KR102300835B1 - System and method of monitoring by analyzing IoT data - Google Patents

Abstract

The present invention relates to IoT data analysis. A monitoring method through IoT data analysis includes the steps of generating an N-dimensional normal data set having N features acquired during a steady state, and using the N-dimensional normal data set, one N-dimensional normal data distribution and N (N -1) generating a dimensional normal data distribution, generating N-dimensional monitoring data having N features obtained after the steady state, and using the N-dimensional normal data distribution, the N-dimensional monitoring data is an outlier determining whether it is, and detecting the feature causing the outlier from the N-dimensional monitoring data using the distribution of the N (N-1)-dimensional normal data.

Description

{System and method of monitoring by analyzing IoT data}

The present invention relates to IoT data analysis.

IoT sensors are used to measure the status of the monitored object. Sensing data generated by a plurality of IoT sensors installed in one monitoring target is not only of various types, but also generated at regular intervals, so big data processing is required. Most users want to know what kind of abnormality has occurred in the state of the monitoring target, and when the abnormal state is identified, they want to take immediate action.

Korean Patent Publication No. 10-1976585

An object of the present invention is to provide a monitoring system and method that can easily determine the status of a monitoring target.

According to one aspect of the present invention, there is provided a monitoring method through IoT data analysis. A monitoring method through IoT data analysis includes the steps of generating an N-dimensional normal data set having N features acquired during a steady state, and using the N-dimensional normal data set, one N-dimensional normal data distribution and N (N -1) generating a dimensional normal data distribution, generating N-dimensional monitoring data having N features obtained after the steady state, and using the N-dimensional normal data distribution, the N-dimensional monitoring data is an outlier determining whether it is, and detecting the feature causing the outlier from the N-dimensional monitoring data using the distribution of the N (N-1)-dimensional normal data.

As an embodiment, the N-dimensional normal data distribution may be generated by applying a Gaussian Mixture Model (GMM) to the N-dimensional normal data set.

In one embodiment, the N (N-1)-dimensional normal data distribution is obtained by removing the n-th feature (1≤n≤N) from the N-dimensional normal data set to obtain N (N-1)-dimensional normal data sets. , and may be generated by applying a Gaussian Mixture Model (GMM) to the N (N-1)-dimensional normal data sets.

In one embodiment, the generating of the N-dimensional normal data set having the N features acquired during the normal state includes: receiving k pieces of first sensing data from an IoT sensor installed in a monitoring target during the steady state period and generating an N-dimensional normal data set with N features associated with each of the k pieces of first sensing data.

In one embodiment, the generating of N-dimensional monitoring data having N features obtained after the normal state includes: receiving k pieces of second sensing data from the IoT sensor after the normal state; The method may include generating N-dimensional monitoring data with N features associated with each of the pieces of second sensing data.

In one embodiment, it is determined whether the N-dimensional monitoring data is an outlier using the N-dimensional normal data distribution, and the characteristic that caused the outlier using the N (N-1)-dimensional normal data distribution is described above. The detecting in the N-dimensional monitoring data includes determining whether the N-dimensional monitoring data is included in the N-dimensional normal data distribution, and if not included in the N-dimensional normal data distribution, in the N-dimensional monitoring data, an nth feature generating N (N-1)-dimensional monitoring features by removing (1≤n≤N), and the N (N-1)-dimensional monitoring features are each of the N (N-1)-dimensional normal data distributions It may include the step of determining whether it is included in the.

In an embodiment, the characteristic may be any one of an absolute value, a slope, and a fast Fourier transform value of the first sensed data or the second sensed data.

According to another aspect of the present invention, a monitoring system through IoT data analysis is provided. The monitoring system through IoT data analysis collects k pieces of first sensed data from the IoT sensors installed in the monitoring target during a steady state section, and a sensing data collection unit that collects k pieces of second sensed data after the steady state section , a feature acquisition unit that acquires N features from the first sensing data and the second sensing data to generate an N-dimensional normal data set and N-dimensional monitoring data, and one N-dimensional normal using the N-dimensional normal data set A steady-state model generator for generating a data distribution and N (N-1)-dimensional normal data distributions, and N-dimensional monitoring data using the N-dimensional normal data distribution and the N (N-1)-dimensional normal data distributions The monitoring unit for determining whether is an outlier may be included.

In an embodiment, the steady-state model generating unit generates the N-dimensional normal data distribution by applying a Gaussian Mixture Model (GMM) to the N-dimensional normal data set, and the n-th feature (1) in the N-dimensional normal data set. ≤n≤N) is removed to generate N (N-1)-dimensional stationary data sets, and GMM is applied to the N (N-1)-dimensional stationary data sets to generate the N (N-1)-dimensional stationary data sets. You can create a data distribution.

In one embodiment, the monitoring unit determines whether the N-dimensional monitoring data is included in the N-dimensional normal data distribution, and if not included in the N-dimensional normal data distribution, in the N-dimensional monitoring data, an nth feature (1 ≤n≤N) is removed to generate N (N-1)-dimensional monitoring features, and whether the N (N-1)-dimensional monitoring features are included in each of the N (N-1)-dimensional normal data distributions. can be judged.

According to the present invention, it is possible to easily grasp the state of the monitoring target.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention is described with reference to the embodiments shown in the accompanying drawings. For ease of understanding, like elements have been assigned like reference numerals throughout the accompanying drawings. The configuration shown in the accompanying drawings is merely an exemplary embodiment for explaining the present invention, and is not intended to limit the scope of the present invention. In particular, the accompanying drawings, in order to help the understanding of the invention, some components are expressed somewhat exaggerated.
1 is a diagram to exemplarily explain the concept of monitoring through IoT data analysis.
2 is a flowchart exemplarily illustrating a monitoring method through IoT data analysis.
3 is a diagram exemplarily illustrating sensing data collected from a monitoring target and characteristics obtained therefrom.
4 is a diagram exemplarily illustrating an N-dimensional and (N-1)-dimensional normal data distribution generated using features.
5 is a diagram exemplarily illustrating a situation in which an abnormal point occurs during a monitoring process.
6 is a diagram exemplarily illustrating a monitoring system through IoT data analysis for implementing the monitoring method illustrated in FIGS. 1 to 5 .

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and will be described in detail through the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents and substitutes included in the spirit and scope of the present invention.

When a component is referred to as being “connected” or “connected” to another component, it is understood that the other component may be directly connected or connected to the other component, but other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that no other element is present in the middle.

Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

The terms used herein are used only to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In this specification, terms such as "comprises" or "have" are intended to designate that the features, numbers, steps, operations, components, parts, or combinations thereof described in the specification exist, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.

In addition, the components of the embodiment described with reference to each drawing are not limitedly applied only to the embodiment, and may be implemented to be included in other embodiments within the scope of maintaining the technical spirit of the present invention, and also Even if the description is omitted, it is natural that a plurality of embodiments may be re-implemented as a single integrated embodiment.

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

1 is a diagram to exemplarily explain the concept of monitoring through IoT data analysis.

Referring to FIG. 1 , a motor is exemplified as a monitoring object. A plurality of IoT sensors are fixed to the motor or arranged to be spaced apart by a predetermined distance (hereinafter, installed) to measure the state of the motor. The sensing data generated by the IoT sensor measuring the motor status is used to determine the motor status. The IoT sensor may measure a factor related to the operation of the monitoring target, such as the rotation speed of the motor, vibration, temperature, etc., or measure a factor related to the surrounding environment of the monitoring target, such as temperature and humidity around the motor. have. Here, the sensing data generated by the IoT sensor may vary depending on the monitoring target, and the monitoring target is not limited to the motor.

One or more features may be obtained from one piece of sensing data. For example, when measuring a speed, an absolute value of the speed, a slope, an acceleration, a fast Fourier transform value (FFT), etc. may be obtained from sensing data called the speed. Sensing data may be different depending on an installation location even if the same element is measured, and characteristics obtained therefrom may also be different.

Whether the monitoring target is normal may be monitored using a plurality of features. A plurality of characteristics acquired while the monitoring target operates normally (hereinafter, referred to as a normal state) may be distributed within a predetermined area. The normal data distribution represents a data set that combines a plurality of features acquired during a steady state in a high-dimensional area. After the steady state, the data set composed of a plurality of features obtained from the sensed data is compared with the normal data distribution. The occurrence of data out of the normal data distribution (hereinafter referred to as an outlier) indicates that the monitoring target is operating abnormally.

FIG. 2 is a flowchart exemplarily illustrating a monitoring method through IoT data analysis, a process performed by a monitoring system through IoT data analysis, and FIG. 3 is sensing data collected from a monitoring target and features obtained therefrom. It is a diagram exemplarily shown, and FIG. 4 is a diagram exemplarily illustrating the distribution of N-dimensional and N-1 dimensional normal data generated using features, and FIG. It is the drawing shown.

The monitoring method through IoT data analysis includes a process of generating a normal data distribution (steps 100 to 150) and a process of monitoring an object based on the generated normal data distribution (steps 160 to 170). The normal data distribution includes one N-dimensional stationary data distribution generated using N features and N (N-1)-dimensional stationary data distributions generated using (N-1) features. The N-dimensional normal data distribution is used as a basis for judging whether an outlier occurs, and the (N-1)-dimensional normal data distribution is used to detect a feature that causes the outlier.

Referring to FIG. 2 , sensing data for a monitoring target is collected ( 100 ). The sensing data is generated by a plurality of IoT sensors installed in the monitoring target. The sensing data may be generated at a predetermined time interval, for example, 1 second interval, or may be continuously generated. 3A illustrates _{sensing data d k@time} generated by k IoT sensors. The sensing data d _k@time is used as a basis for acquiring a plurality of features used for generating a normal data distribution and detecting an outlier.

The first sensing data belonging to the normal state is selected from the input sensing data ( 110 ). The steady state can be set in various ways. For example, the normal state is a specific time point at which monitoring is started, for example, a time period when a predetermined time elapses from _{t 1 , for example, a time period from t 245} is set to a normal state, or the first sensing data is Within the maximum number of times acquired (that is, the first to 245th sensing data) can be set to a normal state. Meanwhile, the steady state may be plural. In the case of a motor, a steady state may be set for each operation mode (eg, continuous duty, short-time duty, intermittent duty, etc.), or a normal data distribution for a single steady state may be generated by combining the steady states for each mode. .

One or more features are obtained from the first sensed data ( 120 ). 3 (b) illustrates acquiring _{m 1} features from the first sensing data generated by the first IoT sensor. When the plurality of IoT sensors respectively generate sensing data, a process of acquiring one or more features for each first sensing data generated by the plurality of IoT sensors is performed. Characteristics related to the sensing data may be set differently according to unique properties of the sensing data. For example, in the case of temperature or humidity, characteristics such as an absolute value, an average value, and a slope may be obtained. As another example, in the case of rotation frequency and vibration, characteristics such as absolute values, acceleration values, and fast Fourier transform values may be obtained. If the number of one or more features acquired from each of the plurality of sensing data is N, a plurality of N-dimensional normal data (hereinafter, N-dimensional normal data set) including the N features as components is generated.

A Gaussian Mixture Model (GMM) is applied to the N-dimensional normal data set (130). A mixture model is a model that assumes that sub-distributions exist in the overall distribution, and is a model that assumes that data is generated from a plurality of distributions having parameters. Among them, the most used GMM is a model that assumes that data are generated from k normal distributions (Gaussian). GMM can approximate the data distribution using the mean, variance, and weight of each normal distribution as parameters.

는

번째 정규 분포의 평균, 분산, 가중치를 나타낸다. 수학식 1과 같은 방법으로, N차원 정상 데이터 셋에 대해 GMM을 적용하면, 확률밀도 값이 같은 점끼리 연결한 등고선(contour)을 경계로 하는 N차원 정상 데이터 분포가 생성된다. 도 4의 (a)는 N차원 정상 데이터 분포를 나타낸다. 정상 상태에서 획득된 복수의 특징들로 구성된 N차원 정상 데이터 셋(210)은, 정상 영역(220) 내부에 위치한다. 정상 영역(220)은, 경계(200)에 의해 비정상 영역으로부터 구분된다.here,

Is

Indicates the mean, variance, and weight of the second normal distribution. When the GMM is applied to the N-dimensional normal data set in the same way as in Equation 1, an N-dimensional normal data distribution is generated with a boundary line connecting points having the same probability density value. 4A shows an N-dimensional normal data distribution. The N-dimensional normal data set 210 including a plurality of features acquired in a steady state is located inside the normal region 220 . The normal region 220 is separated from the abnormal region by the boundary 200 .

Meanwhile, N (N-1)-dimensional normal data sets are generated using the N-dimensional normal data sets (140). The (N-1)-dimensional normal data set may be generated by sequentially removing the n-th feature (1≤n≤N) from the N-dimensional normal data set. For example, the first (N-1)-dimensional normal data set is generated by removing only the first feature from the N-dimensional normal data set, and the second (N-1)-dimensional normal data set is the N-dimensional normal data set. It can be created by removing only the second feature from

GMM is applied to each of the N (N-1)-dimensional normal data sets to generate N (N-1)-dimensional stationary data distributions (150). Since the (N-1)-dimensional normal data set is not affected by the removed features, the normal data distribution generated from the (N-1)-dimensional normal data set also does not reflect the removed features. When one N-dimensional normal data distribution and N (N-1)-dimensional normal data distributions are generated, an anomaly generated in the monitoring target is detected using the second sensing data measured after the steady state. It is determined whether the N features (hereinafter, N-dimensional monitoring data) obtained from the second sensing data measured after the steady state are located within the boundary of the N-dimensional normal data distribution ( 160 ). As illustrated in (a) of FIG. 5 , when the abnormal point 230 occurs, it is determined that the monitoring target is operating abnormally. However, unlike those illustrated in FIGS. 4 and 5 , the N dimension is a space that cannot be visually perceived by humans, and cannot be visually expressed on a monitor or the like. Therefore, the presence or absence of an abnormal point may be output in the form of an alarm or warning message by a monitoring system through IoT data analysis.

If an outlier is detected in the monitoring through the N-dimensional normal data distribution, a characteristic causing the outlier is determined using the (N-1)-dimensional normal data distribution ( 170 ). N pieces of (N-1)-dimensional monitoring data in which one of the N features included in the N-dimensional monitoring data is removed are generated. The N pieces of (N-1)-dimensional monitoring data are 1:1 matched to the (N-1)-dimensional normal data distribution that does not reflect the removed features, and the presence or absence of an outlier is determined. 5B exemplarily shows a case in which an outlier occurs due to the third characteristic. A first (N-1)-dimensional normal data distribution 2201 from which the first feature is removed, a second (N-1)-dimensional normal data distribution 2202 from which the second feature is removed, and a second (N-1)-dimensional normal data distribution from which the N-th feature is removed. In the case of the N (N-1)-dimensional normal data distribution 220N, all of the outliers 230 are detected because they are affected by the third feature that caused the outliers. On the other hand, in the case of the third (N-1)-dimensional normal data distribution 2203 from which the third feature is removed, since the third feature causing the outlier is removed, the outlier does not occur.

6 is a diagram exemplarily illustrating a monitoring system through IoT data analysis for implementing the monitoring method illustrated in FIGS. 1 to 5 .

Referring to FIG. 6 , the monitoring system through IoT data analysis may include a sensing data collection unit 300 , a feature acquisition unit 310 , a steady state model generation unit 320 , and a monitoring unit 330 . . A monitoring system through IoT data analysis may be implemented by a program stored in one or more processors and a storage device (eg, memory) communicatively connected thereto, and FIG. 6 is a functional representation of the processor executing the program. will be.

The sensing data collection unit 300 collects first and second sensing data from a plurality of IoT sensors through a communication network. The communication network may be a wired or wireless communication network. The collected sensing data may be stored in a memory.

The feature acquisition unit 310 acquires one or more features from the collected first and second sensing data. The feature may be obtained directly from the first and second sensing data (eg, an absolute value, etc.) or may be obtained through calculation (eg, a fast Fourier transform value, etc.). When a plurality of IoT sensors generate sensing data, one or more characteristics may be obtained for each sensing data. The acquired N features may be stored in the memory in the form of an N-dimensional normal data set.

The steady-state model generator 320 generates one N-dimensional normal data distribution and N (N-1)-dimensional normal data distributions by using the N-dimensional stationary data set. The N-dimensional normal data distribution is generated by applying GMM to a plurality of N-dimensional normal data sets, and the (N-1)-dimensional normal data distribution is generated by applying GMM to a plurality of (N-1)-dimensional normal data sets. do. Here, the (N-1)-dimensional normal data set may be generated by removing the n-th feature (1≤n≤N) from the N-dimensional normal data set.

The monitoring unit 330 detects the occurrence of the abnormal point and determines the characteristic that caused the abnormal point. The monitoring unit 330 determines whether the N-dimensional monitoring data obtained from the second sensing data input after the normal state is included in the N-dimensional normal data distribution. If the N-dimensional monitoring data is not included in the N-dimensional normal data distribution, the monitoring unit 330 generates N pieces of (N-1)-dimensional monitoring data using the N-dimensional monitoring data. The (N-1)-dimensional monitoring data is generated by removing the n-th feature from the N-dimensional monitoring data. The monitoring unit 320 determines that an anomaly has occurred due to the n-th characteristic when the n-th (N-1)-dimensional monitoring data is included in the normal region of the n-th (N-1)-dimensional normal data distribution.

It goes without saying that the monitoring method through IoT data analysis according to the present embodiment may be performed as an automated procedure according to a time-series sequence by a program built-in or installed in the digital processing device. Codes and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the program is stored in a computer readable media readable by a digital processing device, and is read and executed by the digital processing device to implement the method. The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

The above description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

200: Boundary 210: N-dimensional normal data set
220: normal region 230: outlier
300: sensing data collection unit 310: feature acquisition unit
320: steady state model generation unit 330: monitoring unit

Claims (11)

generating an N-dimensional normal data set having N features acquired during a steady state;
generating one N-dimensional normal data distribution and N (N-1)-dimensional normal data distributions using the N-dimensional normal data set;
generating N-dimensional monitoring data having N features acquired after the steady state; and
It is determined whether the N-dimensional monitoring data is an outlier using the N-dimensional normal data distribution, and the characteristic that caused the outlier using the N (N-1)-dimensional normal data distribution is identified in the N-dimensional monitoring data. detecting,
It is determined whether the N-dimensional monitoring data is an outlier using the N-dimensional normal data distribution, and the characteristic that caused the outlier using the N (N-1)-dimensional normal data distribution is identified in the N-dimensional monitoring data. The detection step is
determining whether the N-dimensional monitoring data is included in the N-dimensional normal data distribution;
if not included in the N-dimensional normal data distribution, generating N (N-1)-dimensional monitoring features by removing an n-th feature (1≤n≤N) from the N-dimensional monitoring data; and
and determining whether the N (N-1)-dimensional monitoring features are included in each of the N (N-1)-dimensional normal data distributions.
According to claim 1,
The N-dimensional normal data distribution is generated by applying a Gaussian Mixture Model (GMM) to the N-dimensional normal data set, a monitoring method through IoT data analysis.
According to claim 1,
The N (N-1)-dimensional normal data distribution is,
Remove the n-th feature (1≤n≤N) from the N-dimensional normal data set to generate N (N-1)-dimensional normal data sets,
A monitoring method through IoT data analysis for generating by applying a Gaussian Mixture Model (GMM) to the N (N-1) dimensional normal data set.
According to claim 1,
The step of generating an N-dimensional normal data set having N features acquired during the steady state comprises:
receiving k pieces of first sensing data from an IoT sensor installed in a monitoring target during the steady state period; and
A monitoring method through IoT data analysis, comprising generating an N-dimensional normal data set with N features associated with each of the k pieces of first sensing data.
5. The method of claim 4,
The step of generating N-dimensional monitoring data having N features obtained after the steady state comprises:
receiving k pieces of second sensing data from the IoT sensor after the normal state; and
A monitoring method through IoT data analysis, comprising generating N-dimensional monitoring data with N features associated with each of the k pieces of second sensing data.
delete 6. The method of claim 5,
The characteristic is any one of an absolute value, a slope, and a fast Fourier transform value of the first sensed data or the second sensed data, a monitoring method through IoT data analysis.
a sensing data collection unit for collecting k pieces of first sensing data from the IoT sensors installed in the monitoring target during the steady state period, and collecting k pieces of second sensing data after the steady state period;
a feature acquisition unit acquiring N features from the first sensing data and the second sensing data to generate an N-dimensional normal data set and N-dimensional monitoring data;
a steady-state model generator for generating one N-dimensional normal data distribution and N (N-1)-dimensional normal data distributions using the N-dimensional normal data set; and
A monitoring unit for determining whether the N-dimensional monitoring data is an outlier by using the N-dimensional normal data distribution and the N (N-1)-dimensional normal data distribution,
The monitoring unit,
It is determined whether the N-dimensional monitoring data is included in the N-dimensional normal data distribution,
If it is not included in the N-dimensional normal data distribution, the n-th feature (1≤n≤N) is removed from the N-dimensional monitoring data to generate N (N-1)-dimensional monitoring data,
A monitoring system through IoT data analysis to determine whether the N pieces of (N-1)-dimensional monitoring data are included in each of the N pieces of (N-1)-dimensional normal data distribution.
9. The method of claim 8,
The steady-state model generation unit,
The N-dimensional normal data distribution is generated by applying a Gaussian Mixture Model (GMM) to the N-dimensional normal data set,
Remove the n-th feature (1≤n≤N) from the N-dimensional normal data set to generate N (N-1)-dimensional normal data sets,
A monitoring system through IoT data analysis that generates the distribution of the N (N-1) dimensional normal data by applying GMM to the N (N-1) dimensional normal data set.
delete 9. The method of claim 8,
The characteristic is any one of an absolute value, a slope, and a fast Fourier transform value of the first sensed data or the second sensed data, a monitoring system through IoT data analysis.

Images