A data detection module that detects respective power consumption data according to the execution of the plurality of times by using the received parameter value; A model generation module for generating a reference power consumption data model for the control facility using the detected power consumption data; And detecting an error degree of the individual power consumption data according to the control of the control facility by using the generated reference power consumption data model, and determining whether the individual power consumption data is abnormal according to the detected error degree. Characterized in that it comprises a determination module.

The data detection module may detect the respective power consumption data based on a start point and an end point of a PLC operation signal corresponding to the control signal.

The model generation module may generate the reference power consumption data model by using a statistical technique on a cumulative result of the detected power consumption data.

According to the present invention, a power consumption measurement data standardized model can be established using a controller operation signal reference such as a PLC, and the conventional simple scan data shows measurement time difference during operation. It is possible to accurately calculate the power consumption data curve of the equipment from which In addition, support for predictive maintenance of facilities in complex processes is possible through facility-specific data curves.

1 is a graph of an embodiment for describing a scan time for control data and power consumption data.
FIG. 2 is a block diagram illustrating an OPC UA (Unified Architecture) among the integrated data collection architectures applied to a power consumption data learning type predictive maintenance system using facility control characteristics according to the present invention.
3 is a block diagram of an embodiment for describing a power consumption data learning type predictive maintenance system using facility control characteristics according to the present invention.
FIG. 4 is a graph for explaining a process of detecting the respective power consumption data based on a start point and an end point of a PLC operation signal.
FIG. 5 is a flowchart illustrating an embodiment of a method for predicting power consumption data learning type facility predictive maintenance using facility control characteristics according to the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Embodiments of the present invention are provided to more fully explain the present invention to those skilled in the art, and the following embodiments may be modified in many different forms, the scope of the present invention It is not limited to the following embodiments. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" may include the plural forms as well, unless the context clearly indicates otherwise. Also, as used herein, "comprise" and / or "comprising" specifies the presence of the mentioned shapes, numbers, steps, actions, members, elements and / or groups of these. It is not intended to exclude the presence or the addition of one or more other shapes, numbers, acts, members, elements and / or groups. As used herein, the term "and / or" includes any and all combinations of one or more of the listed items.

Hereinafter, embodiments of the present invention will be described with reference to the drawings schematically showing embodiments of the present invention. In the drawings, for example, variations in the shape shown may be expected, depending on manufacturing techniques and / or tolerances. Accordingly, embodiments of the present invention should not be construed as limited to the specific shapes of the regions shown herein, but should include, for example, changes in shape resulting from manufacturing.

1 is a graph of an embodiment for describing a scan time for control data and power consumption data.

In automated production systems, control data and power consumption data are collected. In theory, the scan rate (200ms) of the control data and the power consumption data should be the same. However, the scan rate of the power consumption data is increased by about 500 ms due to the performance of the field measuring instrument and the communication equipment, and there is a data difference when matching the control data and the power consumption data. If the scan rate is prolonged and there is a difference, it may be difficult to reflect the changed equipment status quickly, and thus there may be a variety of predictive maintenance.

The present invention can compensate for the missing data due to the time difference between the control data and the power consumption data. As such, the scan rate of the power consumption measuring instrument is increased to provide a technique for supplementing missing power consumption data to enable predictive maintenance that can quickly reflect the status of changing equipment.

The power consumption data learning type predictive maintenance system using the facility control characteristic according to the present invention is an OPC UA (Unified Architecture), Support embedded system, Support variety of sensors via specific driver, Support high security remote communication, Support integrated data collection architectures with current and historical data interfaces, and the like.

FIG. 2 is a block diagram illustrating an OPC UA (Unified Architecture) among the integrated data collection architectures applied to a power consumption data learning type predictive maintenance system using facility control characteristics according to the present invention.

OPC UA is a vendor-independent communication protocol developed for industrial automation applications. Based on the client-server principle, seamless communication from the ERP system to individual sensors and actuators is possible. OPC UA is a flexible, platform-independent, and integrated safety mechanism that makes it an ideal communications protocol for implementing an Industry 4.0 environment.

The basic components of OPC UA consist of transport mechanisms and data modeling. OPC UA's data transfer defines a variety of mechanisms for optimizing different use cases. Through the connection between the IP-based world of IT and the shop floor, all production process data is transferred within a machine, between machines, or between a machine and a cloud database via a single protocol. OPC UA has defined and optimized the mapping to apply high-performance intranet communication as well as security-friendly Internet communication standards (Web services, XML, HTTP).

The data model defines the rules necessary to express the information model of OPC UA and defines the basics used to create address entry points and hierarchy types.

The Companion Information Model consists of different information model layers defined by OPC, IEC, EDDL, FDT, PLCopen, etc., and suppliers.

Built-in Information Models define information models for various process information domains on top of the existing OPC UA Meta Model.

Process Data (DA): Sensors, controls and location recorders generate process data. The data access (DA) information model provides this data to all clients on the network so that they can post-process directly. This can be done for any type of client such as a control unit, SCADA or ERP system.

Alarm Data (AC): The Alarm and Condition (AC) information model defines how alarms and condition conditions are handled. Changes in alarm and condition conditions trigger so-called “events”. Clients can subscribe to these events and select from the available companion figures the value that they want to receive as part of the notification message (eg message text or confirmation action).

Historical data (HA): The historical access (HA) information model provides client access to historical variable values and events. This allows you to read, write and edit the data. The data may be stored in a database, an archive, or various memory devices.

Program and Function (Prog): The Program (Prog) information model calls programs and functions (functions). This can be used, for example, to implement batch (batch) processes. It sends recipes to the line via OPC UA and returns control values back to the application.

The equipment of the automated production system always operates on the output signal of the controller. Accordingly, the same power consumption should occur when the same operation is performed. This feature is used to collect multiple power consumption data for the same facility and to create a power consumption data model for each reference operation. The model can determine the abnormality of the equipment condition by determining the degree to which the newly measured power consumption data deviates from the model in the rapidly changing controller data. To this end, first, the start and end of the control signal PLC operation signal as a reference point. Thereafter, accumulated accumulated power data during the control signal is accumulated. Then, a mathematical model of the accumulated data is derived. Thereafter, the derived mathematical model may be used to determine whether power consumption data is abnormal according to the control signal.

3 is a block diagram of an embodiment for describing a power consumption data learning type predictive maintenance system using facility control characteristics according to the present invention.

Referring to FIG. 3, the facility predictive maintenance system includes an operation control module 100, a communication module 110, a data detection module 120, a model generation module 130, and an abnormality determination module 140.

The operation control module 100 repeatedly executes an operation over a plurality of times for the same control facility by using the same control signal. In this case, the control facility controlled by the operation control module 100 may include an automated production facility. Here, the automated production facility may detect an operating environment according to the control signal and obtain a parameter value of the detected operating environment. The automated production facility may use a variety of sensors, including vibration sensors, temperature sensors, pressure sensors, and the like, to obtain parameter values of the sensed operating environment. The automated production facility may obtain parameter values of the operating environment at predetermined intervals and provide them to the facility predictive maintenance system.

The communication module 110 performs communication for transmitting and receiving data between the control facility and the system. In particular, the communication module 110 may transmit a control signal of the operation control module 100 to the control facility and receive parameter values measured from the control facility. In this case, the communication module 110 may include various communication chips such as a Wi-Fi chip, a Bluetooth chip, an NFC chip, and a wireless communication chip. At this time, the WiFi chip, the Bluetooth chip, and the NFC chip communicate with each other via WiFi, Bluetooth, and NFC. The wireless communication chip may perform communication according to various communication standards such as IEEE, Zigbee, 3rd Generation (3G), 3rd Generation Partnership Project (3GPP), Long Term Evoloution (LTE), and the like.

The data detection module 120 detects respective power consumption data according to execution of a plurality of rounds from respective parameter values received through the communication module 110. The data detection module 110 detects the respective power consumption data based on the start point and the end point of the PLC operation signal corresponding to the control signal.

FIG. 4 is a graph for explaining a process of detecting the respective power consumption data based on a start point and an end point of a PLC operation signal. Referring to FIG. 4, it can be seen that various power consumption data are detected according to the same control signal between the start point and the end point of the PLC operation signal. The detection of such power consumption data can be detected from parameter values measured and provided by various sensors provided in the control facility.

The model generation module 130 generates a reference power consumption data model for the control facility using the detected power consumption data. The model generation module 130 generates the reference power consumption data model by using a statistical technique on the accumulated result of the detected power consumption data. That is, the model generation module 130 may generate a reference power consumption data model using statistical techniques based on state analysis, probability evaluation and reliability analysis, and mathematical algorithms on the detected power consumption data.

The abnormality determination module 140 detects an error degree of individual power consumption data according to the control of a control facility by using the generated reference power consumption data model, and determines whether the individual power consumption data is abnormal according to the detected error degree. Determine.

The abnormality determination module 140 detects an error range of how much difference the individual power consumption data according to the control signal is compared with the reference power consumption data. Thereafter, the abnormality determination module 140 determines whether the detected error range exceeds a predetermined threshold. If the error range exceeds the predetermined threshold, the abnormality determination module 140 determines that the individual power consumption data is abnormal, and if the error range does not exceed the predetermined threshold, the abnormality determination module 140 determines the corresponding individual power. The consumption data is determined to be intact.

The abnormality determination module 140 may generate a message indicating whether or not the determined individual power consumption data is abnormal and output the same through an input / output interface module (not shown). In particular, it is possible to control to output an error message indicating that the individual power consumption data is abnormal.

FIG. 5 is a flowchart illustrating an embodiment of a method for predicting power consumption data learning type facility predictive maintenance using facility control characteristics according to the present invention.

The facility predictive maintenance system repeatedly executes a plurality of times for the same control facility using the same control signal (step S200).

The facility predictive maintenance system may transmit the same control signal to the control facility. Accordingly, the control facility may detect the facility operating environment according to the control signal and obtain a parameter value of the detected operating environment. The control facility may provide the facility prediction maintenance system with parameter values of the operating environment over multiple rounds, respectively.

After step S200, the facility predictive maintenance system detects respective power consumption data according to the execution of the plurality of times (step S202). The facility predictive maintenance system may detect the respective power consumption data based on the start point and the end point of the PLC operation signal as a control signal.

After step S202, the facility predictive maintenance system generates a reference power consumption data module for the control facility by using the detected power consumption data (step S204). The facility predictive maintenance system generates a reference power consumption data model by using a statistical technique on the cumulative result of the detected power consumption data. That is, the facility predictive maintenance system may generate a reference power consumption data model by using statistical techniques based on state analysis, probability evaluation and reliability analysis, and mathematical algorithms on the detected power consumption data.

After the step S204, the facility maintenance maintenance system detects the error degree of the individual power consumption data according to the control of the control facility, using the generated reference power consumption data model (step S206). The facility predictive maintenance system detects an error range on how much difference the individual power consumption data according to the control signal is compared with the reference power consumption data.

After step S206, the facility predictive maintenance system determines whether there is an abnormality of the individual power consumption data according to the detected error level (step S208). The facility predictive maintenance system determines whether the detected error range exceeds a certain threshold. If the margin of error exceeds a certain threshold, the facility maintenance maintenance system determines that the individual power consumption data is abnormal, and if the margin of error does not exceed a certain threshold, the facility prediction maintenance system fails. Determine that there is no.

The present invention can be applied to a variety of playback devices by implementing a software program and recording it on a computer-readable predetermined recording medium. Various playback devices may be PCs, laptops, portable terminals, and the like. For example, the recording medium may be a hard disk, a flash memory, a RAM, a ROM, or the like as an internal type of each playback device, or an optical disc such as a CD-R or a CD-RW, a compact flash card, a smart media, a memory stick, or a multimedia card. have.

While the embodiments of the present invention have been described as described above, the embodiments disclosed in the specification of the present invention do not limit the present invention. The scope of the present invention should be construed by the claims below, and all techniques within the scope equivalent thereto will be construed as being included in the scope of the present invention.

100: motion control module
110: communication module
120: data detection module
130: model generation module
140: abnormality determination module

Claims (6)

Repeatedly executing an operation over a plurality of times for the same control facility using the control signal;
Detecting respective power consumption data according to the execution of the plurality of times;
Generating a reference power consumption data model for the control facility using the detected power consumption data;
Detecting an error degree of individual power consumption data according to the control of the control facility by using the generated reference power consumption data model; And
And determining whether there is an abnormality of the individual power consumption data according to the detected degree of error. The method according to claim 1,
And the control signal detects each of the power consumption data based on a start point and an end point of a PLC operation signal. The method according to claim 1,
The power consumption data learning type predictive maintenance method using facility control characteristics, wherein the reference power consumption data model is generated using a statistical technique based on the accumulated result of the detected power consumption data. An operation control module for repeatedly executing an operation over a plurality of times for the same control facility using the same control signal;
A communication module for transmitting the control signal to the control facility and receiving a parameter value according to the control signal provided from the control facility;
A data detection module that detects respective power consumption data according to the execution of the plurality of times by using the received parameter value;
A model generation module for generating a reference power consumption data model for the control facility using the detected power consumption data; And
Detecting an error degree of the individual power consumption data according to the control of the control facility using the generated reference power consumption data model, and determining abnormality of the individual power consumption data according to the detected error degree Electric power consumption data learning type predictive maintenance system using the facility control characteristics, characterized in that it comprises a module. The method according to claim 4,
The data detection module,
The power consumption data learning type predictive maintenance system using facility control characteristics, characterized in that for detecting the respective power consumption data on the basis of the start point and the end point of the PLC operation signal corresponding to the control signal. The method according to claim 4,
The model generation module,
The power consumption data learning type predictive maintenance system using facility control characteristics, wherein the reference power consumption data model is generated using a statistical technique based on the accumulated result of the detected power consumption data.

Images