DE102021120491A1 - Device for determining the actual state and/or the remaining service life of structural components of a working machine - Google Patents

Abstract

The present invention relates to a device for determining the actual state and/or the remaining service life of structural components, for example large diameter bearings, of a work machine, in particular construction, material handling and/or conveyor machines, with a sensor system for detecting status information relating to the structural component and an evaluation device for evaluating the recorded status information and determining the actual status and/or the remaining service life on the basis of a comparison with predetermined damage characteristics, with an active database device being provided for making the damage characteristics available, to which a determination device for determining the damage characteristics from design data of the structural component and an adjustment device for Adjusting the predetermined damage characteristics based on the status and / or remaining service life information determined by the evaluation device are connected.

Description

The present invention relates to a device for determining the actual state and/or the remaining service life of structural components, for example large diameter bearings, of a work machine, in particular construction, material handling and/or conveyor machines, with a sensor system for detecting status information relating to the structural component and an evaluation device for evaluating the recorded status information and determining the actual status and/or the remaining service life based on a comparison of the recorded status information with predetermined damage characteristics.

With construction machinery such as excavators, cranes, dump trucks, bulldozers, bulldozers or dragline cranes, or material handling machines or conveyor systems such as forklifts and loaders or other larger work machines such as surface milling machines or ship cranes, it is equally important and difficult to determine the remaining service life or the time remaining before a replacement is necessary to predict structural parts. If a construction machine during use on a construction site breaks down, for example because a bearing is seizing or overheating, it is often not possible to procure a suitable replacement machine and deliver it to the construction site, so that delays occur during the repair time required for the repair on the construction site, whereby often Not only are the tasks of the failed construction machine itself left undone, but other processes are also delayed due to the interlocking of the various construction machines. A similar problem also arises with the aforementioned material handling machines and conveyor systems.

It is often only individual components or structural parts that are easy to replace per se and would not cause any major subsequent delays if the replacement can be planned in good time, but on the other hand can also cause greater repair costs if other machine components break when the small structural part breaks be affected.

It is normal in itself for structural components to show a certain degree of wear over the course of the service life, without this immediately requiring the component to be replaced. As long as the function of the component is guaranteed and there is no threat of failure, such signs of wear that are not dangerous for operation are accepted in order to maximize the service life. Such smaller, tolerable signs of wear can be, for example, small hairline cracks that are not critical for the strength or smaller surface breakouts or pittings of less functionally relevant surfaces. However, such signs of wear can also be other component changes such as a reduction in elasticity or increasing bearing play.

Depending on how the structural components of a working machine are monitored, it can be very difficult to distinguish critical damage from non-critical damage, since the signal response of the component-monitoring sensors can show more or less large changes even with minor component changes.

In addition, many different types of damage and associated damage characteristics can occur on the structural components, which can be reflected in correspondingly different abnormalities in the sensor data. For example, vibrations that occur as a result of normal, so to speak, uniform wear on the gear wheels of a gear stage can turn out differently than vibrations that are caused, for example, by a single, more severely damaged tooth of a gear pair. A different vibration picture can result from a lack of lubricant and the associated sluggishness due to overheating or in the case of uneven running due to contamination or excessive bearing clearance.

Sensory monitoring systems for construction machines have therefore already been proposed, which are intended to objectify the determination of the actual states of construction machines using measured sensor data. It is known, for example, to monitor specific operating parameters of the construction machine and to output an error code in the event of irregularities or unusual values in the measured operating parameters, cf JP-OS-8-144 312 . However, such error codes per se are not very meaningful or reliable, since, for example, briefly exceeding a permissible speed, as can occur for example when driving downhill in a construction site access, does not yet allow a reliable statement to be made about engine damage caused as a result.

The font DE 101 45 571 A1 the applicant Komatsu also proposes a monitoring system for construction machines, which intends to predict the degree of damage or abnormality in a more differentiated manner. For this purpose, on the one hand the exhaust pressure and the exhaust gas temperature of the construction machine diesel are monitored by sensors and on the other hand the lubricating oil is analyzed for certain components such as iron particles using a special analysis device. In addition to these sensory monitoring variables, however, the cited document also considers it necessary to include the result of a visual inspection, which is carried out by an experienced maintenance worker, in the automated assessment of the actual condition of the construction machine. On the one hand, this previously known monitoring system for construction machines suffers from a limited reliability of the status statement. The monitored exhaust gas parameters, exhaust gas temperature and exhaust gas pressure, can primarily only be used to determine problems with the diesel. On the other hand, the monitoring system is still relatively complex, since visual inspections have to be carried out by maintenance workers.

The present invention is therefore based on the object of creating an improved device for determining the actual state and/or the remaining service life of a construction machine, which avoids the disadvantages of the prior art and develops the latter in an advantageous manner. In particular, a reliable determination of the actual state and/or remaining service life that can be easily implemented on mobile construction machines is to be achieved, which enables maintenance or repair measures to be initiated and planned in good time even by untrained maintenance employees with sufficient lead time.

According to the invention, the above object is achieved by a device as defined in claim 1. Preferred developments of the invention are the subject matter of the dependent claims.

In order to do justice to the variety of possible damage characteristics, it is therefore proposed to artificially generate synthetic damage characteristics in the form of samples or reference examples based on the construction or the geometry and material data and to continuously adapt these synthetically generated damage characteristics by feeding back real ones from the sensor data determined status information. According to the invention, an active database device is provided for keeping the damage characteristics available, to which a determination device for determining the damage characteristics from design data of the structural component and an adaptation device for adapting the predetermined damage characteristics based on the actual status and/or remaining service life information determined by the evaluation device are connected. Through the synthetic generation of damage characteristics from the design data of the structural component on the one hand and the subsequent adjustment of the synthetically generated damage characteristics through feedback of the evaluated condition and/or remaining service life information, a complex damage characteristics model can be used to determine the current condition and/or the Remaining service life can be created that does justice to the variety of possible damage and provides a high degree of accuracy when determining the current state or the remaining service life.

When determining the synthetic damage characteristics, various design data can be used to calculate the parameters or the data set that makes up the respective damage characteristic sample. For example, relevant damage indicators such as the rollover frequency of the bearing outer ring can be derived for a rolling bearing from design data such as the number of rolling elements, number of rows, specified speed and/or geometric variables such as diameter, raceway width or pitch circle diameter and/or material data of the structural component such as rolling element and raceway hardness or rolling element and raceway material Temperature profile over the operating time, or an acoustic emission pattern can be calculated or the frequency spectrum of the envelope signal can be determined for a specific speed. In particular, the determination device for determining the damage characteristics from design data can have a module for determining and/or calculating kinematic frequencies from geometric data and/or operating data such as rotational speed and/or movement speed and, if necessary, taking into account material data such as weight or hardness, with said module containing the kinematic Frequencies preferably determined independently of acting external forces or energies.

Advantageously, the named determination device can also have an adaptation module in order to adapt and adapt vibration or frequency patterns or spectra that correspond to different damage types or damage patterns to the respective system using the calculated or determined kinematics, in particular using the aforementioned kinematic frequency transform, so that adapted frequency patterns or spectra are generated that do justice to the different damage types or patterns of the specific system or the specific, interested component. The output frequency patterns or spectra that have not yet been adapted can be determined beforehand on other, real components by measurement or are also known in catalog form from damage type libraries. For example, a damage pattern memory can be connected to the determination device from which the determination device can take or read out the frequency pattern that has not yet been adapted and then adapt it using the previously determined kinematic frequency or the kinematics of the specific component. The adapted frequency patterns or spectra can then be stored again in a memory.

If, for example, localized damage occurs in a bearing, e.g. B. rolling element indentations, standstill corrosion or breakouts, these can be determined, for example, by vibration measurements. When rolling over local depressions, shock waves arise which can be calculated or estimated during the synthetic determination on the one hand and can be recorded by distance, speed or acceleration sensors or possibly other sensors on the other hand. For example, by comparing the detected rollover frequency with previously synthetically determined rollover frequency damage patterns, the actual state or damage to the bearing can be determined.

In a similar way, for example, a structure-borne noise pattern of the structural component can also be calculated, estimated or determined in another way from the design data mentioned, and compared with a real structure-borne noise pattern, which can be detected by means of one or more structure-borne noise sensors on the structural component, in order to determine the actual to determine the condition of the structural component. When determining the structure-borne noise pattern, it is also possible to work with a structure-borne noise pattern library from which structure-borne noise patterns characteristic of certain component types and/or system types are stored, which are then adapted and adapted using the previously calculated or determined kinematics of the specific component or the specific system can be transformed into an adapted structure-borne noise pattern.

Similarly, other synthetic damage characteristics of other structural components, such as gear mesh frequency of a driveshaft, can also be synthesized from design data.

Alternatively or additionally, however, synthetic damage characteristics can also be generated on the basis of other system data, for example on the basis of a relationship between speed and temperature using performance data. In particular, it can be assumed that a specific temperature can be expected for a specific transmitted power at a specific speed, and in this respect a speed-temperature-power matrix can be generated. If unusual temperatures occur in certain performance and/or speed ranges, a certain type of damage can be inferred.

In order to do better justice to the complexity of possible damage patterns, a combination module can advantageously be connected to the active database device, which combines the damage characteristics synthetically generated from the design data and thereby creates combined damage characteristics that can be kept ready or stored in the database device. Such combined damage characteristics can correspond to more complex damage patterns, in which not only a special type of damage occurs on the corresponding structural component, such as e.g. spalling in the rolling bearing ring raceway, but various types of damage occur at the same time, for example, in addition to the aforesaid raceway ruptures, there is also bearing contamination or excessive bearing clearance, which can lead to unsteady running.

The combination module mentioned can not only combine different damage patterns of a structural component with one another, but alternatively or additionally also combine the damage characteristics of different structural components with one another in order to be able to take mutual influences of damage characteristics of different structural components into account.

For example, a damaged bearing and the associated rough running of the bearing can also influence vibrations in a gear stage supported by it and/or have an impact on the damage pattern of the gear stage that is rotationally supported by the bearing. The combination module can therefore advantageously be configured to combine different, synthetically generated damage characteristics of a structural component with one another and/or to combine synthetically generated damage characteristics of different structural components with one another. The damage characteristics generated in this way in a combinatorial manner can also be stored in the database device or made available by it.

In order to achieve a more reliable determination of the actual state or the remaining service life, in an advantageous development of the invention, the individual and/or combined damage can be weighted characteristics be provided. In particular, each sample or each synthetically generated damage characteristic can be weighted according to the probability of occurrence, ie the frequency of the cause of failure of the respective component. Damage characteristics with a higher probability of occurrence can be weighted more heavily than damage characteristics that reflect damage that occurs relatively rarely.

In this case, the weighting module can be configured to weight the synthetically generated damage characteristics individually in each case. Alternatively or additionally, the weighting module can also weight the damage characteristics generated in a combinatorial manner, in particular based on the probability with which a specific damage combination occurs.

During the evaluation of the status information determined by sensors and the comparison with the damage characteristics provided, the parameters that make up the status information can fundamentally be evaluated in different ways or compared with the damage characteristics in different ways. For example, different parameters can also be weighted differently here and/or taken into account in different ways.

In this case, a change in a respective parameter can be taken into account in absolute terms, for example to the extent that the predetermined damage characteristics are adapted if a predetermined amount of change is exceeded. As an alternative or in addition, however, summary consideration can also take place, for example to the effect that when a predetermined change is exceeded with summary consideration of a number of parameters, the predetermined damage characteristics are adapted by the adaptation device.

The adaptation device can adapt the damage characteristics in different ways to the evaluated, real status information or using this status information.

In a further development of the invention, the status signal evaluation and/or the damage characteristics provided by the database are adapted to changing status and/or operating conditions of the structural component or aging influences, not based on rigid criteria, but with the help of a self-evolving, variable set of rules . In particular, in a further development of the invention, the named evaluation device and the adjustment device are designed as a self-learning system or the named components form part of a self-learning system that assesses the influence of the recorded status and/or operating information or the real information pattern derived therefrom on the damage characteristics or .On which a damage pattern of the representative set of parameters can be estimated.

In particular, the evaluation device and the said adjustment device can be designed with artificial intelligence or implemented in an AI system, which can have a regression analysis module, for example, to establish a relationship between the detected status and/or operating information and the synthetically generated status or damage - Can estimate reference samples of the work machine or the structural component. For example, the named regression analysis module can adapt or develop a functional relationship between the named parameters or a curve that characterizes the dependency of the actual condition or the remaining service life on the named real condition parameters, preferably using the continuously recorded changes in condition and the actual condition that occurs - or remaining service life prognosis, in particular with the further aid of a training set of the parameters mentioned. Said training set of parameters can originally be specified, for example obtained using one or more test runs, and/or continuously updated or expanded, in particular using data obtained during machine operation.

In particular, the device can use artificial intelligence to compare the constantly measured state response of the structural component or the entire work machine or a subassembly thereof with the state or operating behavior and/or damage pattern characteristic of this in order to identify damage to one or more structural components, in particular crack formation .

By means of the AI system, the reference examples of the damage characteristics of the working machine or of the structural component(s) can be generated, in particular as a function of the recorded, real parameter sets of the working machine and/or of the structural component, which reflect the real machine condition. and their changes are adjusted. As a result, the forecast can always be kept sharp and the error rate minimized.

In order to be able to better monitor the work machine overall for damage, sensors for detecting relevant status parameters such as vibrations or structure-borne noise can be assigned to a number of structural components, for example roller bearings, of the system.

The evaluation device can compare the status information of the various structural components, e.g. different rolling bearings, with one another in order to compare changes in the status or the set of parameters representing this on a structural component with associated changes in the status information on one or more other structural components and in this respect to compare abnormal changes in the status more precisely to be able to capture. Such a comparison can be carried out in addition to the mentioned adaptation of the evaluation criteria and/or the signal reference pattern.

In a further development of the invention, the structural component can be, for example, a roller bearing or large roller bearing, for example a centerless large roller bearing with a diameter of more than 0.5 m or more than 1.0 m and with regard to the rollover frequency pattern and/or the structure-borne noise emissions and/or the temperature curve over the duty cycle and/or a noise emission pattern.

• 1: eine Darstellung einer Vorrichtung zum Bestimmen des Ist-Zustands und/oder der Restlebensdauer von Strukturbauteilen einer Arbeitsmaschine nach einer vorteilhaften Ausführung der Erfindung,
• 2: eine Darstellung der Vorrichtung aus 1 mit ergänzenden Details in den einzelnen Komponenten der Vorrichtung,
• 3: eine Darstellung des Frequenzbilds entsprechend einem Schadmuster und dessen Transformation zu einer spezifisch auf das interessierende System adaptierten Frequenzschadssignatur, und
• 4: eine schematische Darstellung des synthetischen Generierens einer Schadenscharakteristik ausgehend von einem durch Strukturanalyse generierten Grundschwingungsbild und dessen Transformation zu einem spezifischen Schadenscharakteristikbild.

The invention is explained in more detail below using a preferred exemplary embodiment and associated drawings. In the drawings show:

• 1 : an illustration of a device for determining the actual state and/or the remaining service life of structural components of a work machine according to an advantageous embodiment of the invention,
• 2 : a representation of the device 1 with additional details in the individual components of the device,
• 3 : a representation of the frequency image corresponding to a damage pattern and its transformation to a frequency damage signature specifically adapted to the system of interest, and
• 4 : a schematic representation of the synthetic generation of a damage characteristic based on a fundamental vibration image generated by structural analysis and its transformation to a specific damage characteristic image.

As the figures show, the condition monitoring system 1 includes an active, self-implementing database 2 that provides a large number of synthetically generated damage characteristics as reference examples 3 in the form of data sets that specify different damage patterns of a structural component 4 and contain parameters on which the respective Damage pattern fixes or that are characteristic of the respective damage pattern.

As 2 shows, a reference example 3 referred to as "Sample I" can contain the rollover frequency of a bearing ring, for example a bearing outer ring of a roller bearing, with the said rollover frequency containing the mentioned frequency spectrum, for example for different degrees of damage and possibly also for the undamaged state for one or more speeds .

Another reference example 3, referred to as "Sample II", can contain the damage characteristics of a drive module, for example in the form of the meshing frequency of a drive shaft.

Various other reference examples can contain, for example, a temperature profile of a bearing over the running time and/or over the time after it has been switched off, or an acoustic emission spectrum of a roller bearing, or a vibration pattern of a component or other characteristic damage patterns.

As the figures show, a determination device 5 can be connected to the database 2, which automatically generates the synthetic damage characteristics mentioned from the design data provided for the respective structural component 4, in particular from geometry and/or drawing and/or material and/or material data calculated. Such a determination device can also estimate the damage characteristics for a respective structural component, with the help of stored data sets for similar structural components, and/or estimate the synthetic damage characteristics for the structural component of interest from known, typical damage characteristics for a specific structural component, for example by inter- and/or or extrapolation based on the geometry data.

For the synthetic generation of damage characteristics, the kinematics of the specific system of interest can first be calculated. In particular, kinematic frequencies can be calculated for any machine components such as roller bearings, gearing or shafts. Kinematics means a description of the system through its geometry and parameters that change over time, independently of forces and energies.

$$\text{Berechnung der }\ddot{\text{U}}\text{berollfrequenz am Innenring }\to u_j=\frac{z}{2}\cdot \left(1+\frac{D_W\cdot co{s}^2\alpha }{D_{pw}}\right)$$

$$\text{Berechnung der }\ddot{\text{U}}\text{berollfrequenz am Innenring }\to u_A=\frac{z}{2}\cdot \left(1-\frac{D_W\cdot co{s}^2\alpha }{D_{pw}}\right)$$

If, for example, a roller bearing is considered, the kinematics can be determined as follows, with geometric data such as the number of rolling elements z, the rolling element diameter Dw, the pitch circle diameter D _PW under the contact angle α being known or possibly being obtained from a design database. The following steps can be carried out to determine the kinematics of the rolling bearing: $$\text{Calculation of the rotation frequency of the W}\ddot{\text{a}}\text{lzk}\ddot{\text{O}}\text{rpers}\to n_W=\pm \frac{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DE102021120491A1\_0004.png,DE102021120491A1\_0005.png"\; state="\{\{state\}\}">n</figure-callout>}}{2}\cdot \left(\frac{D_{pw}}{D_W}-\frac{D_W\cdot cO{s}^{2}a}{D_{pw}}\right)$$

$$\text{Calculation of}\ddot{\text{u}}\text{rolling frequency on the inner ring}\to {\mathrm{and}}_j=\frac{\mathrm{e.g}}{2}\cdot \left(1+\frac{D_W\cdot cO{s}^{2}a}{D_{pw}}\right)$$

$$\text{Calculation of}\ddot{\text{u}}\text{rolling frequency on the inner ring}\to {\mathrm{and}}_A=\frac{\mathrm{e.g}}{2}\cdot \left(1-\frac{D_W\cdot cO{s}^{2}a}{D_{pw}}\right)$$

• • Generieren von Spektren mit entsprechender Schadinformation
• • Aus Berechnung der Überrollfrequenz am Außenring uA kann im Frequenzbereich Schadenssignatur für beispielsweise Lager Außenringschaden generiert werden (Schadmuster sind bekannt (aus Literatur) bzw. liegen aus anderen Messungen vor.

In order to generate a damage pattern, for example for damage to the outer ring of the roller bearing, one can proceed as follows, cf. also in particular 3 .

• • Generation of spectra with corresponding malicious information
• • From the calculation of the rollover frequency on the outer ring, etc., damage signatures can be generated in the frequency range, for example for bearing outer ring damage (damage patterns are known (from literature) or are available from other measurements.

A standardized data set can be converted to geometric variants using kinematics.

• • Transformieren der Schadsignatur im Frequenzbereich (ƒ̂) in den Zeitbereich (ƒ) → X (jw) -> x (t) zur Erzeugung der Anregungsstöße entsprechend der Maschinen-Schadcharakteristik ƒ(uA), vgl. 3, unterer Teil.

This allows the frequency pattern according to 3 upper part can be obtained.

• • Transforming the damage signature in the frequency range (ƒ̂) into the time range (ƒ) → X (jw) -> x (t) to generate the excitation shocks according to the machine damage characteristic ƒ(uA), cf. 3 , lower part.

• • Generieren von Spektren mit entsprechender Schadinformation
• • Aus Strukturanalyse (bspw. Aus FE-Simulation) kann Eigenschwingungsform der Struktur bestimmt werden → Erzeugen von Grundschwingung ƒ(Sch_Ani), vgl. 4 oberer Teil
• • Überlagerung der Maschinen-Schadcharakteristik f(u_A) und der Grundschwingung ƒ(Sch_Ani) zur Erzeugung von synthetisch generierten Schadenscharakteristiken/Samples, vgl. 4 unterer Teil.

Alternatively or additionally, the following procedure can be used to create a damage pattern that reflects outer ring damage to the rolling bearing:

• • Generation of spectra with corresponding malicious information
• • The mode of vibration of the structure can be determined from the structural analysis (e.g. from FE simulation) → generation of fundamental mode ƒ(Sch _Ani ), cf. 4 upper part
• • Superimposition of the machine damage characteristics f(u _A ) and the fundamental oscillation ƒ(Sch _Ani ) to generate synthetically generated damage characteristics/samples, cf. 4 lower part.

Advantageously, by combining the synthetic damage characteristics or samples, the structure-borne noise signature of a system or an interested component can already be generated in the design phase, which means that the complex training required in previous monitoring systems can be carried out using real machine failures and, if necessary, using artificial intelligence , can be avoided. Without synthetic generation of a structure-borne noise signature or a damage characteristic based on the design data, at least 7 to 8 real systems that correspond to the system in question would have to be examined in tests with known failures. In contrast, due to the synthetically generated damage characteristics, the AI module already has more than 2/3 of the damage patterns to be expected, which considerably reduces the training time and effort required to refine the system. For example, the remaining, still missing or not synthetically generated damage characteristics, for example in the order of 20%, can be returned by a self-learning system.

In order to do justice to more complex damage patterns or signs of wear, the active, self-implementing database 2 is advantageously assigned a combination module 6 which combines the damage characteristics synthetically determined from the design data by the determination device 5 and thereby generates combinatorial damage characteristics.

The damage characteristics, for example the synthetically generated damage characteristics and/or the combinatorially determined damage characteristics, can advantageously be given a weighting that can be generated by a weighting module 7, in particular on the basis of the probability of occurrence of a respective damage event.

Said database 2 or the modules assigned to it for determining the damage characteristics, i.e. in particular the determination device 5 and/or the combination module 6 and/or the weighting module 7 can be part of a self-learning AI system 8 or formed by such an AI system 8 be equipped with artificial intelligence and can estimate or determine a relationship between a specific condition parameter or several condition parameters of a structural component and a damage pattern of the structural component or its current condition and/or remaining service life, the AI system having a regression analysis module, for example can, in order to be able to adjust said relationship between a parameter or a set of parameters and the actual state or the remaining service life of the structural component on the basis of changes that occur.

As the figures also show, the condition monitoring system 1 also includes a sensor system 9, which can include various sensors for measuring or detecting relevant state variables or parameters of the structural component 4 of interest, with the named sensors being able to have different properties depending on the structural component.

For example, said sensor system 9 can comprise a structure-borne noise sensor and/or a displacement sensor and/or a speed sensor and/or an acceleration sensor and/or a temperature sensor in order to be able to detect corresponding state variables on the structural component 4 or surrounding components connected thereto, for example vibration data, temperature data, lubricant data , noise emission data or other relevant status information of the structural component 4.

The state information mentioned, which is recorded and provided by the sensor system 9, can be evaluated by an evaluation device 10 and compared with the damage characteristics provided by the database 2 in order to determine the current state and/or the remaining service life of the structural component 4. As 2 shows, the named evaluation device 10 can comprise an evaluation module 11 which compares actually measured damage or condition characteristics with the synthetic damage characteristics from the reference examples 3 or the combinatorial damage characteristics formed therefrom. A pre-analysis and/or pre-processing module 12 can be provided upstream of such an evaluation module 11, which pre-processes and/or pre-analyzes the status information detected by sensors, for example using a filter or other signal processing modules.

Based on the evaluation of the evaluation device 10, a prognosis and/or trend analysis module 13 can issue a prognosis for the actual state and/or a trend for the actual state of the structural component and/or the entire machine, cf. 1 and 2 .

As the figures show, the condition monitoring system 1 also includes an adaptation device 14, which adapts the damage characteristics held by the database 2 based on the evaluations of the evaluation device 10 or based on the determined condition and/or remaining service life information.

Said adjustment device 14 is preferably part of a self-learning AI system or is formed by such an AI system 8 which provides feedback of real machine status data and/or integration into the existing reference examples 3 using artificial intelligence.

In particular, the AI system 8 can use the synthetic damage characteristics and/or the damage characteristics formed from them in a combinatorial manner as a function of relevant machine status and/or adjust environmental parameters, in particular, for example, depending on the age of the structural component 4, the machine and/or operating condition and the changes in environmental influences. As a result, the forecast can always be kept sharp and the error rate minimized.

The condition monitoring system 1 thus uses in particular a data analysis model based on synthetically generated machine operating characteristics.

For this purpose, based on the construction of a component/product, synthetic operating characteristics are generated artificially in the form of samples. Each sample corresponds to a specific damage characteristic of a certain component (e.g. rollover frequency of bearing outer ring, frequency of gear meshing of drive shaft, etc.) which can be calculated from geometry/drawing data.

By combining the specific, synthetic samples, a complex picture of all possible, measurable forms of damage is generated. This database can be compared afterwards with the measured damage characteristics (e.g. due to structure-borne noise, etc.) and the condition of the component can be evaluated. Due to the large number of variations, the comparison takes place using artificial intelligence (AI) or a comparable feature recognition.

The samples or reference examples of the damage characteristics, as well as the combination of the same, can be stored in advance in a comparison database and assigned to the respective component.

In a further expansion stage, each sample can be weighted according to the probability of occurrence (frequency of the cause of failure of the respective component).

The sample database of damage characteristics can be expanded according to the system's ability to learn. The system can also be "rewarded" by the subsequent damage analysis in order to be able to improve the detection rate for related samples (e.g. other, but similar components).

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

• JP 8144312 [0007]
• DE 10145571 A1 [0008]

Claims (10)

Device for determining the actual status and/or the remaining service life of structural components (4), for example large diameter bearings, of a work machine, in particular a construction machine, material handling machine and/or conveyor machine, with a sensor system (9) for detecting status information relating to the structural component (4 ) and an evaluation device (10) for evaluating the detected status information and determining the actual status and/or the remaining service life on the basis of a comparison with predetermined damage characteristics , characterized in that an active database device (2) is provided for making the damage characteristics available, to which a determination device (5) for determining the damage characteristics from design data of the structural component (4) and an adaptation device (14) for adapting the predetermined damage characteristics based on the condition and/or remaining service life information determined by the evaluation device (10). Device according to the preceding claim, wherein the determination device (5) for determining synthetic damage characteristics is designed to determine kinematic frequencies of the structural component (4) from geometric data of the structural component (4) in order to generate a damage frequency image for the structural component (4) from the kinematic frequencies. to generate. Device according to the preceding claim, wherein the determination device (5) comprises an adaptation module which is designed to convert frequency patterns which correspond to different damage patterns and/or types, based on the geometric data of the structural component (4) into adapted frequency patterns which correspond to different damage patterns and /or - correspond to types of the concrete structural component (4), to transform. Device according to one of the preceding claims, wherein the determination device (5) for synthetic determination of the damage characteristics has a structural analysis module for determining fundamental vibrations of the structural component (4) and a superimposition module for superimposing the determined fundamental vibration with frequency patterns which indicate different damage patterns and/or types , has, in order to produce damage characteristics which are synthetically generated by said superimposition and are adapted to the structural component (4). Device according to the preceding claim, wherein the evaluation device (10) and/or the adaptation device (14) are designed as a self-learning system and/or as part of a self-learning system (8) which is the status information recorded by sensors and/or the status information derived therefrom -Conditions and/or remaining lifetimes returned to the database (2) and/or integrated into the damage characteristics kept ready by the database (2). Device according to the preceding claim, wherein the self-learning system has a regression analysis module for determining the influence of certain damage patterns and/or certain actual states on damage-characterizing parameters of the structural component such as structure-borne noise signal reference patterns, rollover frequency images or tooth meshing frequency images of the structural component using regression analysis. Device according to one of the two preceding claims, wherein the self-learning system (8) has an AI-based estimation module for estimating relationships between recorded current state information patterns and synthetically generated damage characteristics and / or between recorded current state information patterns and a damage pattern or a Having remaining life of the structural component. Device according to one of the preceding claims, wherein the determination device (5) is assigned a combination module (6) for combining the damage characteristics synthetically generated by the determination device (5) and providing combinatorial damage characteristics, the evaluation device (10) being configured to Sensor (9) to compare detected status information with the combinatorial damage characteristics. Device according to one of the preceding claims, wherein the determination device (5) is assigned a weighting module (7) for weighting the damage characteristics on the basis of a probability of occurrence of a damage associated with the damage characteristic. Device according to one of the preceding claims, wherein the sensor system (9) has at least one sensor from the group of vibration sensors, temperature sensors, lubricant sensors, structure-borne noise sensors, acceleration sensors, displacement sensors and speed sensors, and the adaptation device (14) is designed to provided damage characteristics depending on at least one signal to adapt at least one of said sensors.

Images