JPH07159289A - Method for diagnostic cause of abnormal phenomenon - Google Patents
Method for diagnostic cause of abnormal phenomenonInfo
- Publication number
- JPH07159289A JPH07159289A JP5339008A JP33900893A JPH07159289A JP H07159289 A JPH07159289 A JP H07159289A JP 5339008 A JP5339008 A JP 5339008A JP 33900893 A JP33900893 A JP 33900893A JP H07159289 A JPH07159289 A JP H07159289A
- Authority
- JP
- Japan
- Prior art keywords
- cause
- variable
- state
- influence coefficient
- state variable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Landscapes
- Testing And Monitoring For Control Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
Description
【０００１】[0001]
【産業上の利用分野】本発明は、複数の原因群からなる
異常現象に対し、その原因を診断するための異常現象の
原因診断方法に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of diagnosing the cause of an abnormal phenomenon consisting of a plurality of cause groups, for diagnosing the cause.
【０００２】[0002]
【従来の技術】設備の状態、製品品質状態などを常時監
視することは、安定した生産状態を維持するための重要
な要素である。監視された情報から異常の発生を予知す
ると共に原因を識別し、異常が初期の段階で適切なアク
ションをとり、大きな故障を生じない内に対応がとれる
ようにすることが要求される。2. Description of the Related Art Constant monitoring of equipment conditions, product quality conditions, etc. is an important factor for maintaining stable production conditions. It is required to predict the occurrence of an abnormality from the monitored information, identify the cause, and take appropriate action at an early stage so that the abnormality can be dealt with without causing a major failure.
【０００３】そこで、設備・品質の状態変数（設備状
態、操業状態、品質状態）、または状態量を変化させる
複数の原因変数を同時に常時測定し、それらの変数の現
在量自体が許容範囲を越えているかどうかを監視するこ
とにより、劣化程度を診断する提案がなされている（特
開昭５９−９４０１８号）。Therefore, equipment / quality state variables (equipment state, operation state, quality state) or a plurality of causal variables that change the state quantity are constantly measured at the same time, and the present amount itself of these variables exceeds the allowable range. A proposal has been made for diagnosing the degree of deterioration by monitoring whether or not it is present (JP-A-59-94018).
【０００４】例えば、回転機械（例えば、モータ、ブロ
ア等）の診断においては、正常時の振動振幅と現状の振
動振幅との差異から劣化程度を判定し、原因を探るため
に、周波数分析や次数比分析を行い、劣化に関連する特
定スペクトル成分量を監視することにより（どのスペク
トルが一番大きくなったか）、劣化原因を特定してい
る。For example, in diagnosing a rotating machine (for example, a motor, a blower, etc.), a frequency analysis or order is performed to determine the degree of deterioration from the difference between the vibration amplitude under normal conditions and the current vibration amplitude, and to find the cause. The cause of the deterioration is specified by performing a ratio analysis and monitoring the amount of a specific spectrum component related to the deterioration (which spectrum is the largest).
【０００５】[0005]
【発明が解決しようとする課題】しかし、上記した従来
技術にあっては、条件変化により状態変数量そのものが
変化するのに対し、判定のための基準が最初に設定した
ままの一定値である。このため、見かけ上、状態変数量
のみが大きくなり、誤判定される可能性がある。例え
ば、設備の状態を例にとると、その状態変数は回転に伴
う振動、動作音などであるが、回転数変化などで状態変
数の通常値は序々に変化して行き、これに見合って修正
した判定基準が必要であるにもかかわらず、一定のまま
である。この為、状態変数量が正常の範囲内であって
も、異常の判定が出される場合がある。そこで、本発明
の目的は、条件変化により状態変数量が変化しても劣化
等の監視が可能な異常現象の原因診断方法を提供するこ
とにある。However, in the above-mentioned prior art, the state variable amount itself changes due to the change in conditions, but the criterion for determination is a constant value that is initially set. . Therefore, apparently only the state variable amount becomes large, and there is a possibility that an erroneous determination is made. For example, taking the state of equipment as an example, the state variables are vibrations and operation sounds associated with rotation, but the normal value of the state variable gradually changes due to changes in rotation speed, etc., and corrects accordingly. It remains constant despite the need for the criteria set out above. Therefore, even if the state variable amount is within the normal range, an abnormality may be determined. Therefore, it is an object of the present invention to provide a method for diagnosing the cause of an abnormal phenomenon that allows monitoring of deterioration and the like even if the amount of state variables changes due to changes in conditions.
【０００６】[0006]
【課題を解決するための手段】上記の目的を達成するた
めに、この発明は、稼働中の被診断機器または被診断設
備もしくは品質の劣化状態を示す状態変数及び劣化状態
を示す原因変数を測定し、これらに基づいて前記状態変
数に対する原因変数毎の影響係数を算出し、この影響係
数と正常時の影響係数とを比較してその変化割合が最大
の前記原因変数を異常原因として特定するようにしてい
る。In order to achieve the above-mentioned object, the present invention measures a state variable indicating a deteriorated state of a device under diagnosis or equipment under operation or quality and a causal variable indicating a deteriorated state. Then, based on these, the influence coefficient for each cause variable for the state variable is calculated, and this influence coefficient is compared with the influence coefficient at the normal time so that the cause variable having the largest change rate is specified as an abnormal cause. I have to.
【０００７】前記異常原因の特定は、正常時の影響係数
に基づいて状態変数の現在量を予測し、この予測と前記
状態変数の実測値の差に基づいて劣化度判定を行った後
に実行される。The identification of the cause of the abnormality is performed after predicting the current amount of the state variable based on the influence coefficient in the normal state, and determining the degree of deterioration based on the difference between this prediction and the actual measurement value of the state variable. It
【０００８】正確な影響係数を算出するために、前記状
態変数及び前記原因変数の各変数別にタイムラグ補正を
行うようにしている。In order to calculate an accurate influence coefficient, time lag correction is performed for each of the state variable and the cause variable.
【０００９】[0009]
【作用】上記した手段によれば、逐次重回帰分析法を用
いて状態変数を説明する最少の原因変数群が求められ
る。これら原因変数の状態変数に対する影響係数の変化
度が大きく変化したものに対して、異常原因が特定され
る。したがって、条件変化の影響を最小にした劣化監視
が可能になるほか、診断精度の向上及びローコスト化が
可能になる。According to the above-mentioned means, the minimum causal variable group that explains the state variables can be obtained by using the sequential multiple regression analysis method. The cause of the abnormality is specified for those in which the degree of change in the influence coefficient of these cause variables with respect to the state variable has changed significantly. Therefore, it is possible to monitor deterioration while minimizing the influence of change in conditions, improve diagnostic accuracy, and reduce costs.
【００１０】影響係数の変化度のみでは劣化の程度が考
慮されていない。そこで、過去（初期）と現在の関係を
反映させるべく、原因変数別の現在量と正常時の影響係
数から算出される予測値と状態変数の現在量との差を劣
化程度とし、異常か正常かの判別を行う。これにより、
劣化量を考慮した原因診断が可能になり、診断精度を向
上させることができる。例えば、品質情報を状態変数に
した場合、原因変数相互、原因変数と状態変数間であっ
ても、同時刻に変数が変化しない場合がある。そこで、
前記状態変数及び前記原因変数の各変数別にタイムラグ
補正を行うことにより、同じ範疇で評価することができ
るようになり、正確な影響係数の算出が可能になる。The degree of deterioration is not taken into consideration only by the degree of change of the influence coefficient. Therefore, to reflect the relationship between the past (initial) and the present, the difference between the current amount of each cause variable and the predicted value calculated from the influence coefficient in the normal state and the current amount of the state variable is set as the degree of deterioration, and abnormal or normal Is determined. This allows
The cause can be diagnosed in consideration of the deterioration amount, and the diagnostic accuracy can be improved. For example, when the quality information is used as a state variable, the variables may not change at the same time even between the cause variables and between the cause variables and the state variables. Therefore,
By performing the time lag correction for each variable of the state variable and the cause variable, it becomes possible to evaluate in the same category, and it becomes possible to accurately calculate the influence coefficient.
【００１１】[0011]
【実施例】以下，本発明の実施例について、図面を参照
しながら説明する。Embodiments of the present invention will be described below with reference to the drawings.
【００１２】図１は本発明による異常現象の原因診断方
法に基づく処理例を示すフローチャートである。なお、
以下においては、ステップを“Ｓ”で表している。FIG. 1 is a flow chart showing an example of processing based on the method of diagnosing the cause of an abnormal phenomenon according to the present invention. In addition,
In the following, the step is represented by "S".
【００１３】まず、状態変数（被診断機器が回転機であ
れば、振動、振幅等）Ｙと複数の原因変数Ｘ１，Ｘ２，
Ｘ３，・・・Ｘ ｎを測定する（Ｓ１０１）。これは状態
変数Ｙと複数の原因変数Ｘ１，Ｘ２，Ｘ３，・・・Ｘ ｎ
との因果関係を定量的に関係付けるために行うものであ
る。品質情報を状態変数とした場合、状態変数と原因変
数、または原因変数間でも同時刻に変数が変化しない場
合がある。正確な原因診断を行うためには、同時刻に条
件を合わせる必要がある。そこで、変数別にタイムラグ
補正を行う（Ｓ１０２）。First, state variables (the device under diagnosis is a rotating machine)
Then, vibration, amplitude, etc.) Y and a plurality of cause variables X1, X2,
X3, ... X n is measured (S101). This is the state
Variable Y and plural cause variables X1, X2, X3, ... X n
This is done to quantitatively relate the causal relationship with
It When quality information is used as a state variable,
If the variables do not change at the same time between the number or the cause variables,
There is a match. To make an accurate diagnosis,
It is necessary to match the cases. Therefore, time lag for each variable
Correction is performed (S102).
【００１４】ステップ１０２による状態変数Ｙ及び原因
変数Ｘに基づいて、状態変数量に対する原因変数別の影
響係数αを計算する（Ｓ１０３）。ここでは、正常状態
の状態変数を予測するため、必要な最小限の原因変数と
原因変数別の影響係数を逐次重回帰分析法により自動選
別し、選別された原因変数別の影響係数αを同時に算出
する。状態変数Ｙは次式で示される。Based on the state variable Y and the cause variable X in step 102, the influence coefficient α for each cause variable with respect to the state variable amount is calculated (S103). Here, in order to predict a state variable in a normal state, the minimum necessary causal variable and the influence coefficient for each causal variable are automatically selected by the sequential multiple regression analysis method, and the selected influence coefficient α for each causal variable is simultaneously calculated. calculate. The state variable Y is shown by the following equation.
【００１５】 （正常の場合） Ｙ＝（原因変数Ｘ1）α01＋（原因変数Ｘ2）α02＋（原因変数Ｘ1）α03 ・・・＋（原因変数Ｘn）α0n （異常の場合） Ｙ＝（原因変数Ｘ1） α11＋（原因変数Ｘ2）α12＋（原因変数Ｘ1） α13 ・・・＋（原因変数Ｘn） α1n 例えば、モータの場合、Ｙ＝（アンバランス） α01＋
（部品摩耗） α12・・・、で状態変数を表すことができ
る。(Normal case) Y = (Cause variable X1) Α01+ (Cause variable X2) Α02+ (Cause variable X1) Α03 ... + (Cause variable Xn) Α0n (In case of abnormality) Y = (Cause variable X1) α11+ (Cause variable X2) Α12+ (Cause variable X1) α13 ... + (Cause variable Xn) α1n For example, in the case of a motor, Y = (unbalance) α01+
(Part wear) α12・ ・ ・, Can represent the state variable
It
【００１６】ステップ１０３で選別された原因変数量を
常時測定し、正常時に算出された影響係数（α01，α02
・・・α0n）から状態変数の現在量ＹＹを常時予測し、
この予測値（ＹＹ）と状態変数と実績値（Ｙ）（現在
値）との差を劣化程度とし（Ｓ１０４）、この劣化程度
（差値）を予め設定した判定基準値と比較する（Ｓ１０
５）。The amount of causal variables selected in step 103 is constantly measured, and the influence coefficients (α 01 , α 02) calculated under normal conditions are calculated.
... The current amount YY of the state variable is constantly predicted from α 0n ),
The difference between the predicted value (YY) and the state variable and the actual value (Y) (current value) is set as the degree of deterioration (S104), and this degree of deterioration (difference value) is compared with a preset determination reference value (S10).
5).
【００１７】比較の結果、正常であればステップ１０１
に戻って以降の処理を繰り返し実行し、異常であれば、
状態変数Ｙと複数の原因変数Ｘ１，Ｘ２，・・・Ｘｎを
取り込んで変数別にタイムラグ補正を行い（Ｓ１０
６）、劣化状態下での状態変数に対する原因変数別の影
響係数を逐次重回帰分析により算出し（Ｓ１０７）、こ
れらの影響係数と正常時の影響係数とを比較して、その
変化割合が最大の変数を異常原因であると推定する（Ｓ
１０８）。If the result of comparison is normal, step 101
Return to and repeat the subsequent processing.
The state variable Y and a plurality of cause variables X1, X2, ... Xn are taken in and time lag correction is performed for each variable (S10
6), the influence coefficient for each cause variable with respect to the state variable under the deteriorated state is calculated by sequential multiple regression analysis (S107), and these influence coefficients are compared with the influence coefficient at the normal time, and the change rate is maximum. Is assumed to be the cause of the abnormality (S
108).
【００１８】[0018]
【発明の効果】本発明は上記の通り構成されているの
で、次に記載する効果を奏する。Since the present invention is configured as described above, it has the following effects.
【００１９】請求項１の異常現象の原因診断方法によれ
ば、稼働中の被診断機器または被診断設備もしくは品質
の劣化状態を示す状態変数及び劣化状態を示す原因変数
を測定し、これらに基づいて前記状態変数に対する原因
変数毎の影響係数を算出し、この影響係数が最も変化し
た前記原因変数を異常原因として特定するようにしたの
で、条件変化の影響を低減した劣化監視が可能になると
共に診断精度の向上が可能になる。また、従来、経験に
依存していた部分が定量的解析が可能になることで、論
理的な診断ロジックの構築が可能になり、ローコスト化
を図ることができる。According to the method for diagnosing the cause of an abnormal phenomenon according to claim 1, a state variable indicating a deteriorated state of a device or a facility to be diagnosed or quality in operation and a causal variable indicating a deteriorated state are measured, and based on these, By calculating the influence coefficient for each of the cause variables for the state variable and identifying the cause variable with the largest change in the influence coefficient as the cause of abnormality, it is possible to perform deterioration monitoring while reducing the influence of the condition change. The diagnostic accuracy can be improved. In addition, since a part that has conventionally depended on experience can be quantitatively analyzed, it is possible to construct a logical diagnosis logic and reduce cost.
【００２０】請求項２の異常現象の原因診断方法によれ
ば、正常時の影響係数に基づいて状態変数の現在量を予
測し、この予測値と前記状態変数の実施値の差に基づい
て劣化判定を行うようにしたので、劣化量を考慮した原
因診断が可能になり、診断精度を向上させることができ
る。According to the method of diagnosing the cause of the abnormal phenomenon of claim 2, the present amount of the state variable is predicted based on the influence coefficient in the normal state, and the deterioration is caused based on the difference between the predicted value and the actual value of the state variable. Since the determination is made, it is possible to diagnose the cause in consideration of the amount of deterioration and improve the diagnostic accuracy.
【００２１】請求項３の異常現象の原因診断方法によれ
ば、前記状態変数及び前記原因変数の各変数別にタイム
ラグ補正を行うようにしたので、正確な影響係数の算出
が可能になる。According to the method of diagnosing the cause of the abnormal phenomenon of claim 3, since the time lag correction is performed for each of the state variable and the cause variable, it is possible to accurately calculate the influence coefficient.
【図１】本発明による異常現象の原因診断方法に基づく
処理例を示すフローチャートである。FIG. 1 is a flowchart showing a processing example based on a method of diagnosing the cause of an abnormal phenomenon according to the present invention.
Claims (3)
しくは品質の劣化状態を示す状態変数及び劣化状態を示
す原因変数を測定し、これらに基づいて前記状態変数に
対する原因変数毎の影響係数を算出し、この影響係数と
正常時の影響係数とを比較してその変化割合が最大の前
記原因変数を異常原因として特定することを特徴とする
異常現象の原因診断方法。1. A state variable indicating a deteriorated state of a device to be diagnosed or equipment to be diagnosed or quality in operation and a causal variable indicating a deteriorated state are measured, and an influence coefficient for each causal variable to the state variable is measured based on these. A method for diagnosing a cause of an abnormal phenomenon, which comprises calculating and comparing the influence coefficient and an influence coefficient in a normal state to identify the cause variable having the largest change rate as an abnormality cause.
に基づいて状態変数の現在量を予測し、この予測値と前
記状態変数の実測値の差に基づいて劣化度判定を行うこ
とを特徴とする請求項１記載の異常現象の原因診断方
法。2. The normal amount of the state variable is predicted based on the influence coefficient calculated in claim 1 in a normal state, and the deterioration degree determination is performed based on the difference between the predicted value and the actual measured value of the state variable. 2. The method for diagnosing the cause of an abnormal phenomenon according to claim 1.
別にタイムラグ補正を行うことを特徴とする請求項１ま
たは２記載の異常現象の原因診断方法。3. The method for diagnosing the cause of an abnormal phenomenon according to claim 1, wherein time lag correction is performed for each of the state variable and the cause variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5339008A JPH07159289A (en) | 1993-12-03 | 1993-12-03 | Method for diagnostic cause of abnormal phenomenon |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP5339008A JPH07159289A (en) | 1993-12-03 | 1993-12-03 | Method for diagnostic cause of abnormal phenomenon |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH07159289A true JPH07159289A (en) | 1995-06-23 |
Family
ID=18323403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP5339008A Withdrawn JPH07159289A (en) | 1993-12-03 | 1993-12-03 | Method for diagnostic cause of abnormal phenomenon |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH07159289A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002511962A (en) * | 1997-06-11 | 2002-04-16 | ウェスチングハウス・エレクトリック・カンパニー・エルエルシー | Security or protection system utilizing reflective memory and / or another type of processor and communication |
| JP2006350698A (en) * | 2005-06-16 | 2006-12-28 | Taiyo Nippon Sanso Corp | Abnormality diagnostic device, abnormality diagnostic method, and abnormality diagnostic program |
| JP2007156653A (en) * | 2005-12-01 | 2007-06-21 | Kurita Water Ind Ltd | Operation management method and device for water treatment plant |
| WO2016088362A1 (en) * | 2014-12-05 | 2016-06-09 | 日本電気株式会社 | System analyzing device, system analyzing method and storage medium |
| WO2016129218A1 (en) * | 2015-02-09 | 2016-08-18 | 日本電気株式会社 | Display system for displaying analytical information, method, and program |
| US11566967B2 (en) | 2018-11-12 | 2023-01-31 | Nippon Paper Industries Co., Ltd. | Abnormality detection device and abnormality detection method |
-
1993
- 1993-12-03 JP JP5339008A patent/JPH07159289A/en not_active Withdrawn
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002511962A (en) * | 1997-06-11 | 2002-04-16 | ウェスチングハウス・エレクトリック・カンパニー・エルエルシー | Security or protection system utilizing reflective memory and / or another type of processor and communication |
| JP2006350698A (en) * | 2005-06-16 | 2006-12-28 | Taiyo Nippon Sanso Corp | Abnormality diagnostic device, abnormality diagnostic method, and abnormality diagnostic program |
| JP2007156653A (en) * | 2005-12-01 | 2007-06-21 | Kurita Water Ind Ltd | Operation management method and device for water treatment plant |
| JP4604987B2 (en) * | 2005-12-01 | 2011-01-05 | 栗田工業株式会社 | Operation management method and apparatus for water treatment plant |
| WO2016088362A1 (en) * | 2014-12-05 | 2016-06-09 | 日本電気株式会社 | System analyzing device, system analyzing method and storage medium |
| JPWO2016088362A1 (en) * | 2014-12-05 | 2017-09-07 | 日本電気株式会社 | System analysis apparatus, system analysis method, and storage medium |
| US10719577B2 (en) | 2014-12-05 | 2020-07-21 | Nec Corporation | System analyzing device, system analyzing method and storage medium |
| WO2016129218A1 (en) * | 2015-02-09 | 2016-08-18 | 日本電気株式会社 | Display system for displaying analytical information, method, and program |
| JPWO2016129218A1 (en) * | 2015-02-09 | 2017-11-16 | 日本電気株式会社 | Information display system, method and program for analysis |
| US11566967B2 (en) | 2018-11-12 | 2023-01-31 | Nippon Paper Industries Co., Ltd. | Abnormality detection device and abnormality detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109073508B (en) | Fault diagnosis during testing of turbine units | |
| US20060224367A1 (en) | Inspection apparatus, aid device for creating judgement model therefor, abnormality detection device for endurance test apparatus and endurance test method | |
| JP4706316B2 (en) | Tire testing apparatus and tire testing method | |
| CN106052847B (en) | Monitoring of a machine having a rotating machine component | |
| US6763482B2 (en) | Printer diagnostics method | |
| WO2014123443A1 (en) | Method and device for vibration diagnosis and forecasting sudden engine failure | |
| KR101498527B1 (en) | Diagnosis system using vibration frequency analysis program for rotation equipment of power plant | |
| CN111538655A (en) | Software failure detection method, system, storage medium, computer program, and terminal | |
| JP2011191249A (en) | Monitoring method in balance test and dynamic balance measuring device | |
| JPH07159289A (en) | Method for diagnostic cause of abnormal phenomenon | |
| US5710715A (en) | Vibration analysis method | |
| JPH10288546A (en) | Abnormality judging method for rotating apparatus | |
| JP3108405B2 (en) | Device diagnosis method | |
| JP2002214185A (en) | Detecting method and detecting device for sensor abnormality | |
| JP2002090266A (en) | Remaining life-predicting device | |
| CN115855165B (en) | Multi-dimensional precision point inspection method and system for thermal power equipment | |
| JP3102498B2 (en) | Test method and test station for performing the test method | |
| TW202120949A (en) | Diagnostic device and diagnostic method | |
| CN113551764B (en) | Vibration analysis apparatus and method thereof | |
| JP2004020484A (en) | Abnormality monitoring device and program for monitoring abnormality | |
| JP2006300524A (en) | Abnormal sound inspection method | |
| JP3223623B2 (en) | Abnormal inspection equipment for rotating machines | |
| JPH10333743A (en) | Method and device for simple abnormality diagnosis of facility accompanied by load variation | |
| JPH0450731A (en) | Rotary machine fault diagnostic system | |
| JPH0399234A (en) | Diagnosing method for abnormality of rotary machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A300 | Withdrawal of application because of no request for examination |
Free format text: JAPANESE INTERMEDIATE CODE: A300 Effective date: 20010206 |