KR20230049386A - IoT System Including Smart Sensors, Gateways, and Servers for Facility Diagnosis and Method for Diagnosing and Prognosing Facilities - Google Patents

Abstract

An IoT system including a smart sensor, a gateway and a server for diagnosing a facility and a method for diagnosing and predicting a facility are disclosed. A method for diagnosing and predicting facilities processed by a server according to an embodiment of the present invention includes receiving sensing data collected by smart sensors from a gateway; generating processing data by pre-processing the sensing data; diagnosing the condition of the equipment using the processing data; and generating a control command for the smart sensor based on the diagnosis result.

Description

IoT System Including Smart Sensors, Gateways, and Servers for Facility Diagnosis and Method for Diagnosing and Prognosing Facilities}

The present invention relates to an IoT system including a smart sensor, a gateway and a server for diagnosing facilities and a method for diagnosing and predicting facilities.

Recently, with the 4th industrial revolution, research and R&D investment on smart factories are being actively conducted. In particular, in the field of smart factories, maintenance technologies for preventing or preventing equipment failure are being studied.

In maintenance technology, from the Breakdown maintenance (BM) technology that recovers after a failure of a conventional facility, recently, Condition Based Maintenance (CBM) technology that performs customized maintenance by diagnosing and analyzing the state of the facility, and Predictive Maintenance (PdM) technology, which monitors the condition of equipment to diagnose failures and predicts the remaining useful life, is becoming more important.

In smart factories, various types of sensors are used to diagnose the state of equipment. In particular, as research on IIoT (Industrial-IoT) technology becomes active, smart sensors equipped with communication functions or calculation functions are appearing in sensors that have previously only played a role of simply acquiring sensing data.

However, existing smart sensors are difficult to determine the state of facilities in the smart sensor itself due to resource limitations such as batteries and storage capacity, and diagnosis or prediction technology using sensing data still has limitations in terms of performance and accuracy. . Therefore, an integrated IoT system for accurately diagnosing or prognosing the state of facilities using data collected by smart sensors is required.

The present invention provides an integrated IoT system for accurately diagnosing or prognosing the state of facilities by utilizing data collected by smart sensors.

In addition, the present invention provides a smart sensor that maximizes battery efficiency by diagnosing the state of the facility itself and changing operating parameters such as the communication cycle of the smart sensor according to the diagnosis result, despite the limitations of the battery or other resources. .

A method for diagnosing and predicting facilities processed by a server according to an embodiment of the present invention includes receiving sensing data collected by smart sensors from a gateway; generating processing data by pre-processing the sensing data; diagnosing the condition of the equipment using the processing data; and generating a control command for the smart sensor based on the diagnosis result.

The method may further include performing prognosis on a failure or lifespan of the facility based on the processing data.

The step of performing the foreknowledge may include inputting the processed data to a machine learning model to obtain a prediction result, the machine learning model being trained to predict the state of the facility based on the processed data, and the machine learning model may be learned by population based training using hyperparameters extracted from the processed data.

The sensing data may include at least one of vibration data measured by vibration of the facility and current data measured from the facility.

The step of diagnosing the state of the facility may include determining whether the processing data is within a normal range using at least one threshold value among a plurality of threshold values for the type of facility, the cause of failure of the facility, and the normal range of parts of the facility. can decide

The method further includes receiving location data of the smart sensor, and diagnosing the state of the facility comprises determining whether the processing data is normal based on a threshold for a part corresponding to the location data of the smart sensor in the facility. range can be determined.

The method further includes receiving a diagnosis result of the smart sensor for the facility, and generating the control command includes, when the diagnosis result of the smart sensor and the diagnosis result determined by the server are different, the smart sensor determines the diagnosis result. You can create control commands that modify thresholds for diagnostics of your equipment.

Further comprising receiving battery information of the smart sensor, and generating the control command generates a control command for modifying an operating parameter of the smart sensor based on the battery information of the smart sensor, The operating parameter may include at least one of a sensing period of the smart sensor and a transmission period of the sensing data.

In the generating of the control command for modifying the operating parameter, a control command for modifying the operating parameter of the smart sensor may be generated using a battery consumption pattern and remaining battery capacity of the smart sensor included in the battery information.

An IoT system for diagnosing a facility according to an embodiment of the present invention includes a plurality of smart sensors that collect sensing data from the facility and communicate with a gateway or server; a gateway for receiving sensing data from the smart sensors and transmitting the sensing data to the server; and a server that diagnoses the facility (diagnosis) using the sensing data and performs prognosis of the facility;

The server receives sensing data collected by the smart sensors from a gateway, pre-processes the sensing data to generate processed data, diagnoses the state of the facility using the processed data, and diagnoses the state of the facility using the processed data. Results can be sent to the client.

The smart sensor may determine whether the sensing data is within a normal range, determine a state of the facility based on the sensing data, and change an operating parameter of the smart sensor when the sensing data does not fall within the normal range. there is.

A smart sensor for diagnosing a facility according to an embodiment of the present invention includes a sensing terminal for collecting sensing data from the facility; Based on the sensing data, a processor for diagnosing a state of a facility and controlling an operation of the smart sensor; and a communication terminal for communicating with a gateway or server for predictive maintenance of the facility, wherein the processor determines whether the sensing data is within a normal range, and when the sensing data does not belong to the normal range, the sensing data Based on this, it is possible to determine the state of the equipment and change the operating parameters of the smart sensor.

The processor determines an operating mode of the smart sensor according to the state of the facility, and the smart sensor operates according to any one operating mode determined by the processor among a plurality of operating modes using different operating parameters. can do.

The operation parameter may include a transmission period for transmitting the sensing data and the state of the facility to the gateway or server and a sensing period of the sensing terminal.

The plurality of operation modes include a normal mode and a boundary mode, wherein the transmission period and the sensing period of the boundary mode are set shorter than those of the normal mode, and the processor determines that the sensing data is higher than a preset normal threshold. In this case, the state of the facility may be determined as an abnormal state, and the operation mode may be changed from the normal mode to the alert mode.

The plurality of operating modes include a passive mode operating according to a control command received from the gateway, and the processor, when receiving a control request from the gateway, determines the operating mode of the smart sensor as a passive mode, , It can be controlled to transmit the state of the smart sensor or sensing data related to the control command to the gateway.

In the gateway for diagnosing a facility according to an embodiment of the present invention, the gateway includes a processor, and the processor receives sensing data about the facility from smart sensors and generates characteristic data from the sensing data. extract, determine the state of the facility based on the feature data, transmit the state of the facility and the feature data to a server, and the feature data, in the server, machine learning for diagnosis or prediction of the facility can be input into the model.

In a server for diagnosing a facility according to an embodiment of the present invention, the server includes a processor, and the processor receives sensing data collected by smart sensors from a gateway and preprocesses the sensing data. to generate processing data, diagnose the state of the facility using the processing data, and generate a control command for the smart sensor based on the diagnosis result.

The IoT system according to an embodiment of the present invention can accurately diagnose or predict the state of facilities by utilizing data collected by smart sensors in a smart factory.

In addition, the smart sensor according to an embodiment of the present invention diagnoses the state of the facility by itself despite the limitations of the battery or other resources, and changes the operating parameters such as the communication cycle of the smart sensor according to the diagnosis result. efficiency can be maximized.

1 is a block diagram illustrating an IoT system according to an embodiment of the present invention.
2 is a block diagram illustrating a gateway and a server according to one embodiment of the present invention.
3 is a flow chart illustrating a diagnosis and prediction method performed by a server according to an embodiment of the present invention.
4 is a block diagram showing the structure of a smart sensor according to an embodiment of the present invention.
5 illustrates operation modes of a smart sensor according to an embodiment of the present invention.
6 is a flowchart illustrating a diagnosis method performed by a smart sensor according to an embodiment of the present invention.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, since various changes can be made to the embodiments, the scope of the patent application is not limited or limited by these embodiments. It should be understood that all changes, equivalents or substitutes to the embodiments are included within the scope of rights.

Terms used in the examples are used only for descriptive purposes and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that the presence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiment belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

In addition, in the description with reference to the accompanying drawings, the same reference numerals are given to the same components regardless of reference numerals, and overlapping descriptions thereof will be omitted. In describing the embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description will be omitted.

1 is a block diagram illustrating an IoT system according to an embodiment of the present invention.

The present invention may relate to an integrated IoT system for accurately diagnosing or prognosing the state of facilities 101-103 by utilizing data collected by smart sensors 111-113. .

Referring to FIG. 1 , an IoT system according to an embodiment of the present invention, i) collects sensing data from facilities 101-103 and communicates with a gateway 121 or a server 131. of the smart sensors 111-113, ii) a gateway 121 that receives sensing data from the smart sensors 111-113 and transmits the sensing data to the server 131, and a facility 101 using the sensing data -103) It may include a server 131 that diagnoses (diagnosis) and performs prognosis of facilities (101-103). Each of the smart sensors 111 to 113, the gateway 121, and the server 131 may include a processor, and the processor may perform a diagnosis and prediction method according to an embodiment of the present invention.

Referring to FIG. 1 , one or more smart sensors 111-113 may be used to diagnose the state of facilities 101-103 (103) used in a smart factory. Each of the smart sensors 111-113 may be attached to different locations of one facility 101-103 or collect different types of sensing data. At this time, communication between the smart sensors 111-113 is possible to diagnose the state of the facilities 101-103. Alternatively, each smart sensor 111-113 may be attached to several facilities 101-103, respectively.

Even within the facilities 101-103, importance may vary depending on location, and sensing data may vary for each location. In addition, the sensing data may mean data such as vibration, current, and temperature of the facilities 101 to 103, and are not limited to specific examples. The present invention can increase the accuracy of diagnosis results for facilities 101-103 by using a plurality of smart sensors 111-113.

For example, the smart sensors 111-113 are attached to location A of the equipment 101-103, collect vibration data of the equipment 101-103, and the smart sensors 111-113 are connected to the equipment 101-103. 103), and can collect temperature data of facilities 101-103.

Then, the gateway 121 receives the sensing data collected by the plurality of smart sensors 111 to 113 or the results of diagnosing the facilities 101 to 103 using the sensing data, and transmits the results to the server 131 . In addition, the gateway 121 may transmit a control command to the smart sensors 111 to 113 or may transmit updated operating parameters or threshold values of the smart sensors 111 to 113 .

Alternatively, the gateway 121 extracts feature data from the sensing data, determines the state of the facilities 101-103 based on the feature data, and sends the state and feature data of the facilities 101-103 to the server 131. can be sent to

For example, edge computing may be applied to the gateway 121 . For example, the gateway 121 does not simply transmit the sensing data to the server 131, but preprocesses the sensing data to generate processed data or diagnose the state of the facilities 101-103 based on the sensing data. can

The server 131 receives the sensing data collected from the smart sensors 111-113 from the gateway 121 for predictive maintenance of the smart factory, diagnoses the state of the facilities 101-103 based on the sensing data, and , it is possible to predict the failure or remaining life of the equipment (101-103). The server 131 may deliver diagnosis results to the clients 141-143 or perform emergency actions on the facilities 101-103.

The clients 141 to 143 may refer to user terminals capable of communicating with the server 131 . The clients 141-143 may display diagnosis results or prediction results of the facilities 101-103 received from the server 131.

In the present invention, facilities 101-103 may be various types of facilities 101-103 used in production and manufacturing plants. For example, facilities 101 to 103 such as rotating machines and conveyors may be the subject of diagnosis. And, the smart sensors 111-113 can collect different types of sensing data, such as vibration, current, and temperature, from the facilities 101-103, and wired or wireless communication with the gateway 121 is possible.

2 is a block diagram illustrating a gateway and a server according to one embodiment of the present invention.

The gateway 201 and the server 202 may include processors 204 and 203, respectively, and the processors 204 and 203 may perform a diagnosis and prediction method according to an embodiment of the present invention. Referring to FIG. 2, the server 202 receives sensing data 211, location data 212, facility status 213, battery information 214, and operation parameters of the smart sensors from the gateway 201 or smart sensors. 215 and diagnostic results 216 of the smart sensor or gateway 201 for the facility may be received.

The server 202 receives the sensing data 211 collected by the smart sensors from the gateway 201, pre-processes the sensing data 211 to generate processing data, and uses the processing data to generate the state of the facility. It is possible to diagnose and generate a control command 217 for the smart sensor based on the diagnosis result.

The server 202 may perform prognosis on the failure or lifespan of the facility based on the processing data. The server 202 may transmit a diagnosis result 216 , a prediction result 218 , or sensing data 211 to the client 205 . The client 205 may display the sensing data 211, the diagnosis result 216, and the prediction result 218 on the display.

The server 202 may pre-process the sensing data 211 to generate processed data. For example, if the sensing data 211 is vibration data measured by the vibration of equipment, the processing data is root mean square (RMS) and peak level trend for vibration data of a certain time period. , a spectrogram, a power spectrum, a fast Fourier transform, and a cepstrum. Alternatively, the processing data may include harmonics converted into a frequency domain and extracted through signal separation from vibration data.

Alternatively, the server 202 may generate processed data by extracting features of the sensing data 211 as a neural network layer included in the machine learning model. For example, when a machine learning model learned through population-based training is used for diagnosis or prognosis of facilities, processed data may include hyper parameters.

The server 202 may determine whether the processing data is within a normal range by using at least one threshold among a plurality of thresholds for the type of facility, the cause of failure of the facility, and the normal range of parts of the facility.

As an example, the normal range indicates the range of the sensing data 211 that appears when the facility operates normally, is not limited to a specific example, and may be determined differently depending on the embodiment. The normal range may be determined in advance using statistical data on equipment abnormalities. The threshold may indicate a boundary of a normal range.

For example, a plurality of threshold values for the type of sensing data 211, the type of facility, the cause of failure of the facility, and the normal range of parts of the facility may be set in advance. For example, when the type of equipment is A, the normal range of the processing data may be determined in advance according to each case of a, b, and c as the cause of the failure. When the type of facility is A, the server 202 may diagnose the facility by determining whether the processing data falls within a normal range according to each case of failure causes a, b, and c. The server 202 may determine whether the processed data falls within a normal range by comparing the processed data with the threshold value.

For example, when receiving sensing data 211 from a plurality of smart sensors attached to one facility, the server 202 may determine the state of the facility by considering the weighted sum of each smart sensor. For example, based on the location data 212 of each smart sensor, a weight may be set higher for a smart sensor attached to a location having a higher importance, and processing data of the sensing data 211 collected from the smart sensor having a higher weight If is not in the normal range, it can be determined that the facility is in an abnormal state.

For example, when the type of equipment is A, normal ranges of processing data for parts a, b, and c located at different locations may be predetermined. When the type of facility is A, the server 202 may diagnose the facility by determining whether processing data collected for each part falls within each normal range.

For example, sensing data 211 of a high importance level may be determined in advance. For example, in a particular facility, temperature data may represent the state of the facility better than vibration data. In this case, when the processing data generated from the temperature data is not within a normal range, it may be determined that the facility is in an abnormal state.

For example, when the server 202 receives various types of sensing data 211 (eg, vibration data and current data), at least one of processing data of each sensing data 211 does not fall within the normal range. When not in use, equipment can be determined to be in an abnormal state.

For example, when current data and vibration data are received, the server 202 may diagnose a facility based on a relationship between the current data and vibration data. For example, the server 202 may determine an abnormal state of the facility by using a function defining a relationship between a pattern of current data and an RMS of vibration data.

For example, when the sensing data 211 is current data measured from a facility, the processing data may be current usage cycles. A normal range of current usage cycles may be predetermined. The server 202 may diagnose the state of the facility according to whether the current usage cycle falls within a normal range.

The server 202 may perform prognosis on the failure or lifespan of the facility based on the processing data. The server 202 may input the processed data into the machine learning model to obtain the prediction result 218 .

In one example, a machine learning model can be trained to predict the state of a facility based on process data. The machine learning model may be learned through population-based training using hyperparameters included in processed data. The hyperparameter may be feature data for metric analysis of the sensing data 211 . The machine learning model may include a convolutional neural network (CNN) structure.

The smart sensor according to an embodiment of the present invention determines whether the sensing data 211 is in a normal range, and if the sensing data 211 does not belong to the normal range, the state of the facility 213 based on the sensing data 211 , determine the operation mode of the smart sensor according to the state 213 of the facility, and determine the operation parameter 215 of the smart sensor.

The operating principle of the smart sensor will be described later with reference to FIG. 5 . The operation parameter 215 may include a transmission period for transmitting the sensing data 211 and the facility state 213 to the gateway 201 or the server 202 and a sensing period of the sensing terminal.

The server 202 may receive the state 213 of the facility determined by the smart sensor as a diagnosis result. The server 202 may compare the diagnosis result 216 of the facility directly determined based on the sensing data 211 and the diagnosis result determined from the smart sensor.

The server 202 may generate a control command 217 for modifying a threshold value for diagnosis of facilities in the smart sensor when the diagnosis result of the smart sensor and the diagnosis result 216 determined by the server 202 are different. The diagnosis result 216 may mean the state of the facility.

For example, when the smart sensor determines the facility to be in an abnormal state based on a preset threshold, but as a result of integrating and diagnosing the sensing data 211 collected from various smart sensors in the server 202, the facility is in a normal state. , the server 202 may generate a control command 217 that raises a threshold used in the smart sensor.

For example, if the smart sensor determines the facility to be in a normal state based on a preset threshold, but as a result of diagnosing by integrating the sensing data 211 collected from various smart sensors in the server 202, the facility is in an abnormal state. , the server 202 may generate a control command 217 that lowers the threshold used in the smart sensor.

For example, if the smart sensor determines the facility to be in a normal state based on a preset threshold, but as a result of diagnosing by integrating the sensing data 211 collected from various smart sensors in the server 202, the facility is in an abnormal state. , the gateway 201 may send a control command 217 to cause the smart sensor to operate in a passive mode.

For example, the server 202 may generate a control command 217 for modifying an operating parameter 215 of the smart sensor based on battery information 214 of the smart sensor. The operation parameter 215 may include at least one of a sensing period of the smart sensor and a transmission period of the sensing data 211 .

For example, the server 202 may generate a control command 217 for modifying an operating parameter 215 of the smart sensor using the battery consumption pattern and remaining battery capacity of the smart sensor included in the battery information 214. .

As an example, the server 202 may generate a control command 217 to increase the sensing period and the transmission period of the smart sensor when the battery remaining amount of the smart sensor is less than a reference level. For example, the server 202 may generate a control command 217 for increasing a sensing cycle and a transmission cycle of the smart sensor for each operation mode when the battery level of the smart sensor is less than a reference level.

For example, the server 202 may generate a control command 217 to increase the sensing cycle and transmission cycle of the smart sensor in a specific operating mode when the smart sensor excessively consumes battery in a specific operating mode. The server 202 may estimate the remaining battery capacity of the smart sensor after a certain period of time in consideration of the battery consumption pattern. The server 202 may generate a control command 217 for increasing a sensing period and a transmission period of the smart sensor when the predicted remaining amount is less than or equal to the reference amount.

According to one embodiment of the present invention, the gateway 201 may perform edge computing. For example, the gateway 201 may process the sensing data 211 by adjusting the number of retries based on the priority according to the operation mode of the smart sensor or assigning priority within a bandwidth.

The gateway 201 may determine whether the processing data is within a normal range by using at least one threshold value among a plurality of threshold values for the type of facility, the cause of failure of the facility, and the normal range of parts of the facility.

The gateway 201 receives sensing data 211 about equipment from smart sensors, extracts feature data from the sensing data 211, determines a state 213 of the equipment based on the feature data, and of the state 213 and characteristic data may be transmitted to the server 202 . Characteristic data, in server 202, can be input into a machine learning model for diagnosis or prognosis of the facility.

The gateway 201 may perform data polling. For example, the gateway 201 may receive vibration data and current data from smart sensors, and perform data polling according to the diagnosis result 216 of the facility. For example, if a facility is determined to be in an abnormal condition based on current data, gateway 201 may poll for vibration data.

3 is a flow chart illustrating a diagnosis and prediction method performed by a server according to an embodiment of the present invention.

In step 301, the server may receive sensing data collected by the smart sensors from the gateway. In step 302, the server may generate processed data by pre-processing the sensing data.

In the present invention, facilities may be various types of facilities used in production and manufacturing plants. For example, facilities such as rotating machines and conveyors may be subject to diagnosis. In addition, the smart sensor can collect different types of sensing data, such as vibration, current, and temperature, from facilities, and wired or wireless communication with the gateway is possible.

In step 303, the server may diagnose the state of the facility using the processing data. In step 304, the server may generate a control command for the smart sensor based on the diagnosis result.

The server may determine whether the processing data is within a normal range by using at least one threshold among a plurality of thresholds for the type of facility, the cause of failure of the facility, and the normal range of parts of the facility.

4 is a block diagram showing the structure of a smart sensor according to an embodiment of the present invention.

Referring to FIG. 4, the smart sensor 400 includes i) a sensing terminal 401, ii) a micro controller unit (MCU) 402, iii) an external Analog to Digital Converter (ADC) 406, iv) a memory ( 405), v) a communication terminal 408, vi) a switch 407, and vii) a battery 409.

The sensing terminal 401 is a terminal that collects sensing data from facilities, and may collect sensing data such as vibration, current, and temperature of facilities. Sensing data of the present invention is not limited to a specific type of data. Sensing data is collected in time series, converted into digital signals by the ADC, and processed by the processor 403. The sensing period of the sensing terminal 401 is determined by the processor 403 and may vary according to the operation mode of the smart sensor 400 .

The MCU 402 includes a processor 403 for calculating the sensing data and diagnosing the state of the equipment and determining the operation mode of the smart sensor 400 and an internal ADC 404 for converting the sensing data into a digital signal.

The external ADC 406 is an ADC with higher resolution and higher power consumption than the resolution of the internal ADC 404. When the facility is in a normal state, the internal ADC 404 with low power consumption is used, but the facility is in a normal state. If not, an external ADC 406 with high resolution can be used to accurately diagnose the state of the facility. Therefore, accurate diagnosis of facilities and power efficiency of the smart sensor 400 can be increased.

Specifically, the processor 403 may control the switch 407 according to the state of the equipment to convert the sensing data into a digital signal in either the internal ADC 404 or the external ADC 406 .

The memory 405 may store sensing data, operation parameters of the smart sensor 400, and the state of equipment. The communication terminal 408 is a terminal for receiving a control request from the gateway or transmitting sensing data, sensor status, facility status, battery 409 remaining amount, and operating parameters of the smart sensor 400 to the gateway. In addition, the communication terminal 408 is used to transmit and receive other smart sensors 411 and sensing data and facility status.

The battery 409 is used to supply power to the sensing terminal 401 , the MCU 402 , the external ADC 406 , the memory 405 , and the communication terminal 408 . Information about the remaining capacity of the battery 409 is transmitted to the processor 403 . The smart sensor 400 may receive power by wire and may additionally include an energy harvesting terminal that generates power by vibration or heat of a facility. The energy harvesting terminal may transfer power generated by vibration or heat of the facility to the battery 409 .

Also, the circuit board 410 is used for self-diagnosis of the sensing terminal 401 . For self-diagnosis of the sensing terminal 401 , the processor 403 may determine whether the sensing terminal 401 is in a normal state by applying a voltage to the sensing terminal 401 using the circuit board 410 . Also, the processor 403 may determine whether a defect has occurred in the circuit board 410 by controlling a switch between the circuit board 410 and the sensing terminal 401 .

5 illustrates operation modes of a smart sensor according to an embodiment of the present invention.

The smart sensor of the present invention may operate in any one of an active mode, a passive mode, and a test mode.

In the active mode, the processor may diagnose the state of the equipment based on the sensing data collected from the equipment and determine the operation mode of the smart sensor according to the state of the equipment. In the active mode, the operation mode of the smart sensor is determined based on time-series sensing data including abnormal data. The sensing data is data converted by an external ADC or an internal ADC.

In the active mode, the smart sensor operates according to one operation mode determined by a processor among a plurality of operation modes using different operating parameters. The plurality of operating modes may include normal mode, alert mode, transition mode, and risk mode. The processor determines an operating mode of the smart sensor based on the sensing data.

Specifically, when the equipment is in a normal state, the processor determines the operating mode of the smart sensor as the normal mode. And, when the state of the facility is abnormal, the processor may determine an operation mode of the smart sensor as any one mode among alert mode, transition mode, and risk mode. Also, operation modes of the smart sensor in the active mode may be changed to different operation modes according to sensing data.

Different operating parameters are set for each operating mode. The operating parameters include a transmission period for transmitting sensing data and facility status to a gateway or a server, a sensing period for a sensing terminal, and a check period for checking the facility status.

For example, in order to accurately diagnose the state of the facility, the transmission period, check period, and sensing period are set shorter in the alert mode than in the normal mode, and the transmission period, check period, and sensing period in the danger mode are set shorter than those in the alert mode.

At this time, the transition mode is set to have shorter transmission period, check period, and sensing period than the risk mode. The transition mode is controlled to operate with the shortest cycle because it is a point in time when the state of the equipment is changed to a dangerous state based on the sensing data. That is, the transition mode is an operation mode for notifying the server or smart factory operator that a major problem has occurred in the facility.

For example, in the normal mode, the transmission period may be set to 8 times per day, and the sensing period may be set to once per 20 seconds. In the alert mode, the transmission period may be set to 24 times per day, and the sensing period may be set to once per 10 seconds. In the danger mode, the transmission period may be set to once every 10 minutes, and the sensing period may be set to once every 5 seconds. In the transition mode, the transmission period may be set to once per 10 seconds, and the sensing period may be set to once per second.

And, in normal mode, the check cycle can be set to once per minute, in alert mode, the check cycle can be set to once every 30 seconds, and in transition mode and danger mode, the check cycle is 1 second. It can be set once in .

At each sensing period, the processor collects sensing data such as vibration, current, and temperature from the facility using the sensing terminal. For each transmission period, the processor may transmit (i) sensing data, (ii) a root mean square (RMS) calculated from the sensing data, and (iii) an operation mode of the smart sensor to a gateway or server using a communication terminal. As an example, RMS is calculated with 1024 pieces of sensing data (sensing data with a length of 1 second).

At each check period, the processor compares the sensing data with a normal threshold or an abnormal threshold to determine the state of the facility and the operating state of the smart sensor.

In addition, to promote power efficiency, an internal ADC included in the MCU is used in normal mode, and an external ADC with high resolution can be used for precise diagnosis of equipment in alert mode, transition mode, and risk mode.

When the operation mode is determined, the processor controls the switch so that the sensing data is converted into a digital signal by the external ADC or the internal ADC according to the determined operation mode. In this specification, an internal ADC may be defined as a first ADC, and an external ADC may be defined as a second ADC.

At each check period, the processor compares the sensing data with a normal threshold or an abnormal threshold to determine the state of the facility and the operating state of the smart sensor.

When the sensing data falls within the normal range, the processor determines the state of the equipment as the normal state, determines the operating mode as the normal mode, and controls the smart sensor according to the normal mode.

The normal threshold is determined by the range of sensing data that appears when the equipment operates normally. That is, the normal threshold means the upper or lower limit of the normal range of the sensing data when the facility operates normally. The normal threshold represents the upper limit of the normal range.

In addition, when the operation mode is the normal mode, the processor compares the sensing data with the normal threshold at every check period to determine the state of the facility and the operation mode of the smart sensor.

For example, when the sensing data is higher than a preset normal threshold, the processor determines the state of the facility as an abnormal state and changes the operation mode from the normal mode to the alert mode. And, the processor may transmit the changed state of the facility to the gateway or the server through the communication terminal.

Specifically, when the sensing data is higher than the preset normal threshold, a lower trigger occurs, and the processor determines the state of the facility based on the number or ratio of the sensing data higher than the normal threshold for a certain period of time after the lower trigger occurs, and Determine the operating mode of the sensor.

For example, the normal range may be 0-5g, and when the sensing data is greater than 5g, a subtrigger occurs, and the processor sets the equipment according to the number or ratio of sensing data higher than the normal threshold for a certain period of time after the subtrigger occurs. and the operating mode of the smart sensor. For example, the period of time may be 10 minutes.

The processor determines the state of the facility as an abnormal state when the sensing data that does not belong to the normal range among the sensing data for a certain period of time is more than the triggering number or the sensing data that is higher than the normal threshold among the sensing data for a certain period of time is more than the triggering number. The operation mode of the can be changed from normal mode to alert mode.

For example, when 1000 pieces of sensing data are collected every 20 seconds, if the number of sensing data higher than the normal threshold among the sensing data is 250, which is more than the triggering number of 200, the processor determines the state of the facility as an abnormal state, The operating mode can be changed from normal mode to alert mode.

And, when a lower trigger occurs due to sensing data higher than the normal threshold, the processor determines the ratio of the sensing data higher than the normal threshold among the sensing data for every predetermined time interval after the lower trigger occurs, and the ratio is higher than the triggering ratio, If it increases from the previous time interval, the state of the facility may be determined as an abnormal state, and the operation mode may be changed from the normal mode to the alert mode.

For example, when sensing data is collected every 20 seconds, if the ratio of sensing data higher than the normal threshold among the sensing data was 15% in section N, but increased to 20% in section N+1, the processor The state may be determined as an abnormal state, and the operation mode may be changed from normal mode to alert mode.

Alternatively, if the ratio of the sensing data higher than the normal threshold is 25% higher than the triggering ratio in section N, the processor may determine the state of the facility as an abnormal state and change the operation mode from the normal mode to the alert mode.

Alternatively, when the ratio of the sensing data higher than the normal threshold among the sensing data continuously increases for a predetermined time, the processor may determine the state of the facility as an abnormal state and change the operation mode from the normal mode to the alert mode. And, the processor may transmit that the state of the facility is in an abnormal state to the gateway or the server through the communication terminal. And, the processor may transmit the operation mode of the smart sensor to the server or gateway using the communication terminal.

However, when a lower trigger occurs and the ratio of sensing data higher than the normal threshold among the sensing data in the current time section does not increase than the previous time section and the ratio is lower than the triggering ratio, the lower trigger ends and the equipment state (normal state) or the operation mode (normal mode) of the smart sensor is maintained.

However, when the state of the facility is determined to be abnormal and the ratio of the sensing data higher than the normal threshold among the sensing data of the current time section is lower than the ratio of the previous time section in the ratio of the current time section, a lower trigger occurs.

After the occurrence of the lower trigger, if the ratio of the sensing data higher than the normal threshold decreases, the processor may determine the state of the facility as the normal state and change the operation mode from the alert mode to the normal mode. And, the processor may transmit that the facility is in a normal state to the gateway or the server through the communication terminal.

And, when the operating mode of the smart sensor is determined to be the alert mode and the sensing data is higher than a preset abnormal threshold, an upper trigger occurs, and the processor changes the operating mode from the alert mode to the transition mode. And, the processor may transmit the operation mode of the smart sensor to the server or gateway using the communication terminal.

In addition, since the sensing data may deviate more than the normal range as the number of defects in the facility increases, the abnormality threshold is set higher than the normal threshold.

When the operation mode of the smart sensor is determined to be the transition mode, the processor determines the number and ratio of sensing data higher than the abnormal threshold among the sensing data for a certain period of time for a certain period of time, the RMS calculated from the sensing data, the state of the facility, and the operation mode of the smart sensor to the gateway or server.

The processor may change the operation mode of the smart sensor from the transition mode to the danger mode when the number of sensing data higher than the abnormal threshold is greater than or equal to the triggering number among the sensing data for a predetermined time.

For example, when 1000 pieces of sensing data are collected every 20 seconds, and the number of sensing data higher than the abnormality threshold (8g) among the sensing data is 250, which is greater than the triggering number of 200, the processor changes the operation mode to the transition mode. You can change to risk mode.

In addition, the processor determines the ratio of sensing data higher than the abnormal threshold among the sensing data for each predetermined time interval, and when the ratio is higher than the triggering ratio and increases from the previous time interval, the operation mode may be changed from the transition mode to the dangerous mode. .

For example, when sensing data is collected every 10 seconds, if the ratio of sensing data higher than the abnormal threshold among the sensing data is 15% in section N but increases to 20% in section N+1, the processor enters the operating mode can be changed from transition mode to risk mode.

Alternatively, when the ratio of the sensing data higher than the abnormal threshold among the sensing data continuously increases during a plurality of consecutive time intervals, the processor may change the operation mode from the transition mode to the dangerous mode. And, the processor may transmit the operation mode of the smart sensor to the server or gateway using the communication terminal.

However, when the sensing data of the current time interval is lower than the abnormal threshold, an upper trigger occurs and the processor changes the operation mode from the danger mode to the transition mode.

And, when the ratio of the sensing data higher than the abnormality threshold decreases in the transition mode, the processor may change the operation mode from the transition mode to the boundary mode. And, the processor may transmit the operation mode of the smart sensor to the server or gateway using the communication terminal.

Also, when a control command from the gateway is received, the smart sensor operates in a passive mode rather than an active mode. In the passive mode, the processor may transmit the most recent operation mode of the smart sensor, transmit sensing data for a specific event, or transmit information about the version of firmware included in the smart sensor, according to the request of the gateway. operates in active mode again upon request of the gateway.

When the smart sensor operates in an active mode, it is changed to a test mode every self-diagnosis period by a processor. In the test mode, the processor may determine whether the sensing terminal is in a normal state by applying a voltage to the sensing terminal using a circuit board.

5(a) is a diagram illustrating the state of the facility and the operation mode determined by the smart sensor when the state of the facility is in a normal state. When the state of the facility is in the normal state, the processor determines the state of the facility as the normal state and determines the operation mode of the smart sensor as the normal mode (A). And, the state of the smart sensor may be determined in the test mode. When the state of the smart sensor is not normal, the processor may notify the user by transferring the state of the smart sensor to the gateway.

Referring to (a) of FIG. 5 , as the operation mode is determined to be the normal mode (A), an internal ADC with low power consumption is used, the sensing cycle is set to LOW, and the communication cycle is also set to LOW.

In addition, as the state of the facility is determined to be normal, the priority for distinguishing the importance of the sensing data received from each smart sensor in the gateway or server is set to a lower level.

5(b) is a diagram showing the state of the facility and the operation mode determined by the smart sensor when the state of the facility is in an abnormal state. When the state of the facility is in an abnormal state, the processor determines the state of the facility to be in an abnormal state, and determines the operation mode as the alert mode (B) according to the state of the facility.

Referring to (b) of FIG. 5 , as the operation mode is determined to be the boundary mode (A), an external ADC with high resolution is used, the sensing period is set to phase, and the communication period is set to phase. The sensing period and the communication period are shorter when it is high than when it is low.

In addition, as the state of the facility is determined to be abnormal, the priority for classifying the importance of the sensing data received from each smart sensor in the gateway or server is set to 0. When the priority is determined as high, the server takes precedence in diagnosing the facility rather than when the priority is lower.

Referring to (b) of FIG. 5 , when the remaining capacity of the battery is 60% and the reference capacity is 50%, the processor may use an internal ADC or set a longer sensing period and communication period in consideration of the battery state.

5(c) is a diagram showing that the operation mode of the smart sensor is determined as a transition mode based on sensing data. Referring to (c) of FIG. 5 , the state of the facility is maintained in an abnormal state, and the processor determines the operation mode as a transition mode based on the sensing data.

Referring to (c) of FIG. 5, as the operation mode is determined to be the transition mode, an external ADC with high power consumption and resolution is used, the sensing period is set to the best, and the communication period and check period are also set to the best. The transition mode is controlled to operate with the shortest cycle because it is a point in time when the state of the equipment is changed to a dangerous state based on the sensing data.

In addition, as the operation mode is determined as the transition mode, the priority for distinguishing the importance of the sensing data received from each smart sensor in the gateway or server is set to the highest.

5(d) is a diagram showing that the operation mode of the smart sensor is determined to be a risk mode based on sensing data. Referring to FIG. 5(d), the state of the facility is maintained in an abnormal state, and the processor Based on the data, the operation mode is determined as a risk mode.

Referring to (d) of FIG. 5 , as the operation mode is determined to be the critical mode, an external ADC with high power consumption and resolution is used, the sensing period is set to phase, and the communication period is set to phase.

In addition, as the operation mode is determined to be the risk mode, the gateway or server sets a priority level for classifying the importance of the sensing data received from each smart sensor.

6 is a flowchart illustrating a diagnosis method performed by a smart sensor according to an embodiment of the present invention.

In step 601, the processor of the smart sensor identifies the sensing data collected by the sensing terminal from the facility. The sensing data may refer to time-series data such as vibration, current, and temperature of a facility, and is not limited to a specific example.

In step 602, the processor of the smart sensor determines whether the sensing data is within a normal range. When the sensing data of a certain time interval falls within the normal range, the processor identifies the sensing data of the next time interval.

However, when the sensing data of the predetermined time period does not fall within the normal range, the processor determines the state of the facility based on the sensing data in step 603 . When the sensing data is higher than the preset normal threshold, a sub-trigger occurs, the processor determines the state of the equipment as an abnormal state based on the number of triggers or the triggering ratio, and changes the operation mode from the normal mode to the alert mode.

Alternatively, the processor changes the operation mode of the smart sensor from the alert mode to the transition mode when the sensing data is higher than a preset abnormal threshold.

In step 604, the processor of the smart sensor changes operating parameters of the smart sensor in consideration of the state of the equipment.

Specifically, when the state of the facility is determined to be abnormal, the processor controls the smart sensor to operate in alert mode. The alert mode has shorter transmission and sensing cycles than the normal mode, and the danger mode has shorter transmission and sensing cycles than the alert mode.

Meanwhile, the method according to the present invention is written as a program that can be executed on a computer and can be implemented in various recording media such as magnetic storage media, optical reading media, and digital storage media.

Implementations of the various techniques described herein may be implemented in digital electronic circuitry, or in computer hardware, firmware, software, or combinations thereof. Implementations may be a computer program product, i.e., an information carrier, e.g., a machine-readable storage, for processing by, or for controlling, the operation of a data processing apparatus, e.g., a programmable processor, computer, or plurality of computers. It can be implemented as a computer program tangibly embodied in a device (computer readable medium) or a radio signal. A computer program, such as the computer program(s) described above, may be written in any form of programming language, including compiled or interpreted languages, and may be written as a stand-alone program or in a module, component, subroutine, or computing environment. It can be deployed in any form, including as other units suitable for the use of. A computer program can be deployed to be processed on one computer or multiple computers at one site or distributed across multiple sites and interconnected by a communication network.

Processors suitable for processing a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from read only memory or random access memory or both. Elements of a computer may include at least one processor that executes instructions and one or more memory devices that store instructions and data. In general, a computer may include, receive data from, send data to, or both, one or more mass storage devices that store data, such as magnetic, magneto-optical disks, or optical disks. It can also be combined to become. Information carriers suitable for embodying computer program instructions and data include, for example, semiconductor memory devices, for example, magnetic media such as hard disks, floppy disks and magnetic tapes, compact disk read only memory (CD-ROM) ), optical media such as DVD (Digital Video Disk), magneto-optical media such as Floptical Disk, ROM (Read Only Memory), RAM (RAM) , Random Access Memory), flash memory, EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), and the like. The processor and memory may be supplemented by, or included in, special purpose logic circuitry.

In addition, computer readable media may be any available media that can be accessed by a computer, and may include both computer storage media and transmission media.

Although this specification contains many specific implementation details, they should not be construed as limiting on the scope of any invention or what is claimed, but rather as a description of features that may be unique to a particular embodiment of a particular invention. It should be understood. Certain features that are described in this specification in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable subcombination. Further, while features may operate in particular combinations and are initially depicted as such claimed, one or more features from a claimed combination may in some cases be excluded from that combination, and the claimed combination is a subcombination. or sub-combination variations.

Similarly, while actions are depicted in the drawings in a particular order, it should not be construed as requiring that those actions be performed in the specific order shown or in the sequential order, or that all depicted actions must be performed to obtain desired results. In certain cases, multitasking and parallel processing can be advantageous. Further, the separation of various device components in the embodiments described above should not be understood as requiring such separation in all embodiments, and the program components and devices described may generally be integrated together into a single software product or packaged into multiple software products. You have to understand that you can.

On the other hand, the embodiments of the present invention disclosed in this specification and drawings are only presented as specific examples to aid understanding, and are not intended to limit the scope of the present invention. In addition to the embodiments disclosed herein, it is obvious to those skilled in the art that other modified examples based on the technical idea of the present invention can be implemented.

101-103: Equipment
111-113: smart sensor
121: gateway
131: server
141-143: Client

Claims (18)

In the method for diagnosing and predicting facilities processed by a server,
Receiving sensing data collected by smart sensors from a gateway;
generating processing data by pre-processing the sensing data;
diagnosing the condition of the equipment using the processing data; and
Generating a control command for the smart sensor based on the diagnosis result
Diagnosis and prognosis method of facilities including. According to claim 1,
The diagnostic and prognostic method further comprising the step of performing prognosis on the failure or life of the facility based on the processing data. According to claim 2,
The step of performing the foreknowledge,
Obtaining a prediction result by inputting the processed data into a machine learning model;
The machine learning model,
Trained to predict the state of the equipment based on the processing data,
The machine learning model is learned by population-based training using hyperparameters included in the processed data. According to claim 1,
The sensing data,
A diagnosis and prediction method comprising at least one of vibration data measured by the vibration of the facility and current data measured from the facility. According to claim 1,
The step of diagnosing the state of the facility,
Diagnosis and prediction method for determining whether the processing data is in a normal range by using at least one threshold value among a plurality of threshold values for the type of equipment, the cause of failure of the equipment, and the normal range of parts of the equipment. According to claim 1,
Further comprising receiving location data of the smart sensor,
The step of diagnosing the state of the facility,
Based on a threshold for a part corresponding to the location data of the smart sensor in the facility, determining whether the processing data is in a normal range, diagnosis and prediction method. According to claim 1,
Further comprising receiving a diagnostic result of the smart sensor for the facility,
Generating the control command,
When the diagnosis result of the smart sensor and the diagnosis result determined by the server are different, the smart sensor generates a control command for modifying a threshold for diagnosis of the facility. According to claim 1,
Further comprising receiving battery information of the smart sensor,
Generating the control command,
Based on the battery information of the smart sensor, a control command for modifying an operating parameter of the smart sensor is generated;
The operating parameters are,
A diagnosis and prediction method comprising at least one of a sensing period of the smart sensor and a transmission period of the sensing data. According to claim 8,
Generating a control command for modifying the operating parameter,
Generating a control command for modifying an operating parameter of the smart sensor using a battery consumption pattern and remaining battery capacity of the smart sensor included in the battery information, diagnosis and prediction method. In the IoT system for diagnosing equipment,
A plurality of smart sensors that collect sensing data from the facility and communicate with a gateway or server;
a gateway for receiving the sensing data from the smart sensors and transmitting the sensing data to the server; and
A server that diagnoses the facility (diagnosis) using the sensing data and performs prognosis of the facility
including,
The server,
The sensing data collected by the smart sensors is received from the gateway, the sensing data is preprocessed to generate processed data, the status of the facility is diagnosed using the processed data, and the diagnosis result is sent to the client. transmitting,
IoT system. According to claim 10,
The smart sensor,
Determining whether the sensing data is within a normal range, determining a state of the facility based on the sensing data when the sensing data does not belong to the normal range, and changing an operating parameter of the smart sensor. IoT system. In a smart sensor for diagnosing equipment,
a sensing terminal that collects sensing data from the facility;
Based on the sensing data, a processor for diagnosing a state of a facility and controlling an operation of the smart sensor; and
A communication terminal for communicating with a gateway or server for predictive maintenance of the facility
including,
the processor,
determining whether the sensing data is within a normal range, determining a state of the facility based on the sensing data when the sensing data does not belong to the normal range, and changing an operating parameter of the smart sensor;
smart sensor. According to claim 12,
the processor,
Determine the operating mode of the smart sensor according to the state of the facility;
The smart sensor,
A smart sensor that operates according to any one operation mode determined by the processor among a plurality of operation modes using different operation parameters. According to claim 13,
The operating parameters are,
A smart sensor comprising a transmission period for transmitting the sensing data and the state of the facility to the gateway or server and a sensing period of the sensing terminal. According to claim 13,
The plurality of operation modes include a normal mode and an alert mode,
In the alert mode, the transmission period and the sensing period are set shorter than those of the normal mode,
the processor,
When the sensing data is higher than a preset normal threshold, the smart sensor determines the state of the facility as an abnormal state and changes the operation mode from the normal mode to the alert mode. According to claim 13,
The plurality of operating modes,
A passive mode operating according to a control command received from the gateway;
the processor,
When a control command of the gateway is received, determining an operation mode of the sensor of the smart as a passive mode, and controlling to transmit a state of the smart sensor or sensing data associated with the control command to the gateway, the smart sensor. In the gateway for the diagnosis of equipment,
The gateway includes a processor;
the processor,
Receives sensing data about the facility from smart sensors, extracts feature data from the sensing data, determines the state of the facility based on the feature data, and transmits the state of the facility and the feature data to a server. do,
The feature data is input to a machine learning model for diagnosis or prognosis of the facility in the server.
gateway. In the server for diagnosing equipment,
The server includes a processor,
the processor,
Sensing data collected by smart sensors is received from a gateway, processing data is generated by pre-processing the sensing data, the state of the facility is diagnosed using the processing data, and the state of the facility is diagnosed based on the diagnosis result. generating control commands for smart sensors;
server.

Images