DE102020133684A1 - Condition monitoring for a drive train - Google Patents

Abstract

The invention relates to a method and a system for a device for condition monitoring. With the measure of recording a measured value to be monitored and associated main influencing factors and providing for a segmentation of the working area, it is also possible to record changes in the operating status that are gradual over time and are far removed from a set limit value. Both a gradual change in the main influencing factors and a gradual change that is not due to the main influencing factors can be identified. A gradual change in the main influencing factors is recognized by a change in the chronological sequence of the segments assigned to the main influencing factors during operation. A change caused by other influencing factors is recognized by a deviation from the reference value of the respective segment.

Description

Condition monitoring, also known as condition monitoring, has found its way into many applications. The Schäffler company offers condition monitoring of bearings. The system uses vibration, noise and temperature from each bearing. Based on these signals, damage can be detected early and maintenance of the bearings can be planned at an early stage. Other companies also offer condition monitoring systems, e.g. MAN for steam turbines.

So e.g. B from the U.S. 9,091,588 B2 a system and a method for the detection of a mechanical failure based on a spectral analysis of frequencies and a comparison with theoretical frequencies is known.

Condition monitoring is usually carried out using limit values. If the first limit value is exceeded, an alarm is triggered and if another limit value is exceeded, the machine is shut down.

The behavior of the measured variables for determining whether a limit value has been exceeded changes differently depending on the respective current operating point. The behavior and the level of the measured variables often depend heavily on the operating location and the conditions in the drive train, e.g. foundations, engine, driven machine, ambient conditions and cooling, to name just a few influencing factors. The limit values are selected in such a way that the limit values cover the entire operating range. At operating points with a large distance to set limit values, the distance to the limit value or the limit values is therefore quite high. Changes are not perceived as a result.

It is the object of the invention to improve status monitoring. In particular, the task is also to be able to better recognize changes below the set limit value and to be able to provide a system for their recognition.

A measurand is measured during operation. Here, a measured variable is understood to mean a measured variable that is provided with a limit value. If the limit value is exceeded, an alarm signal is generated. A currently measured variable is referred to as an ACTUAL value.

The main influencing factors are operating parameters that have an impact on the measured variable. The measured value is recorded, with the instantaneous values of the main influencing factors also being recorded in parallel with the measured value and being assigned to the measured value or the measured value being assigned to the main influencing factors.

Map is an n-dimensional array, where n is the number of key drivers. So if two main influencing factors are used, the map is a 2-dimensional field. If, for example, torque and speed are used as the main influencing factors, the possible range of values for speed and torque is covered by the map. This map is divided into sub-areas. Segments or also referred to as grids in the 2-dimensional are spanned by the sub-areas.

Reference values of the measurand are assigned to each segment. These reference values are taught separately.

Steady state is staying in a segment for a predetermined minimum amount of time without interruption.

The object is achieved by the method with the features according to claim 1 and the system according to claim 10. Further advantageous features are specified in the dependent claims.

In the state monitoring method according to the invention, it is not provided to carry out a comparison with an alarm value. Alarm values are stored independently of an operating range and represent an absolute upper or lower limit for a stored measured value. In contrast, the method according to the invention provides for a comparison with a stored reference value to be carried out. The stored reference value is stored depending on the main influencing factors. The main influencing factors have an impact on the measured value. This makes it possible to store reference values that deviate significantly from the alarm value. By referring to the main influencing factors, a relation to the current operational situation. As a result, both minor changes and a change in the operating situation can be identified. If there is a significant change in the main influencing factors, the measured value is compared with another reference value.

It can be provided to subdivide the main influencing factors into areas and thereby carry out a segmentation. The smaller the ranges of the main influencing factors are selected, the finer the subdivision into segments and the more specific reference values can be stored. It is also possible to provide for the subdivision of the individual main influencing factors for a segmentation with different fineness over the operating range of the respective main influencing factor and also differently for the individual main influencing factors themselves.

By comparing the measured value with the reference value, a status value is generated as a function of the deviation of the measured value from the reference value. This status value also makes gradual changes clear.

In a preferred embodiment, it is provided that n main influencing factors span an n-dimensional map and this map is subdivided into n-dimensional segments, with each segment being able to be assigned exactly one reference value. By assigning reference values to the segments, it is possible to store a large number of reference values. By considering n main influencing factors, a change outside of these main influencing factors can also be recognized. The more main influencing factors are taken into account, the more specifically a reference value can be assigned. Condition monitoring becomes more accurate. Changes in the main influencing factors in operation also become comprehensible, which means that maintenance related to a main influencing factor can be better predicted and planned.

In one embodiment, it is provided that reference values are generated for the respective individual segments after commissioning, preferably within a predetermined learning time, and are stored assigned to the respective segment. This ensures that the reference values record a status that comes close to being new.

In one embodiment, it is provided that reference values are generated for segments even after the teach-in time has elapsed, with these segments not yet being assigned a reference value. However, these "late" reference values are marked separately so that it is clear that there is no comparison with a reference value representing a new system. This makes it possible for a change to be detected even for operating ranges to which no reference value was initially assigned. This means that status monitoring is also possible in the event of a change in use.

In one embodiment, it is provided that the reference values can only be generated in a steady state. In the steady state, all main influencing factors assigned to a segment must remain in the area assigned to the respective segment continuously for a predetermined period of time. The predetermined period of time is preferably at least a few minutes, particularly preferably at least 10 minutes and very particularly preferably at least 30 minutes.

In one embodiment, it is provided that a deviation from a current measured value to the reference value of the respective segment is evaluated using a rating function. Depending on the result of the rating, a display of the rating is triggered. This means that there is only an indication when the status monitoring becomes more critical. A constant signal is avoided in order to attract increased attention when the rating is displayed.

A preferred embodiment provides for the rating function to carry out a rating depending on the stored reference value and depending on the distance between the reference value and an alarm value. The alarm value is independent of the main influencing factors.

A preferred embodiment provides for the rating function to generate a rating value for the measured value, with a rating such as green, yellow or orange depending on the percentage difference between the measured value and the alarm value. The rating areas are based on a percentage difference between the reference value and the alarm value.

In one embodiment variant, it is provided that a rating dependent on the reference value is omitted when the reference value exceeds a predetermined percentage of the alarm value. A rating is preferably omitted if the reference value is at least 90% of the alarm value. This makes it clear that the alarm value is already very close in this operating range.

A preferred application of the method provides for the method to be used to detect a change in the vibration behavior of an output shaft. In particular, it has turned out to be an advantageous use for detecting a vibration of an output shaft of a drive unit with a variable transmission.

The invention relates to a system for a trend analysis of a device with a controller and a display and a sensor system of main influencing factors and a sensor system for recording a measured value. The controller within the meaning of this invention includes data processing and a data memory. Signals can be transmitted wired or wirelessly from the controller. An alarm value is stored in the controller for the measured value. This alarm value is an independent value and applies to the entire operating range. In addition, a recording of n main influencing factors is planned. An n-dimensional field is assigned to the n main influencing factors. The n-dimensional field is preferably divided into segments and a reference value can be stored for each segment. In this case, the reference value represents a reference point dependent on the main influencing factors. This reference value is intended to reflect a state of the device in the “new state”. Due to the dependence of these reference values on the main influencing factors, a change in the behavior of the device can be detected. The change then recognizable is independent of the main influencing factors. A change in the main influencing factors in ongoing operation can be recognized by a change in the time-related segment during operation.

In a preferred embodiment of the system, an alarm value is stored in the system and when the alarm value is exceeded by the measured value, a display indicates that the alarm value has been reached.

In a particularly preferred embodiment, it is provided that this proximity to the alarm value is already signaled when a measured value is in a predetermined range surrounding the alarm value. When the alarm value is exceeded, the drive power is often shut down or at least a reduction in the drive power is provided. This clear signaling can counteract this and prevent the alarm value from being reached.

In one embodiment, it is provided that the controller has a rating function. The rating function can be used to evaluate the measured value depending on the respective reference value. The rating is a classification. The result of at least one critical rating can appear on a display. As a result, necessary maintenance measures can be planned particularly early and thus efficient operation over the service life can be achieved, which has an advantageous effect on the cost-effectiveness of the device.

In an embodiment variant it is provided that the rating function uses the alarm value and the ratio of reference value and alarm value is included in the evaluation of the measured value. As a result, the operating behavior can be made particularly tailored to the current operating range. Measures can be initiated in a particularly targeted manner.

It has turned out to be advantageous that the device is a drive train and the measured value detects the vibration of a transmission shaft. As a result, a change in the vibration behavior of the output shaft can be detected early and thus increased stress on the device caused by these vibrations can also be reduced, which has an advantageous effect on the operating time and the operating behavior of the device. In addition, such a method can also be used to monitor housing vibrations and bearing temperatures and/or oil pressures. As a result, the condition of the device can be monitored in detail and necessary or recommended measures can be planned at an early stage.

In a preferred embodiment it is provided that a main influencing factor is the torque transmitted by the output shaft and a main influencing factor is the speed of the output shaft. These characteristic values can be recorded and, on the other hand, their influence on the vibration behavior of the output wave big. In a preferred embodiment, the temperature of the lubricant is also included.

In particular, it is advantageous to be able to carry out a comparison with the usual signal levels of the measured value in the different operating ranges defined by the main influencing factors and to be able to display changes accordingly at an early stage.

By using this system in a state of equilibrium, it is achieved that influences from the previous operating state no longer have any influence on the measured value.

With this system and method, gradual changes or a "running away" of the measured values can be recognized and displayed at an early stage. Maintenance work can also be carried out early and in a targeted manner. By monitoring depending on the main influencing factors, it is also possible to show which of the main influencing factors change significantly during operation. The assigned segment changes as a result of the change in at least one of the main influencing factors. By changing the assigned segment, a comparison is made with the measured value assigned to this segment.

By recognizing changes, it is possible to counteract these changes through maintenance. This means that the entire system can be protected.

• 1: Signalzuordnung
• 2: Datenkennfeld
• 3: Antriebsstrang
• 4: Rating innerhalb eines Segmentes
• 5: 2-dimensionales Kennfeld mit Ratingfunktion
• 6: Verfahren einer Trendanalyse

Further advantageous features of the invention are explained on the basis of exemplary embodiments with reference to the drawings.

• 1 : signal mapping
• 2 : data map
• 3 : powertrain
• 4 : Rating within a segment
• 5 : 2-dimensional map with rating function
• 6 : Method of trend analysis

Based on 3 the invention is explained using a specific application in a drive train 1 as a device. The drive train 1 includes a drive 3. This drive 3 is assigned a temperature sensor 11 for detecting a lubricating oil temperature. In addition, a speed sensor 13 and a torque sensor 9 are provided. However, an indirect determination of speed and torque can also be provided, which are represented only symbolically by these sensors 9, 13. A vibration sensor 15 is provided for determining a vibration of an output shaft 5 . This vibration can also be detected indirectly.

In a drive train 1, a maximum permissible vibration, referred to as an alarm value 75, is predetermined for the output shaft 5. Factors such as output speed, output torque as the main influencing factors 45 have a great influence on the vibration behavior of the output shaft 5 . The torque sensor 9 and the speed sensor 13 and the temperature sensor 11 act as sensors for the main influencing factors, as shown in FIG 1 shown. This drive train is provided with a controller 17 to which the signals from the sensors are supplied. The controller 17 is connected to a display 19 . A critical state or the current state can be displayed by the display 19 . It can also be provided that the current operating point is displayed using the display 19 .

A permissible range of operation is limited and predetermined by the set alarm value 75. The allowable operating range is an n-dimensional array when n main influencing factors are recorded. In 2 a three-dimensional field is shown with the main influencing factors 45 of output torque, output speed and lubricating oil temperature. If two main influencing factors 45 are to be considered, then a 2-dimensional characteristic map 30, see FIG 5 , based on. The characteristic map 30 is subdivided into segments 31, wherein the segments 31 in the 2-dimensional characteristic map 30 can also be referred to as grid fields. In this specific case, the torque transmitted by the output shaft 5 and the speed of the output shaft 5 are recorded as the main influencing factors 45, 112. After commissioning, the vibration of the output shaft 5 is recorded as a measured value 55 in the drive arrangement in stationary operation, referred to as the steady state, 130. The conditions for stationary operation are subject to change it's correct. It can be sufficient if the rotational speed and the torque remain in a segment 31 for at least 10 minutes, better 30 minutes. Then the vibration of the output shaft 5 is detected and assigned 140 to this segment 31 as a reference value. In the following, persistence or a state of inertia is also spoken of when the main influencing factors 45 in the area that is assigned to a segment 31 remain without interruption over the predetermined minimum period of time. This ensures that the vibration of the output shaft 5 is not caused or influenced by a previous operating situation.

It can be provided that this storage of reference values 65 is only permitted within a predetermined period of time from the first start-up. This period of time is referred to as "learning time active", reference number 118 . Designated after this period of time has elapsed, segments 31 that have not yet been provided with a reference value can also be provided with a reference value 65, but this can be marked separately. This makes it possible to recognize whether the operating range has changed during operation. This change can also be related to gradual wear or indicate a change in operation or use.

• Wird im weiteren Betrieb eines der eingelernten Felder angefahren und befinden sich die Haupteinflussfaktoren 45 in der definierten Beharrung, so findet eine Bewertung, auch als Rating bezeichnet, des aktuell gemessenen Messwertes 55 unter Berücksichtigung des zuvor eingelernten Referenzwertes 65 statt. Diese Verfahrensschritte sind in 6 in der Abfolge Referenzwert ermitteln 150, Rating durchführen 152 und ggf. Anzeige 154 des Ratings vorgesehen.

Rating function 33:

• If one of the taught-in fields is approached during further operation and the main influencing factors 45 are in the defined persistence, then an evaluation, also known as a rating, of the currently measured measured value 55 takes place, taking into account the previously taught-in reference value 65 . These procedural steps are in 6 provided in the sequence determine reference value 150, perform rating 152 and, if necessary, display 154 the rating.

In addition, the predetermined alarm value 75 can also be included. It can be taken into account how far away the segment 31 approached is from the predetermined alarm value 75, 5 . If the segment 31 approached is very close to the limit value, it can be provided that a determination of a deviation from the reference value 65 is no longer carried out. There is no separate rating. It is then intended to directly signal an indication of a critical situation. In the case of a color rating, a preliminary stage to red, namely an orange rating, can be provided. An evaluation of the distance between the measured value and the alarm value 75 then has priority in status monitoring. The rating result is categorized according to its deviation, preferably divided into three categories, rating green / rating yellow / rating orange. A color coding of 35 is provided here as a rating. Recommendations for action, eg additional inspections, etc., can be stored or defined for the respective categories.

An example of such a stored rating, which also includes the fixed, predetermined alarm value 75, is in 5 shown.

If, based on the specific exemplary embodiment, in the following operation the drive train 1 is again operated with a speed and a torque that lead to a segment 31 in which a reference value 65 is stored, then the instantaneous vibration value is used as a measured value 55 of the output shaft 5 with the stored Reference value 65 compared. If this deviation is large, then this is an indication of an error or defect in the drive train 1. Since this deviation is not influenced by the main influencing factors 45, the change must have other causes. In this way, gradual changes in particular can be detected.

A classification, also referred to as a rating, of the deviation of the measured value 55 from the reference value 65 can be provided. Provision can also be made for a rating 60 to depend on the ratio of reference value 65 to alarm value 75 . For example, dependencies as listed in the following tables are selected and stored in the controller. Table 1: Evaluation of the difference between the measured value and the reference value green Reference value in % of the alarm value yellow Reference value in % of the alarm value orange Reference value in % of the alarm value ACTUAL value > reference value from until 0 <0.1 No calculation! 0.1 <0.55 <12.5% (linear) of alarm value 12.5-25% > 25% (linear) of alarm value 0.55 <0.8 <12.5%-0% (linear) < 20% >20% (linear) 0.8 <0.9 <0% 0-10% >around 10% (linear) 0.9 1 No calculation! ACTUAL value < reference value from until 0 <0.3 No calculation! 0.3 <0.9 >-30% of alarm <-30% omitted 0.9 1 No calculation!

For example, if the deviation is the ratio of the measured value 55 to the reference value between 0.1 and 0.55 and if the reference value is less than 12.5% of the alarm value, the rating is "green".

A corresponding characteristics map 30 is defined on the basis of the main influencing factors 45 and subdivided into segments 31 of reasonable size. The size depends on the application. For this purpose, the respective output variable, measured value, reference value is modeled as a scalar field of the input variables (influencing factors), ie as y ≙ f(x ₁ ,x ₂ , ...).

The learning function of the reference value is in 6 shown in simplified form.

• Um eine Trendanalyse durchführen zu können, sollte zuerst ein Referenzwert 65 ermittelt werden. Dazu wird ein Filter implementiert, der eine stationäre Betriebssituation „Beharrungszustand“ erkennt. Der Filter mit Funktionsfeld 120 soll für die zu analysierende Haupteinflussfaktoren 45 einen eindeutig definierten Beharrungszustand erkennen, 120. Der Beharrungszustand tritt dann ein, wenn alle festgelegten Haupteinflussfaktoren 45 über einen vorbestimmten Zeitraum t_prüf innerhalb eines gewählten Toleranzbereichs bleiben. Für diesen Fall werden die Werte der Haupteinflussfakten fortlaufend in einem Ringspeicher zur weiteren Verarbeitung abgespeichert.

Test steady state 120:

• In order to be able to carry out a trend analysis, a reference value 65 should first be determined. For this purpose, a filter is implemented that recognizes a stationary operating situation "steady state". The filter with function field 120 should recognize a clearly defined steady state for the main influencing factors 45 to be analyzed, 120. The steady state occurs when all defined main influencing factors 45 remain within a selected tolerance range over a predetermined period of time t_test. In this case, the values of the main influencing factors are continuously stored in a ring memory for further processing.

• Werden die Bedingungen für den Beharrungszustand von allen Haupteinflussfaktoren 45 gleichzeitig über den gesamten Beobachtungszeitraum eingehalten, werden Daten des Messwertes 55 abgespeichert. Die Daten des Messwertes werden gespeichert und ggf. über einen vorbestimmten Zeitraum gemittelt. Der Messwert 55 wird als Referenzwert 65 dieser Paarung der Haupteinflussfaktoren 45 zugeordnet. Durch die Paarung an Haupteinflussfaktoren wird das zugeordnete Segment bestimmt und der bestimmte Referenzwert wird diesem Segment zugeordnet, 140.

Determination of the current segment 114:

• If the conditions for the steady state are met by all main influencing factors 45 simultaneously over the entire observation period, data of the measured value 55 are stored. The measured value data is stored and, if necessary, averaged over a predetermined period of time. The measured value 55 is assigned as a reference value 65 to this pairing of the main influencing factors 45 . The assigned segment is determined by the pairing of main influencing factors and the determined reference value is assigned to this segment, 140.

learning process

• - Anzahl der Messpunkte
• - Min Signal Wert im Segment
• - Max Signal Wert im Segment
• - Signal Mittelwert → Referenzwert
• - Standard Abweichung
• - optional Variationskoeffizient

During the learning process, the following values are stored based on the measured value 55 in the active segment 31:

• - Number of measuring points
• - Min signal value in the segment
• - Max signal value in the segment
• - Mean signal → reference value
• - Standard Deviation
• - optional coefficient of variation

The segments 31 can be taught with interruptions, e.g.

The learning process is defined for the individual segments 31 by the following criteria and must finally meet the following conditions: - Map learning duration t_Lern_Kf_ges is the period within which the machine/the drive train is considered "new", 118. A reference value 65 may be taught .

After this time period 118 has expired, a segment 31 can no longer be taught in as a “new reference value”. If, after this period of time, segments 31 are approached for which no reference value has yet been stored, reference values continue to be determined 154 and 158 . These “late reference values” are stored with a separate identifier 156 . This applies to the entire characteristics map 30. This also makes it possible to separately consider changes in which segments 31 are approached for the first time.

Learning duration limit, 118, for the individual map grid element t_Lern_EI: T_Lern_EL is a predetermined time period and restriction that defines when there is sufficient data for an individual segment 31 in the map 30 so that a reference value 65 can be determined. The reference value 65 is generally formed by averaging the available measurement data for the respective segment 31 . If a segment 31 is assigned a reference value, then the learning process for this segment 31 is complete 140. The learning time can limit the learning of reference values 118 to a predetermined period of operating hours after the initial start-up. It can thereby be ensured that a reference value 65 is determined when a segment 31 is approached for the first time, extensive wear and tear of the device 1 not already being included in the reference value 65 determined.

The following states can exist for each segment 31 in the map: 1. Not trained → "no reference" 2. Learning → "in progress" 3. Done learning → "completed / yyyy.mm.dd" The date is also used to distinguish whether the reference value 65 was learned in the “new state” 140 after commissioning, or was learned at a later point in time 156 after t_Lern_Kf_ges has expired.

• Vor dem Abspeichern ist vorgesehen den Referenzwert 65 hinsichtlich seiner Plausibilität zu prüfen. Hierzu wird der Alarmwert 75 des Signals herangezogen. Für die Fälle des sehr kleinen oder sehr großen Referenzwerts 65 bezogen auf den Alarmwert 75 ist vorgesehen schon bei der Ablage des Referenzwerts eine Warnung auszugeben, dass der Referenzwert 65 bedenklich ist und eine nähere Untersuchung der Ursache erforderlich ist, siehe Tabelle.

Plausibility assessment 145 of the reference value 65:

• Before storing, the reference value 65 is checked with regard to its plausibility. The alarm value 75 of the signal is used for this. If the reference value 65 is very small or very large in relation to the alarm value 75, a warning is provided when the reference value is stored that the reference value 65 is questionable and that the cause needs to be investigated more closely, see table.

• Durch die Ratingfunktion findet ein Vergleich des aktuellen Messwerts 55 mit dem entsprechenden Referenzwert 65 des aktiven Segments 31 statt. Die Differenz wird unter Berücksichtigung der festen Signal Alarmwertes 75 einer Bewertung unterzogen.

Rating function:

• The rating function compares the current measured value 55 with the corresponding reference value 65 of the active segment 31 . The difference is subjected to an evaluation taking into account the fixed signal alarm value 75 .

Only measured values below the fixed alarm value of 75 are evaluated. A precise statement can be made about the change in a measurement signal by determining the permissible percentage deviation between the current measured value and the reference value. The percentage ratio of the reference value in relation to the fixed alarm value is also included. The evaluation key thus refers to the ratio of reference value 65 and alarm value 75. The The range between the alarm value and the reference value is divided into different zones. (e.g. green, yellow, orange).

If the reference value 65 is very small or even 0, no rating is carried out, as this would not provide a meaningful result. If the reference value is very large, for example >80% of the alarm value, the measured values must be evaluated much more strictly. Such a high reference value 65 should be identified and reported as quickly as possible in order to be able to search for the cause at an early stage.

If the reference value is already very high, greater than 90% of the alarm value, no calculation is carried out because the added value of such a calculation is very small.

In cases where the reference value 65 is very small or very large in relation to the alarm value 75, a warning must be issued when the reference value 65 is stored that the reference value 65 is questionable and that the cause needs to be investigated more closely.

If the current measured value is significantly below the reference value, a message is also generated. In this case, either the reference value 65 or the measured value 55 must be critically examined.

In order to avoid a possible constant jumping back and forth of the evaluation / rating in the border area between the different categories of the rating, green and yellow or yellow and orange, an additional condition can be stored. A rating must be present without interruption for at least a predetermined period of time, for example in excess of 10 minutes, before a message indicating that rating is generated and issued.

• Als zusätzliches Bewertungskriterium kann vorgesehen sein, dass die Streuung des Messwertes in einem bestimmten Betriebsbereich durch die Bestimmung der Standardabweichung δ ermittelt wird. Dabei kann vorgesehen sein, dass aus Erfahrungswerten gewonnen Streuwerte bereits in der Steuerung hinterlegt sind und bei einem Rating berücksichtigt werden.

Evaluation of the scatter of the measured values:

• As an additional evaluation criterion, it can be provided that the scatter of the measured value in a specific operating range is determined by determining the standard deviation δ. It can be provided that scattered values obtained from empirical values are already stored in the controller and are taken into account in a rating.

The standard deviation always refers to an associated mean value µ. A direct comparison is not given. The coefficient of variation V is the quotient of the running standard deviation and the running mean in %. It is therefore a clear measure of the scatter of the measured values. The smaller the coefficient of variation, the smaller the spread of the values. $$V=\frac{\delta }{\mu }\times 100\%$$

The coefficient of variation is formed both for the reference value (V ₆₅ ) and for the current measured value (V ₅₅ ). If you put the two coefficients of variation in relation, you get a measure of the deviation of the scatter of both data series and can classify them again.

Evaluation limits spread V65/V55 <1.333 green 1.333-2.0 yellow >2.0 red

Reference List

1

powertrain

3

drive

5

output shaft

7

Sensor main influencing factor

9

torque sensor

11

temperature sensor

13

speed sensor

15

vibration sensor

17

steering

19

advertisement

21

Limit reading

23

reference value

25

operating range

30

map

31

segment

33

rating function

35

color coding

45

Main Influencing Factors

55

reading

60

Rating/Condition Value

65

reference value

75

alarm value

112

Determine the main influencing factors

114

identify associated segment

118

Learning time active

120

Determination of residence time/steady state

130

Determine measured value

140

reference value formation

145

plausibility check

150

Query reference value for segment

152

carry out a rating

154

View or save rating

156

Reference value creation after learning time

158

Allocation of "Reference value after teach-in time" to the segment

160

warning

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• US9091588B2 [0002]

Claims (15)

Method of status monitoring in which a measured value (55) is recorded during operation and a status value is generated as a function of the recorded measured value (55) by comparison with a stored value, characterized in that the measured value (55) is recorded and that main influencing factors ( 45) are recorded, which have an influence on the measured value (55), depending on the main influencing factors (45), a reference value (65) as a stored value being assigned to the recorded main influencing factors (45) and the measured value (55) to the reference value (65) is related and depending on a deviation of the measured value (55) from the reference value, a status value (60) is generated. procedure after claim 1 , characterized in that n main influencing factors (45) span an n-dimensional characteristic map and this characteristic map (30) is divided into n-dimensional segments (31), each segment (31) being able to be assigned exactly one reference value (65). Method according to claim claim 1 or 2 , characterized in that reference values (65) are generated after commissioning for each of the individual segments, preferably within a predetermined learning time, and assigned to the respective segment (31) are stored. procedure after claim 3 , characterized in that after the training time has elapsed, reference values (65) generated for the first time for segments (31) are assigned to the respective segment (31) in a marked manner. Method according to one of the preceding claims, characterized in that the reference values (65) are only generated in a steady state, in which state all the main influencing factors (45) assigned to a segment are consistently over a predetermined period of time in the value range assigned to the respective segment (31). remain continuous, preferably the predetermined period of time is at least a few minutes, preferably 10 minutes and particularly preferably at least 30 minutes. Method according to one of the preceding claims, characterized in that a deviation from a current measured value (55) to the reference value (65) of the respective segment (31) is evaluated by means of a rating function (33), depending on the result of the rating ( 60) a display of the rating (60) is triggered. procedure after claim 6 , characterized in that the rating function (33) generates a rating value (60) for the measured value ( 55) is generated, the rating (60) preferably depending on the percentage difference between the reference value (65) and the alarm value (75). procedure after claim 6 or 7 , characterized in that a rating (60) dependent on the reference value (65) is omitted if the reference value (65) exceeds a predetermined percentage of the alarm value (75), preferably at least 90% of the alarm value (75). Method according to one of the preceding claims, characterized in that the method is provided for detecting a change in the vibration behavior of a transmission shaft (5), housing vibrations, bearing temperatures and/or oil pressures. System for a trend analysis of a device (1), preferably a drive arrangement, for a method according to one of Claims 1 until 9 with a controller (17) and a display (19) and sensors (9, 11, 13) of main influencing factors and sensors (15) of a measured value, with the measured value (55) being assigned an alarm value (75) in a controller (17) is assigned, characterized in that an n-dimensional field (30) is stored in the controller (17) for a number of n main influencing factors (45) and the n-dimensional field (30) is dependent on the main influencing factors (45). Segments (31) is divided, the segments (31) are each defined by value ranges of the main influencing factors (45) and assigned to each segment (31) a reference value can be stored. system after claim 10 , characterized in that the alarm value (75) independent of the main influencing factors is stored in the system, when the measured value exceeds the alarm value, the reaching of the alarm value (75) is indicated on a display. system after claim 10 or 11 , characterized in that the controller (17) has a rating function (33), with the rating function (33) evaluating (60) the measured value (55) as a function of the respective reference value (65) and showing it on a display (19 ) the result is at least one critical rating (yellow, orange). system after claim 12 , characterized in that the rating function (33) accesses the alarm value (75) and the ratio of reference value (65) and alarm value (75) is included in the evaluation of the measured value (55). system according to one of the Claims 10 until 13 , characterized in that the device is a drive train (1) and the measured value (55) detects the vibration of a transmission shaft (5), housing vibrations, bearing temperatures and/or oil pressures. system according to one of the Claims 10 until 14 , characterized in that a main influencing factor (45) is the torque transmitted by the output shaft (5) and a main influencing factor is the speed of the output shaft and the lubricating oil temperature of the bearing of the output shaft.

Images