EP1259941B1 - Method and device for determining the remaining serviceable life of a product - Google Patents

Abstract

A method and a device for acquiring performance quantities of a product, in particular until its technical failure, and for determining the remaining service life of the product are described. The determination of the remaining service life of the product, acquisition of service lives of the products and determination of service life threshold values are performed on the basis of performance quantities subdivided into classes (classified). Weighting factors are determined first and then these weighting factors are used to determine weighted, cumulative service lives and service life threshold values. The reliability of products is monitored in mass production.

Description

State of the art

The present invention relates to a method for determining the remaining service life of a product; The invention further relates to methods for detecting operating times until the technical failure of the product, and methods for determining operating duration thresholds of products in dependence on certain time-varying operating variables for monitoring the reliability of products, and finally the invention also relates to a product whose reliability to be monitored, arranged device for comparing the actual operating life of the product with operating time thresholds according to the preambles of the independent claims.

From the DE 195 16 481 A1 a method for a lifetime determination is known. A control device for a motor vehicle is described, which has an operating data memory in which operating variables of the motor vehicle are stored, which can give information about the probability of failure or future reliability of the control device. Essential data of the life history of a control device are stored in the operating data memory in order to be able to make a statement regarding the reliability of the control device if necessary.

From the EP 863490 A2 A method for determining a remaining operating time is already known in which quantized operating data are stored.

Purpose and advantages of the invention

The object of the present invention is to enable the most accurate, non-model-based lifetime estimation for any products that have or have access to an operating data memory. Another task is an optimal acquisition of data and storage in an operating data storage to use the memory optimally, in particular to save storage space.

In order to achieve this object, the invention proposes a method for determining an operating time threshold value of a product which has a microcomputer or microcontroller and an operating data memory connected by a communication system (KS). A monitoring of the reliability of the product is carried out by comparing an operating time with a threshold value, whereby values of value ranges of predefinable operating quantities of the product are detected. The respective value range of the respective operating size is in classes. The values and / or the operating times (t_ijk) are stored in the operating data memory (BSi, BSe) assigned to the product corresponding to the classes (j). From a set of products, a first subset of the products is operated until technical failure, whereby the service life of the classes of the predetermined operating sizes of the product are determined. For each class and farm size, a weighting factor is determined which reflects the influence of the technical failure of the product of the respective class and farm size. A second subset of products is operated until technical failure, with the weighting factors determined from the first subset applied to the second subset. In the case of the second subset of the product, a critical service life is determined for all operating classes across all classes, and the operating time threshold value is determined from the critical operating times across all classes of all operating variables.

The product whose operating time is detected until the technical failure is, for example, designed as a control unit or a subsystem (eg brake, engine, transmission, steering, etc.) of a motor vehicle. The products have an operating data memory or are assigned to one in which the recorded operating variables or the operating periods can be stored and recalled if necessary. The operating data memory preferably has a nonvolatile memory (eg an EEPROM or a flash EEPROM) as well as means for detecting the operating variables or the operating times. In a motor vehicle, the operating data memory can be realized, for example, in one or more control units.

The operating data memories are used to record discrete system states (eg number of starts, number of emergency starts, number of thermal shutdowns, etc.) as well as the time-varying operating variables. As operating variables, for example, sensor data such as temperature, current, voltage, pressure u. a. detected.

In the value range of the operating variables permissible under operating conditions, a linear or non-linear subdivision of the value range into several classes is carried out in each case. Extreme values that lead to immediate destruction of the product are outside the permissible value range. The class assignment is based on the division of the entire value range into relevant load groups. The individual classes have a different influence on aging / wear of the product. The operating data memory records the operating life of the product for each operating quantity in each class.

According to the present invention, the determination of the individual technical operating life of a product and the calculation of the degree of wear at any given time are performed on the basis of classed operating variables. Due to the classified operating variables, a particularly reliable and accurate determination of the operating life of a product is possible, whereby the memory requirement for the operating data memory is minimized, as it is possible to dispense with a recording of time profiles of the operating variables. This allows a particularly reliable preventive maintenance / repair shortly before reaching the end of the technical service life.

• der Anzahl der Betriebsgrößen,
• der durchschnittlichen Anzahl der Klassen pro Betriebsgröße und
• der durchschnittlichen Anzahl Bytes pro Klassenzähler.
  Das erfindungsgemäße Verfahren zum Erfassen von Betriebsdauern auf der Basis klassierter Betriebsgrößen hat insbesondere beim Bestimmen von Betriebsdauer-Schwellenwerten von Produkten für eine Überwachung der Zuverlässigkeit von Produkten besondere Vorteile. Deshalb wird gemäß einer vorteilhaften Weiterbildung der vorliegenden Erfindung ein Verfahren zum Bestimmen von Betriebsdauer-Schwellenwerten der eingangs genannten Art vorgeschlagen, das dadurch gekennzeichnet ist, dass
• die Betriebsdauern der Produkte für die Klassen der Betriebsgrößen bis zum technischen Versagen des Produkts durch Verwendung des Verfahrens nach Anspruch 1 oder 2 bestimmt werden;
• den Klassen der Betriebsgrößen Gewichtungsfaktoren zugewiesen werden;
• die Gewichtungsfaktoren aus der Lösung eines Optimierungsproblems $$\min \left\{\mathrm{f}\left(\mathrm{x}\right)\right\},\mathrm{mit\; x}=\left\{\mathrm{a\_ij},,\mathrm{t\_ijk}\right\}$$
  
  
  unter Berücksichtigung der Korrelation zwischen den einzelnen Betriebsgrößen ermittelt werden;
• für die Produkte kritische kumulierte Betriebsdauern für die einzelnen Betriebsgrößen aus der Gleichung $$\mathrm{P\_iz\_krit}\mathrm{=}\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijz}\right\}$$
  
  
  ermittelt werden; und
• für die einzelnen Produkte die Betriebsdauer-Schwellenwerte aus der Gleichung $$\begin{array}{c}\min \left\{\mathrm{P\_iz\_krit}\right\},\end{array}$$
  
  
  mit i = 1...N oder $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\},\mathrm{mit\; i}=1\dots \mathrm{N}$$
  
  
  ermittelt werden.

According to a preferred development of the present invention, it is proposed that the values of the operating variables are recorded at regular time intervals and a class counter of a certain class is increased if the value of a recorded operating variable falls within this class. Each operating quantity of a particular product may thus be assigned, after detecting the operating times, an operating time histogram which gives the operating life of the product for the operating quantity within a particular class. The size of the operating data memory in bytes required for the operating data storage results from the product

• the number of farm sizes,
• the average number of classes per farm size and
• the average number of bytes per class counter.
  The method according to the invention for detecting operating times on the basis of classified operating variables has particular advantages, in particular when determining operating time threshold values of products for monitoring the reliability of products. Therefore, according to an advantageous development of the present invention, a method for determining operating time threshold values of the type mentioned in the introduction is proposed, which is characterized in that
• the operating lives of the products are determined for the classes of operating quantities until the technical failure of the product by using the method according to claim 1 or 2;
• the weighting factors are assigned to the classes of farm sizes;
• the weighting factors from the solution of an optimization problem $$\min \left\{\mathrm{f}\left(\mathrm{x}\right)\right\}.\mathrm{with\; x}=\left\{\mathrm{a\_ij},,\mathrm{t\_ijk}\right\}$$
  
  
  taking into account the correlation between the individual farm sizes;
• for the products critical cumulative operating times for the individual operating variables from the equation $$\mathrm{P\_iz\_krit}\mathrm{=}\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijz}\right\}$$
  
  
  be determined; and
• for the individual products, the operating time thresholds from the equation $$\begin{array}{c}\min \left\{\mathrm{P\_iz\_krit}\right\}.\end{array}$$
  
  
  with i = 1 ... N or $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\}.\mathrm{with\; i}=1...\mathrm{N}$$
  
  
  be determined.

The individual classes have a different influence on aging / wear of the products. Therefore, weighting factors are assigned to the classes of farm sizes that express the relative impact of a particular class of a particular farm size on the aging or wear of the product. The invention provides to determine the weighting factors from a subset K of the products and then apply these to the subset Z of the products. As a result, the critical weighted cumulative operating periods of the operating variables for series production can be determined for the products from the subset S, upon reaching which an end of the technical service life can be concluded.

unter Berücksichtigung der Korrelation zwischen den einzelnen Betriebsgrößen ermittelt, wobei a_ij der Gewichtungsfaktor ist, der der Klasse j der Betriebsgröße i zugewiesen ist, und t_ijk die Betriebsdauer des Produkts k für die Klasse j der Betriebsgröße i ist. Die Korrelation zwischen den Betriebsgrößen kann bspw. dadurch berücksichtigt werden, dass die Gewichtungsfaktoren aus einem Gleichungssystem bestimmt werden, in dem die gewichteten kumulierten Betriebsdauern für jede Betriebsgröße mittels Operatoren miteinander verknüpft werden. Die Operatoren können bspw. eine UND-Verknüpfung (Produktbildung), eine ODER-Verknüpfung (Summenbildung) oder eine Fuzzy-Verknüpfung (z. B. einen Zwischenzustand zwischen UND und ODER) darstellen.The weighting factors become the solution of an optimization problem $$\min \left\{\mathrm{f}\left(\mathrm{x}\right)\right\}.\mathrm{with\; x}=\left\{\mathrm{a\_ij},,\mathrm{t\_ijk}\right\}$$

taking into account the correlation between the individual operating quantities, where a_ij is the weighting factor assigned to the class j of the operating quantity i, and t_ijk is the operating life of the product k for the class j of the operating variable i. The correlation between the operating variables can, for example, be taken into account by determining the weighting factors from an equation system in which the weighted cumulative operating times for each operating variable are linked to one another by means of operators. The operators may, for example, represent an AND operation (product formation), an OR operation (summation) or a fuzzy operation (eg an intermediate state between AND and OR).

für alle Betriebsgrößen und für alle Z Produkte ermittelt, wobei P_iz_krit die kritische kumulierte Betriebsdauer des Produkts z der Betriebsgröße i ist und t_ijz die Betriebsdauer des Produkts z für die Klasse j der Betriebsgröße i ist. Damit erhält man Z Vektoren der gewichteten kumulierten Betriebsdauern $$\begin{array}{c}\mathrm{Y\_z}=\left(\mathrm{P\_}1\mathrm{z\_krit},\mathrm{P\_}2\mathrm{z\_krit},\dots ,\mathrm{P\_Nz\_krit}\right),\mathrm{mit\; z}=1\dots \mathrm{Z}\end{array}$$

After the weighting factors by solving an optimization problem with appropriate mathematical Optimization algorithms were determined, the critical accumulated operating times for the individual operating variables are set, which can be concluded when reaching an end of the technical service life. For this purpose, with the help of K products, a number of Z products are operated until technical failure, whereby the weighting factors calculated from the K products are applied to the classified operating variables of the Z products. It will $$\mathrm{P\_iz\_krit}\mathrm{=}\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijz}\right\}$$

for all operating variables and for all Z products, where P_iz_crit is the critical cumulative operating life of the product z of the operating variable i and t_ijz is the operating life of the product z for the class j of the operating variable i. This yields Z vectors of the weighted accumulated operating lives $$\begin{array}{c}\mathrm{Y\; Z}=\left(\mathrm{P\_}1\mathrm{z\_krit}.\mathrm{P\_}2\mathrm{z\_krit}.....\mathrm{P\_Nz\_krit}\right).\mathrm{with\; z}=1...\mathrm{Z}\end{array}$$

oder aus dem Durchschnitt der Spaltenelemente der Matrix Y_z gemäß der Gleichung $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\},\mathrm{mit\; i}=1\dots \mathrm{N}$$

ermittelt. Das funktioniert mit der geforderten Zuverlässigkeit, wenn die einzelnen Spaltenelemente hinreichend dicht beieinander liegen, d. h. wenn die Standardabweichung der Spaltenelemente nicht zu gross ist. Ausreisser sollen bei der Auswahl der Spaltenminima nicht berücksichtigt werden.For the individual products, the operating time thresholds, upon reaching which a technical end of life of the product can be concluded, are calculated from the column minimums of the matrix Y_z according to the equation $$\min \left\{\mathrm{P\_iz\_krit}\right\}.\mathrm{with\; i}=1...\mathrm{N}$$

or from the average of the column elements of the matrix Y_z according to the equation $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\}.\mathrm{with\; i}=1...\mathrm{N}$$

determined. This works with the required reliability when the individual column elements are sufficiently close to each other, ie if the standard deviation of the column elements is not too large. Outliers should not be taken into account when selecting the column minima.

After the critical accumulated operating times have been determined for the individual operating variables, for all series products equipped with operating data memories shortly before reaching the critical threshold value, the need for repair, replacement or maintenance by the product can be signaled. Alternatively, the operating variables stored in the product are evaluated as part of regular product maintenance.

In summary, therefore, k = 1... K products are initially operated until technical failure in order to be able to determine the weighting factors a_ij. Thereafter, the weighting factors a_ij are integrated into the operating data memories of z = 1... Z products, which are again operated until technical failure, the critical cumulative operating times P_iz_crit and the operating time threshold values via a minimum selection or the average of the critical cumulative operating times P_iz_crit to investigate. Thereafter, the reliability of s = 1 ... S products is monitored in series production, with the actual operating life of a product s being compared to a threshold value.

ermittelt werden, mit der Ungleichungsnebenbedingung a_ij > 0, wobei a_ij der Gewichtungsfaktor ist, der der Klasse j der Betriebsgröße i zugewiesen ist, und t_ijk die Betriebsdauer des Produkts k für die Klasse j der Betriebsgröße i ist. Gemäß dieser Ausführungsform wird bei der Berechnung der Gewichtungsfaktoren keine Korrelation zwischen den einzelnen Betriebsgrössen berücksichtigt. Es wird also von der Annahme ausgegangen, dass jede Betriebsgröße unabhängig von den Werten der anderen Betriebsgrößen zur technischen Zerstörung des Produktes führen kann.According to a preferred embodiment of the present invention, it is proposed that the weighting factors from the solution of the optimization problem $$\min \left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{k}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\mathrm{SECTION}\{\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\{\mathrm{a\_ij}\times \mathrm{t\_ijk}\}-1\}\right\}$$

to be detected, with the inequality constraint a_ij> 0, where a_ij is the weighting factor assigned to the class j of the operation quantity i, and t_ijk is the operation time of the product k for the class j of the operation quantity i. According to this embodiment, no correlation between the individual operating variables is taken into account in the calculation of the weighting factors. It is therefore assumed that each size of operation can lead to the technical destruction of the product, regardless of the values of the other operating variables.

If no correlation between the individual operating variables is used to determine the weighting factors, the largest ratio of a weighted cumulative operating period for an operating variable to the critical threshold of the operating variable can be interpreted as a degree of wear. The remaining life in% is then calculated according to $$\mathrm{Remaining\; life}\left[\%\right]=1-\mathrm{degree\; of\; wear}\left[\%\right]$$

ermittelt werden, mit der Ungleichungsnebenbedingung a_ij > 0. Bei dieser Ausführungsform wird die Korrelation zwischen den einzelnen Betriebsgrößen berücksichtigt. Es wird also von der Annahme ausgegangen, dass mehrere Betriebsgrößen gemeinsam zur technischen Zerstörung des Produktes führen. Gemäß dieser Ausführungsform sind die Betriebsgrößen mittels reiner UND-Verknüpfungen (Produktbildung) miteinander verknüpft. Die Gewichtungsfaktoren werden derart bestimmt, dass die durch den UND-Operator verknüpften gewichteten Klassensummen eines jeden Produktes einen minimalen "Abstand" zueinander besitzen.According to an alternative embodiment of the present invention, it is proposed that the weighting factors result from the solution of the optimization problem $$\begin{array}{c}\min \left\{\stackrel{\mathrm{K}}{\underset{\mathrm{v}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{\mu }\mathrm{\ne }\mathrm{v}}{\underset{\mathrm{\mu }\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}}\mathrm{SECTION}\left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\left\{\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ij\mu }\right\}\right\}\right\}\mathrm{-}\mathrm{...}\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\mathrm{\{}\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijv}\right\}\mathrm{\}}\mathrm{\}}\right\}\end{array}$$

to be determined, with the inequality constraint a_ij> 0. In this embodiment, the correlation between the individual operation quantities is taken into account. It is therefore assumed that several Operating variables together lead to the technical destruction of the product. According to this embodiment, the operating variables are linked to one another by means of pure AND operations (product formation). The weighting factors are determined such that the weighted class sums of each product associated with the AND operator have a minimal "distance" from each other.

According to a third alternative embodiment is intended to link multiple operating variables at the level of individual classes. It is assumed that several operating variables within certain classes lead to a technical destruction of the product.

To solve the object of the present invention, starting from a device for detecting the operating times until the technical failure of a product further proposed that the device comprises first means for detecting the values of certain operating variables at regular time intervals, the range of values of the individual operating variables in Classes is divided and the device comprises second means for detecting the operating times depending on the class in which the detected value of the operating variable falls.

According to an advantageous development of the invention, it is proposed that the second means increase a class counter of a certain class if the value of a recorded operating variable falls within this class.

The device according to the invention for detecting operating times on the basis of classified operating variables has in particular when determining operating time threshold values of products for monitoring the reliability of products special benefits. Therefore, according to an advantageous development of the present invention, a device for determining operating time threshold values of the type mentioned in the introduction is proposed, which is characterized in that the device has means for carrying out the method according to one of claims 5 to 8.

To achieve the object of the present invention, starting from a device of the type mentioned above, which is arranged in a product to be monitored, it is proposed that the operating time threshold values are determined according to the method according to one of claims 5 to 8. The operating data memory of the device can be designed to be particularly small, since a determination of the operating time threshold values according to the invention can be dispensed with a memory-intensive detection of time profiles of the operating variables.

In addition, an operational data acquisition in classes has the particular advantage that the memory can be used optimally, ie in particular only little storage space is needed, since no complex acquisition of operating variables over the entire time axis, or with respect to the time axis must be performed. As a result, the invention, in particular the operating data acquisition, may be realized as additional functionality in a control device or in a dedicated device.

Further advantages and advantageous embodiments will become apparent from the description and the features of the claims.

drawings

Fig. 1

ein Ablaufdiagramm eines erfindungsgemäßen Verfahrens zum Erfassen von Betriebsdauern bis zum technischen Versagen eines Produkts gemäß einer bevorzugten Ausführungsform; und

Fig. 2

ein Ablaufdiagramm eines erfindungsgemäßen Verfahrens zum Bestimmen von Betriebsdauer-Schwellenwerten von Produkten gemäß einer bevorzugten Ausführungsform.

A preferred embodiment of the present invention will be explained in more detail below with reference to the drawings. It shows:

Fig. 1

a flowchart of a method according to the invention for detecting operating times to the technical failure of a product according to a preferred embodiment; and

Fig. 2

a flowchart of a method according to the invention for determining operating duration thresholds of products according to a preferred embodiment.

Description of the embodiment

In Fig. 1 1 is a flow chart of a method according to the invention for detecting operating times t_ijk of a product k = 1... K until the technical failure of the product k according to a preferred embodiment is shown. The product k, whose operating time t_ijk is detected, is designed, for example, as a control device or a subsystem (eg brake, engine, transmission, steering, etc.) of a motor vehicle. The product k has an operating data memory in which recorded operating variables i = 1... N or the operating times t_ijk can be stored and recalled if necessary. The operating data memory preferably has a nonvolatile memory (eg an EEPROM or a flash EEPROM) as well as means for detecting the operating variables or operating times. In a motor vehicle, the operating data memory can be realized, for example, in one or more control units.

The operating data memories are used to record discrete system states (eg number of starts, number of emergency starts, number of thermal shutdowns, etc.) and the time-varying operating variables i. As operating variables i, for example, sensor data such as temperature, current, voltage, pressure u. a. detected.

The method begins in a function block 10. In a function block 11, the range of values of the individual operating variables i that are permissible under operating conditions is subdivided linearly or nonlinearly into classes j = 1... M_i. Extreme values that lead to the immediate destruction of the product k are outside the permissible value range. The class assignment is based on the division of the entire value range into relevant load groups. The individual classes j have a different influence on aging / wear of the product k.

In a subsequent function block 12, values of the operating variables i are acquired at regular time intervals. The operating times t_ijk are detected as a function of the class j, in which the detected value of the operating variable i falls. For this purpose, in a function block 13, a class counter of a certain class j is incremented if the value of the acquired operating variable i falls within this class j. Each operating quantity i of a specific product k can thus be assigned, after detecting the operating times t_ijk, an operating time histogram from which the operating time t_ijk of the product k for the operating quantity i within a specific class j results. The operating times t_ijk result from the product of the state of the class counters and the time interval between the recorded values of the operating variables i.

In a subsequent query block 14, it is checked whether the detection of the operating times t_ijk has ended. If not, a branch is made to the function block 12 again. If the detection of the operating times t_ijk has ended, the program branches to the end of the method in function block 15.

In Fig. 2 FIG. 3 shows a flowchart of a method according to the invention for determining operating time threshold values of the products z according to a preferred embodiment. The method according to the invention begins in a functional block 20. Then, first the operating times t_ijk of the products k for the class j of the operating variables i up to the technical failure of the product k are determined by using the method according to FIG Fig. 1 certainly.

Subsequently, in a function block 21, the classes of the operating variables i are assigned weighting factors a_ij. Since the individual classes j have a different influence on the aging / wear of the products k, the classes j of the operating variables i are assigned weighting factors a_ij which determine the relative influence of a particular class j of a specific operating variable i on the aging or wear of the product k expresses.

unter Berücksichtigung der Korrelation zwischen den einzelnen Betriebsgrößen i ermittelt. Die Gewichtungsfaktoren a_ij können bspw. aus der Lösung des Optimierungsproblems $$\min \left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{k}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\mathrm{ABS}\{\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\{\mathrm{a\_ij}\times \mathrm{t\_ijk}\}-1\}\right\}$$

ermittelt werden, mit der Ungleichungsnebenbedingung a_ij > 0. Dabei wird keine Korrelation zwischen den einzelnen Betriebsgrößen berücksichtigt und von der Annahme ausgegangen, dass jede Betriebsgröße i unabhängig von den Werten der anderen Betriebsgrößen i zur technischen Zerstörung des Produktes k führen kann.In a subsequent function block 22, the weighting factors a_ij become the solution to an optimization problem $$\min \left\{\mathrm{f}\left(\mathrm{x}\right)\right\}.\mathrm{with\; x}=\left\{\mathrm{a\_ij},,\mathrm{t\_ijk}\right\}$$

taking into account the correlation between the individual operating variables i determined. The weighting factors a_ij can, for example, from the solution of the optimization problem $$\min \left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{k}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\mathrm{SECTION}\{\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\{\mathrm{a\_ij}\times \mathrm{t\_ijk}\}-1\}\right\}$$

In this case, no correlation between the individual operating variables is taken into account and it is assumed that each operating variable i, irrespective of the values of the other operating variables i, can lead to the technical destruction of the product k.

ermittelt werden, mit der Ungleichungsnebenbedingung a_ij > 0. Es wird die Korrelation zwischen den einzelnen Betriebsgrößen i berücksichtigt und von der Annahme ausgegangen, dass mehrere Betriebsgrößen i gemeinsam zur technischen Zerstörung des Produktes k führen. Die Betriebsgrößen i sind in dem Ausführungsbeispiel mittels reiner UND-Verknüpfungen (Produktbildung) miteinander verknüpft.Alternatively, the weighting factors a_ij can also be obtained from the solution of the optimization problem $$\begin{array}{c}\min \left\{\stackrel{\mathrm{K}}{\underset{\mathrm{v}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{\mu }\mathrm{\ne }\mathrm{v}}{\underset{\mathrm{\mu }\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}}\mathrm{SECTION}\left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\left\{\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ij\mu }\right\}\right\}\right\}\mathrm{-}\mathrm{...}\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\mathrm{\{}\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijv}\right\}\mathrm{\}}\mathrm{\}}\right\}\end{array}$$

with the inequality constraint a_ij> 0. The correlation between the individual operating variables i is taken into account and it is assumed that several operating variables i together lead to the technical destruction of the product k. The operating variables i are linked together in the exemplary embodiment by means of pure AND operations (product formation).

According to a third alternative, a combination of a plurality of operating variables i on the level of individual classes j is conceivable. It is assumed that several operating variables i within certain classes j lead to a technical destruction of the product k.

The invention provides for determining the weighting factors a_ij from a subset K of the products k and then applying them to the subset Z of the products z. As a result, critical cumulative operating periods P_iz_krit of the operating variables i can be determined for series production, which can be concluded when they reach an end of the technical operating period.

ermittelt, indem die Produkte z bis zum technischen Versagen betrieben werden. Damit erhält man Z Vektoren der gewichteten kumulierten Betriebsdauern $$\begin{array}{c}\mathrm{Y\_z}=\left(\mathrm{P\_}1\mathrm{z\_krit},\mathrm{P\_}2\mathrm{z\_krit},\dots ,\mathrm{P\_Nz\_krit}\right),\mathrm{mit\; z}=1\dots \mathrm{Z}\end{array}$$

In a function block 23, critical cumulative operating times P_iz_crit for the individual operating variables i for the products z are then obtained from the equation $$\mathrm{P\_iz\_krit}\mathrm{=}\stackrel{\mathrm{m\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijz}\right\}$$

determined by operating the products z until technical failure. This yields Z vectors of the weighted accumulated operating lives $$\begin{array}{c}\mathrm{Y\; Z}=\left(\mathrm{P\_}1\mathrm{z\_krit}.\mathrm{P\_}2\mathrm{z\_krit}.....\mathrm{P\_Nz\_krit}\right).\mathrm{with\; z}=1...\mathrm{Z}\end{array}$$

oder aus dem Durchschnitt der Spaltenelemente der Matrix Y_z gemäß der Gleichung $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\},\mathrm{mit\; i}=1\dots \mathrm{N}$$

ermittelt. Das funktioniert dann mit der geforderten Zuverlässigkeit, wenn die einzelnen Spaltenelemente hinreichend dicht beieinander liegen, d. h. wenn die Standardabweichung der Spaltenelemente klein ist.For the individual products z, finally, in function block 24, the operating time threshold values, upon reaching which an immediate technical end of life of the product can be concluded, are determined from the column minimums of the matrix Y_z according to the equation $$\min \left\{\mathrm{P\_iz\_krit}\right\}.\mathrm{with\; i}=1...\mathrm{N}$$

or from the average of the column elements of the matrix Y_z according to the equation $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\}.\mathrm{with\; i}=1...\mathrm{N}$$

determined. This then works with the required reliability if the individual column elements are sufficiently close to each other, ie if the standard deviation of the column elements is small.

Outliers, if present, should therefore not be taken into account when selecting the column minima. In function block 25, the method for determining operating time thresholds of the products z is completed. In addition to absolute or relative minimum selection and simple averaging, other methods and procedures, such as sliding or empirical or harmonic averaging or meridian formation, etc., may be used to determine the operating duration thresholds.

After the critical accumulated operating times P_iz_krit have been determined for the individual operating variables i, all the data items s equipped with operating data memories can be signaled by the product s shortly before reaching the critical threshold value, the need for repair, replacement or maintenance. This can be done in particular in the form of a self-diagnosis of the series product. Alternatively, the operating variables stored in the product s are evaluated as part of regular product maintenance. This product maintenance may then also be e.g. in a partial product of a vehicle, or the vehicle itself in operation itself be carried out in the form of an on-board diagnosis.

FIG. 3 schematically shows a possible device according to the invention. P is the product itself. This is connected by a communication system KS, in particular a line or bus system, with a product external operating data memory BSe. Alternatively, an internal operating data memory BSi may be provided in the product itself. Both memories can also be present at the same time and, for example, a virtual memory can be formed from BSe and BSi. In M, the funds are summarized, for example in the form of a Microcomputers or microcontroller, which are used for carrying out the inventive method as shown above. These means may, for example, be present or incorporated in a control unit of a motor vehicle.

The product P whose operating time is detected is embodied, for example, as a control device or a subsystem (eg brake, engine, transmission, steering, and the like) of a motor vehicle. The products P have an operating data memory BSi or are assigned to one (BSe) in which the recorded operating variables or the operating periods can be stored and recalled if necessary. The operating data memory preferably has a nonvolatile memory (eg an EEPROM or a flash memory) as well as means EM for detecting the operating variables or the operating times. In a motor vehicle, the operating data memory can be realized, for example, in one or more control units. The detection means EM obtain their information e.g. via the communication system KS or other interfaces of the product, e.g. to other sensors or actuators. The evaluation, operating time detection, operating time determination by threshold value comparison, etc. is carried out in particular by the means M, which also initiate or carry out the signaling or the initiation of further measures. The detection means EM and the means M can also be present in combination, that is to say in a combined manner, and can likewise be assigned to the operating data memories in a targeted manner or integrated into them.

The operating data memories are used to record discrete system states (eg number of starts, number of emergency starts, number of thermal shutdowns, etc.) as well as the time-varying operating variables. When Operating variables are, for example, sensor data such as temperature, current, voltage, pressure, etc. recorded. The necessary sensor technology is connected, for example, via the communication system KS or coupled with the product via further interfaces. Depending on the product, the sensor system may also be partially or completely integrated in the product. The same applies to particular inventive information delivering actuator.

Thus, in all serial products s equipped with operating data memories, the need for repair, replacement or maintenance by the product s can be signaled shortly before the critical threshold value is reached. This may in particular also take the form of a self-diagnosis of the series product s, e.g. by operating data memory with integrated means M or detection means EM.

Claims (4)

1. Method for determining an operating life threshold value for a product which has a microcomputer or microcontroller, and an operating data memory which is connected by means of a communication system (KS) or is internal, for monitoring the reliability of the product by comparing an operating life with a threshold value, with values of value ranges of operating variables of the product which can be predetermined being recorded, with the respective value range of the respective operating variable being subdivided into classes, with sensor data being recorded as operating variables, characterized in that the values and/or the operating lives (t_ijk) are stored on the basis of the classes (j) in the operating data memory (BSi, BSe) which is associated with that product, and in that from a set of products a first subset of the products is operated until technical failure, by which means the operating lives of the classes of the operating variables of the product which can be predetermined are determined, with this being used to determine a weighting factor for each class and operating variable, which weighting factor reflects the influence on technical failure of the product of the respective class and operating variable, and a second subset of the products is operated until technical failure, with the weighting factors which have been determined from the first subset being applied to the second subset, and a critical operating life for all the classes being determined for each operating variable for the second subset of the product, and the operating life threshold value for all the classes of all the operating variables being determined from the critical operating lives.
2. Method according to Claim 1, characterized in that the weighting factors (a_ij) are determined from the solution of an optimization problem $$\min \left\{\mathrm{f}\left(\mathrm{x}\right)\right\},\mathrm{where\; x}=\left(\mathrm{a\_ij},,\mathrm{t\_ijk}\right)$$

   

   
   taking account of the correlation between the individual operating variables;

   - critical accumulated operating lives (P_iz_krit) for the individual operating variables (i) of the products (z) are determined from the equation $$\mathrm{P\_iz\_krit}=\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}=1}{\mathrm{SUM}}}(\mathrm{a\_ij}\times \mathrm{t\_ijz});$$

   

   
   and

   - the operating life threshold values for the individual products (z) are determined from the equation $$\min \left\{\mathrm{P\_iz\_krit}\right\},\mathrm{where\; i}=1\dots \mathrm{N}$$

   

   
   or $$\frac{1}{\mathrm{N}}\times \stackrel{\mathrm{N}}{\underset{\mathrm{i}=1}{\mathrm{SUM}}}\left\{\mathrm{P\_iz\_krit}\right\},\mathrm{where\; i}=1\dots \mathrm{N}\mathrm{.}$$

   
3. Method according to Claim 1, characterized in that the weighting factors (a_ij) are determined from the solution of the optimization problem $$\min \left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{k}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\mathrm{ABS}\{\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\{\mathrm{a\_ij}\times \mathrm{t\_ijk}\}-1\}\right\},$$

   

   
   with the secondary inequality condition a_ij > 0.
4. Method according to Claim 1, characterized in that the weighting factors (a_ij) are determined from the solution of the optimization problem: $$\begin{array}{c}\min \left\{\stackrel{\mathrm{K}}{\underset{\mathrm{v}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\stackrel{\mathrm{K}}{\underset{\mathrm{\mu }\mathrm{\ne }\mathrm{v}}{\underset{\mathrm{\mu }\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}}\mathrm{ABS}\left\{\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\left\{\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ij\mu }\right\}\right\}\right\}\mathrm{-}\mathrm{\dots }\stackrel{\mathrm{N}}{\underset{\mathrm{i}\mathrm{=}\mathrm{1}}{\mathrm{PROD}}}\mathrm{\{}\stackrel{\mathrm{M\_i}}{\underset{\mathrm{j}\mathrm{=}\mathrm{1}}{\mathrm{SUM}}}\left\{\mathrm{a\_ij}\mathrm{\times }\mathrm{t\_ijv}\right\}\mathrm{\}}\mathrm{\}}\right\}\end{array},$$

   

   
   using the secondary inequality condition a_ij > 0.

Images