JPS6019207A - Monitor device for control system - Google Patents

Monitor device for control system

Info

Publication number
JPS6019207A
JPS6019207A JP58126563A JP12656383A JPS6019207A JP S6019207 A JPS6019207 A JP S6019207A JP 58126563 A JP58126563 A JP 58126563A JP 12656383 A JP12656383 A JP 12656383A JP S6019207 A JPS6019207 A JP S6019207A
Authority
JP
Japan
Prior art keywords
control system
neutral point
control
gain
regression coefficient
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.)
Pending
Application number
JP58126563A
Other languages
Japanese (ja)
Inventor
Yasunobu Hayama
葉山　安信
Sadamu Terado
寺戸　定
Mitsuhiro Abe
阿部　光博
Makoto Watanabe
誠 渡辺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mitsubishi Heavy Industries Ltd
Original Assignee
Mitsubishi Heavy Industries Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Mitsubishi Heavy Industries Ltd filed Critical Mitsubishi Heavy Industries Ltd
Priority to JP58126563A priority Critical patent/JPS6019207A/en
Publication of JPS6019207A publication Critical patent/JPS6019207A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
    • G05B23/024—Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0267—Fault communication, e.g. human machine interface [HMI]
    • G05B23/0272—Presentation of monitored results, e.g. selection of status reports to be displayed; Filtering information to the user

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Human Computer Interaction (AREA)
  • Mathematical Physics (AREA)
  • Indication Of The Valve Opening Or Closing Status (AREA)
  • Fluid-Pressure Circuits (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
  • Testing And Monitoring For Control Systems (AREA)
  • Servomotors (AREA)

Abstract

PURPOSE:To detect a fault or a defect with high sensitivity by monitoring at all times the regression coefficients corresponding to the control neutral point and the control system gain respectively and monitoring these coefficients. CONSTITUTION:A regression coefficient (a) corresponding to the control neutral point calculated by a regression coefficient operator 23 is supplied to a neutral point fault detector 25. If the coefficient (a) gets out of a prescribed neutral point range of -L1-L1, an occurrence of a fault is displayed to a neutral point display device 27. While a regression line (b) corresponding to the control system gain calculated by the operator 23 is supplied to a gain fault detector 26. If this line (b) gets out of a prescribed gain range L2-L3, the detector 26 functions to display a fault. In this case, both coefficient (a) and line (b) are always recorded to a recorder 29. Thus even a small fault is never overlooked despite the monitor for a control system having high response performance. As a result, a fault or a defect can be detected with high sensitivity.

Description

【発明の詳細な説明】 この発明は、例えば油圧制御系の異常動作および故障等
を高感度に検出する制御系監視装置に関する。DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a control system monitoring device that detects, for example, abnormal operation and failure of a hydraulic control system with high sensitivity.

例えに油圧制御系における制御弁の故障および油漏れ等
を検出するには、第１図に示すような故障検出装置１１
が使用される。この故障検出装置１１は、制御系１２０
制御目標値Ｘ？入力し、制御系モデル１３によって制御
系１２の出力予測値’ｆａを算出するもので、この予測
値”（ａと実際の制御系１２の出力値Ｙとから制御偏差
Ｅ（＝ＹＪＩ−Ｙｉ求める。次に、故障利足装置１４に
よって上記制御偏差Ｅが予め設足される偏差許容範囲内
であるか否かｔ利足するもので、この場合、制御偏差Ｅ
が許容範囲全逸脱すると、制御系１２は故障であると利
足され表示装置１５に表示されるようになる。１−なわ
ち、この故障検出装置ｊｌｌｌは、制御系モデル１３と
実際の制御系１２とに、それぞれ実時間で同じ人力傷号
金与え、そして、その時の歯刃信号の差？監視すること
により、制御系１２の故障會検出するものである。この
場合、一般に、上記のような演算処理および利足処理等
は、ディジタルコンピュータで行なわれるものである。For example, in order to detect a control valve failure, oil leak, etc. in a hydraulic control system, a failure detection device 11 as shown in FIG.
is used. This failure detection device 11 includes a control system 120
Control target value X? input, and calculates the predicted output value 'fa of the control system 12 using the control system model 13. From this predicted value '(a) and the actual output value Y of the control system 12, the control deviation E (= YJI - Yi .Next, the failure counting device 14 checks whether the control deviation E is within a deviation tolerance range set in advance.In this case, the control deviation E
If the error exceeds the allowable range, the control system 12 is recognized as having a failure and is displayed on the display device 15. 1-In other words, this failure detection device jllll gives the same human power signal to the control system model 13 and the actual control system 12 in real time, and then determines the difference between the tooth blade signals at that time. By monitoring, a failure of the control system 12 is detected. In this case, the above-mentioned arithmetic processing, foot-scanning processing, etc. are generally performed by a digital computer.

ここで、このような故障検出装置１１ｋ、例えば、油圧
制御系のような数１０　（Ｈｚ）〜数１００（七）の高
速応答性を有する制御系１２に対して使用する場合には
、ディジタルコンピュータによる演算処理時間全高速化
しなければならないのて、制御系モデル１３を簡略化し
て、その実質的な計算時間？短縮しなければならない・
しかし、このように制御系モデル１３ケ簡略化すると、
このモデル１３自体の精度が低下する状態となｐ、実際
の制御系１２との特性誤差が大きくなってしまう。した
がって−制御系１２の正常時においても、制御偏差Ｅが
大きい値となるため、予め故障判定装置１４の許容範囲
１広く設足しなけれはならない。この場合、制御偏差Ｅ
が非常に大きくなるような制御系１２の重故障は検出可
能であるが、逆に、制御偏差Ｅが小壜い軽故障は検出す
ることができない。Here, when such a failure detection device 11k is used for a control system 12 having a high-speed response of several tens (Hz) to several hundreds (seven), such as a hydraulic control system, a digital computer is used. Since the calculation processing time must be completely accelerated, the control system model 13 is simplified to reduce its actual calculation time. Must be shortened.
However, when the 13 control system models are simplified in this way,
The accuracy of this model 13 itself is reduced, and the characteristic error with the actual control system 12 becomes large. Therefore, even when the control system 12 is normal, the control deviation E has a large value, so the tolerance range 1 of the failure determination device 14 must be set wide in advance. In this case, the control deviation E
Although it is possible to detect a major failure in the control system 12 where E is extremely large, it is not possible to detect a minor failure where the control deviation E is small.

この発明は、上記のような問題点に鑑みなされたもので
、例えば油圧制御系のように、高速応答性′Ｉｆ：有す
る制御系に対して使用する場合でも、軽故障を見逃丁こ
となく、感度の高い故障検出を行うことができるように
なる制御系監視装置全提供すること？目的とする。This invention was made in view of the above-mentioned problems, and even when used in a control system that has high-speed response, such as a hydraulic control system, it can be used without overlooking minor failures. , to provide complete control system monitoring equipment that will be able to perform highly sensitive fault detection? purpose.

すなわち、この発明に係る制御系監視装置は、制御系の
制御偏差信号とフィードバック信号と會それぞれ一足時
間間隔毎に抽出する抽出器と、この抽出器によυ抽出さ
れるフィードバック信号からフィードバック変化速度全
算出する微分器と、この微分器によシ算出されるフィー
ドバック変化速度と上記抽出器にＪニジ抽出される制御
偏差値とにより上記制御系の制御中立点と制御系利得と
に対応する回帰係数？算出する回帰係数演算器と、この
回帰係数演算器によＶ算出される上記制御中立点に対応
する回帰係数が予め設足される中立点範囲會逸脱したこ
とを検出する中立点異常検出器と、上記回帰係数演算器
により算出される制御系利得に対応する回帰係数が予め
設足される利得範囲全逸脱したことを検出する利得異常
検出器と全具備し、上記制御中立点と制御系利得とに対
応するそれぞれの回福係数の変イヒ會監視するようにし
たものである〇以下図面によｐこの発明の一実施例？説
明する。That is, the control system monitoring device according to the present invention includes an extractor that extracts a control deviation signal and a feedback signal of the control system at each time interval, and a feedback change rate from the feedback signal extracted by this extractor. A regression that corresponds to the control neutral point and control system gain of the control system is performed using a differentiator that calculates the total, a feedback change rate calculated by this differentiator, and a control deviation value extracted by the extractor. coefficient? a regression coefficient calculator for calculating; a neutral point abnormality detector for detecting that the regression coefficient corresponding to the control neutral point calculated by the regression coefficient calculator deviates from a neutral point range set in advance; , is fully equipped with a gain abnormality detector that detects when a regression coefficient corresponding to the control system gain calculated by the above-mentioned regression coefficient calculator deviates from the entire gain range set in advance, and the above-mentioned control neutral point and control system gain The following figure shows an embodiment of the present invention. explain.

第２図はこの監視装置？、例えば位置制御用油圧制御系
に使用する場合のブロック構成？示すもので、２０は位
置制御系、２ノはこの実施例による制御系監視装置であ
る。Is this the monitoring device in Figure 2? For example, what is the block configuration when used in a hydraulic control system for position control? In the figure, 20 is a position control system, and 2 is a control system monitoring device according to this embodiment.

まず、上記位置制御系２０は、偏差演算部２０ａ１増幅
部２０ｂ１サーボ弁２０Ｃ，シリンダ２０ｄおよびこの
シリンダ２９ｄＩ７）ｆｋ３刀側と上記偏差演算部２０
Ｂとの間に介在される位置検出器２０ｅにより構厄され
る〇 ここで、この実施例に↓＾監視装置２ノは、２つの信号
抽出器２２ａ、２２ｂ　ｗ＠え°Ｃいる。この信号抽出
器２２８，２２ｂは、上記位置制御系２０の偏差演算部
２０８から増幅！１ｌＯｂに供給される制御偏差信号Ｅ
と、シリング゛２０ｄの出刃側から位置検出器２０ｅ　
ｋ介して偏差演算部２０ａに帰還されるフィードバック
信号Ｙと會、そ扛ぞれ一足時間間隔毎電毎に同時抽出す
るもので、この抽出器２２Ｂ、２２ｂからの制御偏差値
ｅｉおよびフィードバック信号ｙｉｋそれぞれ回帰係数
演算器２３および微分器２４に供給する。この微分器２
４は抽出器２２ｂから供給さ扛るフィードバック信号ｙ
ｉからフィードバック変化速度−１？算出するもので、
この微分器２４からのフィードバック変化速度信号；１
を上記回帰係数演算器２３に供給する。First, the position control system 20 includes a deviation calculating section 20a1, an amplifying section 20b1, a servo valve 20C, a cylinder 20d and this cylinder 29dI7)fk3 sword side, and the deviation calculating section 20
Here, in this embodiment, the monitoring device 2 includes two signal extractors 22a and 22b w@E°C. The signal extractors 228 and 22b amplify the signal from the deviation calculation section 208 of the position control system 20! Control deviation signal E supplied to 1lOb
and the position detector 20e from the cutting edge side of the syringe 20d.
The control deviation value ei and the feedback signal yik from the extractors 22B and 22b are simultaneously extracted at each time interval. These are supplied to a regression coefficient calculator 23 and a differentiator 24, respectively. This differentiator 2
4 is the feedback signal y supplied from the extractor 22b.
Feedback change rate -1 from i? It is calculated by
Feedback change rate signal from this differentiator 24; 1
is supplied to the regression coefficient calculator 23.

この回帰係数演算器２３は、上記抽出器２２ａからの制
御偏差値ｅｉと、微分器２４からの７１一ドバツク変化
速度ｙｉとにより上記位置制御系２０の制御中立点と制
御系オリ得とに対応するそれぞれの回帰係数８およびｂ
２算出するもので、この回帰係数演算器２３からの回帰
係数ａおよびｂを、それぞれ中立点異常検出器２５およ
び利得異常検出器２６に供給する。この中立点異常検出
器２５は回帰係数演算器２３により算出される制御中立
点に対応する回帰係数ａが、予め設足される中立点範囲
（−Ｌ１〜Ｌ１）？逸脱したことを検出するもので、こ
の中立点異常検出器２５からの検出信号を中立点異常表
示器２７に供給して表示する。また、上記利得異常検出
器２６は回帰係数演算器２３により算出される制御系利
得すに対応する回帰係数が予め設足される利得範囲（Ｌ
２〜Ｌｓ）ｋ逸脱したこと？検出するもので、この利得
異常検出器２６からの検出信号全利得異常表示器２８に
供給して表示する。そして、上記回帰係数演算器２３か
らのそｎぞれの回帰係数８およびｂを、常時記録計２９
に供給してその経時変化全記録する。This regression coefficient calculator 23 corresponds to the control neutral point and control system original point of the position control system 20 using the control deviation value ei from the extractor 22a and the 71-back change rate yi from the differentiator 24. The respective regression coefficients 8 and b
The regression coefficients a and b from this regression coefficient calculator 23 are supplied to a neutral point abnormality detector 25 and a gain abnormality detector 26, respectively. This neutral point abnormality detector 25 is configured such that the regression coefficient a corresponding to the control neutral point calculated by the regression coefficient calculator 23 is a neutral point range (-L1 to L1) set in advance. This detects deviation from the neutral point abnormality detector 25, and supplies the detection signal from the neutral point abnormality detector 25 to the neutral point abnormality display 27 for display. The gain abnormality detector 26 also operates in a gain range (L
2~Ls) Have you deviated? The detection signal from the gain abnormality detector 26 is supplied to the total gain abnormality display 28 and displayed. Then, the regression coefficients 8 and b from the regression coefficient calculator 23 are recorded in a constant recorder 29.
and record all changes over time.

すなわち、このように構成される装置において、まず、
位置制御系２０の偏差演算部２０ｍには制御目標値Ｘ′
？を供給する。これにより、偏差演算部２０８は、制御
−目標値Ｘとフィードバック信号Ｙとにより制御偏差Ｅ
（＝Ｘ−ＹＪ　ｋ算出するもので、この制御偏差Ｅは増
幅部２０ｂで増幅されサーボ弁電流工となってサーボ９
Ｆ−２０Ｃに供給される。このサーボ弁２０Ｃは、サー
ボ弁電流■に比例した油の流量Ｑ會流入または流出し、
シリンダ２０ｄｆその流量Ｑに対応して駆動するもので
、このシリンダ２０ｄの移動付値°−は、位置検出器２
０ｅで検出されフィードバック信号Ｙとして上記偏差演
算部２０Ｂに帰還される。この場合、偏差演算部２０ａ
において、制御目標値Ｘとフィードバック信号Ｙとが等
しく（Ｘ＝Ｙ）なり、制御偏差Ｅ’＝０となれば、増幅
部２０ｂからのサーボ弁電流■およびサーボ弁２０Ｃか
らの流量Ｑは、それぞれＩ＝Ｏ，Ｑ＝０となりシリンダ
２０ｄの動作は停止するようになる。That is, in the device configured in this way, first,
The deviation calculation section 20m of the position control system 20 receives the control target value X'
? supply. As a result, the deviation calculation unit 208 calculates the control deviation E based on the control target value X and the feedback signal Y.
(=X-YJ k is calculated, and this control deviation E is amplified by the amplifying section 20b and becomes the servo valve current
Supplied to F-20C. This servo valve 20C allows an oil flow rate Q proportional to the servo valve current to flow in or out,
The cylinder 20df is driven in accordance with its flow rate Q, and the movement value °- of this cylinder 20d is determined by the position detector 2.
0e and fed back to the deviation calculation section 20B as a feedback signal Y. In this case, the deviation calculation unit 20a
If the control target value X and the feedback signal Y are equal (X=Y) and the control deviation E'=0, then the servo valve current ■ from the amplifier 20b and the flow rate Q from the servo valve 20C are respectively I=O, Q=0, and the operation of the cylinder 20d comes to stop.

ここで、サーボ増幅ｍ２０ｂを比例要素、サーボ弁２０
Ｃおよび位置検出器２０６　ｋ　１次遅れ要素、また、
シリンダ２０６ケ積分要素として近φすると、制御偏差
Ｅとフィードバック信号Ｙとの関係は次式のようになる
。Here, the servo amplification m20b is a proportional element, and the servo valve 20
C and position detector 206 k first-order delay element, and
When 206 cylinders are approximated to φ as integral elements, the relationship between the control deviation E and the feedback signal Y is as shown in the following equation.

但し、Ｓニラプラス演算子 Ｋｌ：サーボ増幅部２０ｂの利得 均：サーボ弁２０Ｃの利得 Ｋｓニジリンダ２０ｄの利得 Ｔ！＝サーボ５Ｆ　２０Ｇの時定数 Ｔ２：位置検出器２０ｅの時定数 である。However, S nila plus operator Kl: Gain of servo amplifier 20b Average: gain of servo valve 20C Gain of Ks Nijilinda 20d T! = Servo 5F 20G time constant T2: Time constant of position detector 20e It is.

ここで、この実施例による監視装置２１の２つの抽出器
２２Ｂおよび２２ｂによシ、それぞれ位置制御系２００
制御偏差信号Ｅおよびフィードバック信号Ｙを、一定時
間間隔Δ電毎に抽出する。この場合、抽出器２２１４．
２２ｂにおける一定抽出時間間隔Δｔ？、制御系２０の
サーボ弁２０Ｃお工び位置検出器２０ｅそれぞれの時定
数Ｔ＋およびＴ２に対して数倍長く設足する。これによ
シ、サーボ弁２０Ｃおよび位置検出器２０６の時間遅れ
要素會無視することができ、上記１式は次式のように近
似されるよりになる。Here, the two extractors 22B and 22b of the monitoring device 21 according to this embodiment have a position control system 200, respectively.
A control deviation signal E and a feedback signal Y are extracted at regular time intervals Δ. In this case, extractor 2214.
Constant extraction time interval Δt at 22b? , the time constants T+ and T2 of the servo valve 20C and the position detector 20e of the control system 20 are set several times longer. As a result, the time delay elements of the servo valve 20C and the position detector 206 can be ignored, and the above equation 1 can be approximated as the following equation.

Ｓ　−Ｙ（Ｓ）中に−Ｅ（Ｓ）　・・・・・・・・・２
式但し、Ｋ＝Ｋｔ’に！・に３であｐ１制御系２０のル
ーグゲイン（利得）會表わす。-E(S) in S -Y(S) ・・・・・・・・・2
However, K=Kt'!・3 represents the Roug gain of the p1 control system 20.

次に、抽出器２２Ｂからの制御偏差値Ｊおよび抽出器２
２ｂから微分器２４全介して出力さ扛るフィードバック
変化引０ｉは、それぞ扛回帰係数演算器２３に供給さ牡
、上記制御系２θの制御中立点と制御系利得とに対応す
る、そ扛ぞ扛ａ＝　・・・・・・４式 ｂ＝−一−−−−−−−−−−−−・・・・・・５式ｎ
Σｅｉ′１−（Σｅ　ｒ　）２ １霞１　ｉツー ０、を 但し、この回帰係数８およびｂは、Σ（ｙｉ−Ｙｉ）”
の１厘１ 計算値が最小になるように決足される。Next, the control deviation value J from the extractor 22B and the extractor 2
The feedback changes 0i outputted from 2b through the differentiator 24 are respectively supplied to the regression coefficient calculator 23, and the feedback changes 0i corresponding to the control neutral point and control system gain of the control system 2θ are zo扛a=...4 formula b=-1-----------5 formula n
Σei′1−(Σe r )2 1 Kasumi 1 i20, however, this regression coefficient 8 and b are Σ(yi−Yi)”
1 厘1 The calculated value is determined to be the minimum.

このように、回帰係数演算器２３により算出した制御中
立点に対応する回帰係数ａは、中立点異常検出器２５に
供給さｎ１予め設定される中立点範囲（−Ｌｉ−Ｌ＋　
３　ｋ逸脱した場合に中立点異常表示器２７は異常表示
動作するようになる、また、上記回帰係数演算器２３に
よυｐ−出した制御系利得に対応する回帰係数すは、利
得異常検出器２６に供給され、予め設定される利得範囲
（Ｌ２〜Ｌ３）ｋ逸脱した場合に利得異常検出器２８１
４異常表示動作するようになる。ここで、回帰係数ａお
よびｂは、常時記録計２９ＶＣ記録される。In this way, the regression coefficient a corresponding to the control neutral point calculated by the regression coefficient calculator 23 is supplied to the neutral point abnormality detector 25, and the regression coefficient a corresponding to the control neutral point calculated by the regression coefficient calculator 23 is
3 k deviation, the neutral point abnormality indicator 27 will operate to display an abnormality, and the regression coefficient corresponding to the control system gain υp- output by the regression coefficient calculator 23 will be detected by the gain abnormality detector. 26, and if the preset gain range (L2 to L3) k deviates from the gain abnormality detector 281.
4 Abnormal display starts to work. Here, the regression coefficients a and b are constantly recorded on a recorder 29VC.

すなわち、制御系２０が正常であれば、制御中立点に対
応する回帰係数Ｂは零となり、また−制御系オリ得に対
応する回帰係数すはループゲインＫＶｃ等しくなるもの
で、この場合、フィードバック変化速度９１と制御偏差
Ｅとは、前記２式より次式のようになる。That is, if the control system 20 is normal, the regression coefficient B corresponding to the control neutral point is zero, and the regression coefficient B corresponding to the control system original gain is equal to the loop gain KVc, and in this case, the feedback change The speed 91 and the control deviation E are determined by the following equation based on the above two equations.

：Ｖｉ＝Ｋ　＠　ｅｊ　・・・・・・・・・６式また、
逆に、回帰係数ａおよびｂがそ扛ぞれａＮＯおよびｂ（
［となｎは、制御系２０には伺らかの異常が生じている
ことになる。:Vi=K @ej ・・・・・・・・・6 formula also,
Conversely, the regression coefficients a and b are aNO and b(
[This means that some kind of abnormality has occurred in the control system 20.

ここで、例えば制御系２０のサーボ弁２０Ｃが劣化して
中立点がドリフトしｆｃｎ、シリンタ。Here, for example, the servo valve 20C of the control system 20 deteriorates and the neutral point drifts, causing the cylinder fcn.

２０ｄが劣化して油漏ｎが発生する場合には、この制御
系２０は上記中立点ドリフト量や油漏れ量全補償するた
めに、余分なサーボ油流Ｉｏｋ流しておく必要がある。20d deteriorates and oil leakage n occurs, the control system 20 needs to flow an extra servo oil flow Iok in order to fully compensate for the neutral point drift amount and oil leakage amount.

この場合、偏差演算部２０ａには上記サーボ弁電流ＩＯ
に対応して、余分な制御偏差Ｅｏが常に生じるようにな
るため、この場合、前記２式は次式のようになる。In this case, the deviation calculation section 20a has the above-mentioned servo valve current IO.
Correspondingly, an extra control deviation Eo always occurs, so in this case, the above two equations become as shown below.

Ｓ　＠Ｙ（Ｓ）キに・（Ｅ（Ｓ）　−Ｅｏ　）　・・・
・・・・・・７式これによｐ１上記６式は次式のように
なる。S @Y(S) Kini・(E(S) -Eo)...
. . . Equation 7 With this, p1 Equation 6 above becomes the following equation.

マｉ＝Ｋ・（ｅｉ−ｅｏ）　・・・・・・・・・８式つ
まり、上記３式（回帰式）によりａ＝−Ｋ”　ｅｇ　＊
１）＝にとなり、正常時に比較して回帰係数Ｂ力Ｓ変化
するようになる。ｉた、例えはサーボ弁２０Ｃの劣化等
に、よって、サーボ弁２０Ｃ自体のオリ得に２が変化す
る場合には、ループゲインＫか変化して回帰係数すが変
化するようになる。My=K・(ei-eo) ・・・・・・・・・Equation 8 In other words, using the above 3 equations (regression equation), a=−K” eg *
1)=, and the regression coefficient B force S changes compared to the normal state. Additionally, if the servo valve 20C itself changes due to deterioration of the servo valve 20C, the loop gain K changes and the regression coefficient S changes.

したがって、上記のように算出される回帰係数ａおよび
ｂｋｓそｎぞれ中立点異常検出器２５および利得異常検
出器２６で監視することにより、例えばサーボ弁２０ｃ
の劣化による中立点ドリフトや利得の変化等Ｏ制御系２
０の異常？検出することができる。また、上記のような
回帰係数ａおよびｂを、そｎぞｎ第３図囚および（Ｂ）
に示すように、記録計２９によジ経過時間Ｔに伴なって
連続記録して監視することにより、制御系２０の劣化の
進み具合音知ることができると共に、寿命の予測等が可
能となる。この場合、第３図（４）においては、回帰係
数ａと、破線で示すような中立点範囲限界Ｌｌまたは−
Ｌｔ　との交点Ｘ１から中立点異常表示器２７が表示動
作？開始し、−１′た、同図（Ｂ）　においては、回帰
係数すと破線で示すよりな利得範囲限界Ｌ２またはＬ３
との交点Ｘ２から利得異常検出器２８が赤水動作？開始
するようになる。Therefore, by monitoring the regression coefficients a and bks calculated as described above with the neutral point abnormality detector 25 and the gain abnormality detector 26, for example, the servo valve 20c
O control system 2, such as neutral point drift and gain changes due to deterioration of
0 abnormality? can be detected. In addition, the regression coefficients a and b as described above are calculated as shown in Figure 3 and (B).
As shown in the figure, by continuously recording and monitoring the elapsed time T using the recorder 29, it becomes possible to know the progress of deterioration of the control system 20 and also to predict the life span. . In this case, in FIG. 3 (4), the regression coefficient a and the neutral point range limit Ll or −
Is the neutral point abnormality indicator 27 indicating operation from the intersection X1 with Lt? In the same figure (B), the regression coefficient is the gain range limit L2 or L3 shown by the dashed line.
Is the gain abnormality detector 28 operating from the intersection X2? Get started.

尚、上記実施例の鵠２図においては、監視装ｆｔｔ、２
１の一方の抽出器２２ａに対して制御偏差Ｅ全供給して
いるが、この抽出器２２ａＫは、上記制御偏差Ｅｔ−単
にＫ１倍したサーボ非電流Ｉ？供給してもよい。この場
合、前記２式は次式のようになる。In addition, in Figure 2 of the above embodiment, the monitoring devices ftt, 2
The entire control deviation E is supplied to one of the extractors 22a of 1, but this extractor 22aK is supplied with the control deviation Et minus the servo non-current I? simply multiplied by K1? May be supplied. In this case, the above two equations become as follows.

ｓ　＠　ｙ（ｓ）＝　Ｋｓ　番　Ｋｓ　・　Ｉ　（Ｓ）
　・・・・・・・・・　９　式以上のようにこの発明に
よれば、制御中立点および制御系利得にそれぞれ対応す
る回帰係数を常時算出して監視するようにしたので、例
えば補圧制御系のように、高速応答性を有する制御系を
監視する場合でも、軽故障？見逃すことなく、高感度な
故障および異常の検出が可能となる。s @ y(s) = Ks number Ks ・ I (S)
......9 As shown in the equation above, according to the present invention, the regression coefficients corresponding to the control neutral point and the control system gain are constantly calculated and monitored. Even when monitoring a control system with high-speed response, such as a control system, does it cause minor failures? Highly sensitive failures and abnormalities can be detected without being overlooked.

【図面の簡単な説明】[Brief explanation of drawings]

第１図は制御系の従来の故障検出装置全示すブロック構
成図、第２図はこの発明の一実施例に係る制御系監視装
置を示すブロック構成図、第３図囚および（Ｂ）はそｎ
ぞｎ上記監視装置の記録計に記録される回帰係数の経時
変化を示す図である。 ２０・・・制御系、２１・・・制御系監視装置、　２２
ｍ。 ２２ｂ・・・信号抽出器、２３・・・回帰係数演算器、
２４・・・微分器、２５・・・中立点異常検出器、２６
・・・利得異常検出器、２７・・・中立点異常表示器、
２８・・・利得異常表示器、２９・・・記録計。 出願人復代理人　弁理士　鈴　江　武　綾第　２図 Ｌ　、Ｊ ３Ｆ１− 第３図 ｂ（Ｂ）FIG. 1 is a block diagram showing a conventional failure detection device for a control system, FIG. 2 is a block diagram showing a control system monitoring device according to an embodiment of the present invention, and FIG. n
FIG. 3 is a diagram showing changes over time in regression coefficients recorded on the recorder of the monitoring device. 20... Control system, 21... Control system monitoring device, 22
m. 22b...signal extractor, 23...regression coefficient calculator,
24... Differentiator, 25... Neutral point abnormality detector, 26
...gain abnormality detector, 27...neutral point abnormality indicator,
28...Gain abnormality indicator, 29...Recorder. Applicant Sub-Agent Patent Attorney Takeshi Suzue Figure 2L, J 3F1- Figure 3b (B)

Claims (1)

【特許請求の範囲】[Claims] 制御系の制御偏差信号とフィードバック信号とをそれぞ
れ一足時間間隔毎に抽出する抽出器と、この抽出器によ
り抽出されるフィードバック信号からフィードバック変
化速度を算出する微分器と、この微分器により算出され
るフィードバック変化速度と上記抽出器により抽出され
る制御偏差値とにより上記制御系の制御中立点と制御系
利得とに対応する回帰係数？算出する回帰係数演算器と
、この回帰係数演算器により算出される上記制御中立点
に対応する回帰係数が予め設足される中立点範囲？逸脱
したこと全検出する中立点異常検出器と、上記０２１帰
係数演算器によｐ算出される制御系オリ得に対応す；ｂ
Ｉｇｌ帰係数が予め設足される利得範囲を逸脱したこと
を検出する利得異常検出器と全具備した仁とｔ％徴とす
る制御系監視装置。An extractor that extracts a control deviation signal and a feedback signal of a control system at each step time interval, a differentiator that calculates a feedback change rate from the feedback signal extracted by this extractor, and a differentiator that calculates a feedback change rate from the feedback signal extracted by this extractor. What is the regression coefficient corresponding to the control neutral point and control system gain of the control system based on the feedback change rate and the control deviation value extracted by the extractor? A regression coefficient calculator to be calculated and a neutral point range in which a regression coefficient corresponding to the control neutral point calculated by this regression coefficient calculator is set in advance? Corresponds to the neutral point abnormality detector that detects all deviations and the control system original gain p calculated by the 021 regression coefficient calculator; b
A control system monitoring device that is fully equipped with a gain abnormality detector that detects that the Igl return coefficient deviates from a preset gain range.
JP58126563A 1983-07-12 1983-07-12 Monitor device for control system Pending JPS6019207A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP58126563A JPS6019207A (en) 1983-07-12 1983-07-12 Monitor device for control system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP58126563A JPS6019207A (en) 1983-07-12 1983-07-12 Monitor device for control system

Publications (1)

Publication Number Publication Date
JPS6019207A true JPS6019207A (en) 1985-01-31

Family

ID=14938261

Family Applications (1)

Application Number Title Priority Date Filing Date
JP58126563A Pending JPS6019207A (en) 1983-07-12 1983-07-12 Monitor device for control system

Country Status (1)

Country Link
JP (1) JPS6019207A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS63172004A (en) * 1987-01-09 1988-07-15 Kawasaki Steel Corp Online automatic diagnosis method and device for hydraulic control equipment
JPH05106613A (en) * 1991-10-16 1993-04-27 Nippon Steel Corp Hydraulic system diagnosis method
JP2021025809A (en) * 2019-08-01 2021-02-22 カヤバ システム マシナリー株式会社 Servo valve characteristics evaluation system and servo valve characteristics evaluation method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS63172004A (en) * 1987-01-09 1988-07-15 Kawasaki Steel Corp Online automatic diagnosis method and device for hydraulic control equipment
JPH05106613A (en) * 1991-10-16 1993-04-27 Nippon Steel Corp Hydraulic system diagnosis method
JP2021025809A (en) * 2019-08-01 2021-02-22 カヤバ システム マシナリー株式会社 Servo valve characteristics evaluation system and servo valve characteristics evaluation method

Similar Documents

Publication Publication Date Title
JPS5847042B2 (en) Anomaly detection method
JP5965406B2 (en) System, method and apparatus for detecting irregular sensor signal noise
JPS6132689B2 (en)
CN111929618A (en) Transformer monitoring method and monitoring system
US20220177151A1 (en) Device for controlling an aircraft engine comprising two redundant control channels
KR101160661B1 (en) diagnostic device of a electronic equipment from radiation induced degradation using the CMOS semiconductor devices
JPH0217488A (en) Measuring apparatus of radiation
JP2018131923A (en) Hydraulic drive valve monitoring device, hydraulic drive valve monitoring method, and control system
JPS6019206A (en) Monitor device for control system
WO2020090715A1 (en) Process management device, process management method, and process management program storage medium
JPS642203B2 (en)
JPS6127110A (en) Fault diagnosing method of hydraulic rolling reduction type rolling mill
JP3438903B2 (en) Abnormality diagnosis device by vibration waveform observation
JP2777275B2 (en) Power plant abnormality detection device
RU2304060C2 (en) Device for diagnosing condition of current-conduction rail joint elements
JPS6227672A (en) Sensor self-diagnosis device
JPS59202511A (en) Fault detector of control system
JPS63111207A (en) Turbine clogging monitor
JPH04325489A (en) Abnormality analizer for process chamber
JPS5850433A (en) Vibration monitor
SU1129565A1 (en) Device for quantitative evaluation of variable resistor contacting quality
JPH0442746Y2 (en)
JPS5850435A (en) Vibration monitor
DE3146374A1 (en) Method and arrangement for detecting and indicating cooling malfunctions in reactor cores
GB1013048A (en) Improvements in or relating to monitoring systems for fault detection