Method for investigating a functional behavior of a component of a technical installation, computer program, and computer-readable storage medium

Abstract

An improved method for investigating a functional behavior of a component of a technical installation includes comparing a signal of the component to be investigated and representing the functional behavior of the component with a reference signal which describes an average functional behavior of identical components. During the comparison, a comparison variable describing the deviation of the signal from the reference signal is determined. In addition, a probability of the occurrence of the comparison variable is determined by using a predefinable distribution of a multiplicity of such comparative variables. A computer program and a computer readable storage medium are also provided.

Claims

1. A method for investigating a functional behavior of a component of a technical installation, the method comprising the following steps: comparing a signal of the component to be investigated and representing the functional behavior of the component to be investigated with a reference signal describing an average functional behavior of identical components; including a plurality of operating parameter values as a function of time in the signal of the component to be investigated; including a plurality of reference values as a function of time in the reference signal; accumulating the operating parameter values over time; accumulating the reference values over time; determining a maximum cumulative deviation between the accumulated operating values X.sub.Ri and the accumulated reference values as the comparison variable; and during the comparing step: determining a comparison variable describing a deviation of the signal from the reference signal, determining a probability of occurrence of the comparison variable by using a definable distribution of a plurality of such comparison variables; and detecting an abnormal behavior of the component based on the probability of occurrence of the comparison variable.

2. The method according to claim 1, which further comprises determining the comparison variable by using a statistical test.

3. The method according to claim 1, which further comprises determining the comparison variable by using a Kolmogorov-Smirnov test.

4. The method according to claim 1, which further comprises defining the comparison variable as a maximum Euclidean distance between the signal of the component to be investigated and the reference signal.

5. The method according to claim 1, which further comprises: using the signal of the component to be investigated to describe the functional behavior of the component within a specified time interval; and using the reference signal to describe an average functional behavior of identical components within the same time interval.

6. The method according to claim 1, which further comprises defining the reference signal as an average over a plurality of signals of a plurality of identical components of the same technical installation.

7. The method according to claim 1, which further comprises using another component of the same component type as an identical component.

8. The method according to claim 1, which further comprises using another component of the same technical installation as an identical component.

9. The method according to claim 1, which further comprises selecting an identical component as a component of another component type or of the same technical installation, the identical component responding in the same way as the component to be investigated to influences including external influences.

10. The method according to claim 1, which further comprises detecting an abnormal behavior of the component to be investigated, when the probability of occurrence of the comparison variable is lower than a specified limit value.

11. The method according to claim 1, which further comprises determining the distribution of the multiplicity of such comparison variables as follows: for multiple time intervals: comparing a signal of a respective time interval representing the functional behavior of at least one identical component with a respective reference signal describing an average functional behavior of identical components within the same respective time interval; and during each comparison: determining a comparison variable describing the deviation of the signal from the reference signal in each case, and determining the distribution of the plurality of such comparison variables based on the multiplicity of comparison variables.

12. The method according to claim 1, which further comprises determining the distribution of the plurality of such comparison variables as follows: for multiple components of the same component type: comparing a signal of a respective component representing the functional behavior of the respective component with a reference signal describing an average functional behavior of identical components; and during each comparison: determining a comparison variable describing the deviation of the signal from the reference signal in each case, and determining the distribution of the plurality of such comparison variables based on the multiplicity of comparison variables.

13. A non-transitory computer program product comprising instructions that when executed by a processor, perform the steps according to claim 1.

14. A non-transitory computer readable storage medium comprising instructions stored thereon, that when executed by a processor, perform the steps according to claim 1.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

(1) Shown are:

(2) FIG. 1 a first graph showing a signal of a component to be investigated representing the functional behavior of the component, and a reference signal which describes an average functional behavior of identical components,

(3) FIG. 2 a second graph for determining a comparison variable which describes the deviation of the signal of the component to be investigated from FIG. 1 from the reference signal of FIG. 1, and

(4) FIG. 3 a third graph for determining a probability of occurrence of the comparison variable of FIG. 2 using a specified distribution of a plurality of such distribution variables.

DETAILED DESCRIPTION OF THE INVENTION

(5) FIG. 1 shows a diagram 2 with a signal 4 of a component to be investigated, representing the functional behavior of the component. The diagram further comprises a reference signal 6, which describes an average functional behavior of identical components.

(6) On the x-axis 8 of the graph 2 the time t is plotted. On the y-axis 10 of the graph 2 an operating parameter is plotted.

(7) In this example, the component to be investigated is a wheelset bearing of a rail vehicle. The identical components are other wheelset bearings of the same rail vehicle, in particular all the other wheelset bearings of the same rail vehicle. Each wheelset bearing comprises a sensor which detects values of an operating parameter. The operating parameter is sensitive to damage to the wheelset bearing.

(8) In this example, the operating parameter is a temperature of the respective component, in this case the respective wheelset bearing.

(9) The sensors determine (quasi-)continuous values of the operating parameter, operating parameter values for short, in this example with a frequency of f=1 min.sup.−1.

(10) Each wheelset bearing can also comprise a plurality of sensors whose signals are then averaged to form a signal of the respective component. Further, the operating parameter values can be smoothed.

(11) For each wheelset bearing R.sub.i with i=1, 2, . . . N each signal comprises operating parameter values x.sub.Ri(t) as a function of time t.

(12) In this example, the component R.sub.1 (i.e. i=1) is the component to be investigated. The signal 4 of the component to be investigated comprises the operating parameters x.sub.Ri with i=1.

(13) The reference signal 6 is calculated by calculating the expected curve {circumflex over (X)}.sub.Ri(t) of all other wheelset bearings R.sub.j with j≠i. To this end a mean value is determined, here an arithmetic mean, of the operating parameters x.sub.Rj of the wheelset bearings R.sub.j with j≠i:
{circumflex over (x)}.sub.R.sub.i(t)−avg(x.sub.R.sub.i(t)) with J≠i, here j=2 . . . N

(14) In principle, the mean value could also be a median, a modal value or a quantile.

(15) In principle, it is possible that statistical and/or systematic fluctuations of at least one signal can be allowed for using a correction factor for the fluctuating signal. For the sake of better clarity, no correction factors have been introduced here.

(16) As an example of the signal 4 of the component to be investigated, FIG. 1 shows the progression over time of the temperature of the wheelset bearing to be tested (R.sub.i with i=1) within a specified time interval as a continuous line. Also, as a reference signal 6 FIG. 1 shows the progression over time of the temperature of all other wheelset bearings (R.sub.j with j=2 . . . N) on the same rail vehicle within the same time interval, as a dashed line.

(17) The signal 4 of the component to be investigated is compared with the reference signal 6.

(18) In the comparison, a comparison variable 14 describing the deviation of the signal 4 from the reference signal 6 is determined.

(19) FIG. 2 shows a graph 12 for determining the comparison variable 14, which comparison variable 14 describes the deviation of a signal 4 of the component to be investigated from a reference signal 6. The signal 4 of the component to be investigated and the reference signal 6 are similar to the signals 4 and 6 shown in FIG. 1. With regard to the characteristics of the signal 4 of the component to be investigated and the reference signal 6, reference is hereby made to FIG. 1.

(20) FIG. 2 explains an example of the calculation of the comparison variable 14. FIG. 2 does not show, in particular, the calculation of the comparison variable for the operating parameter values x.sub.Ri(t) shown in FIG. 1 for the signal 4 of the component to be investigated and for the reference values {circumflex over (X)}.sub.Ri(t) of the reference signal 6 shown in FIG. 1.

(21) In FIG. 2 the time t is plotted on the x-axis 16 of the graph 12. On the y-axis 18 of the graph 12, the cumulative operating parameter is plotted. In this example, the cumulative temperature is plotted on the y-axis 18 of the graph 12.

(22) To determine the comparison variable 14 the operating parameter values x.sub.Ri(t) are accumulated over time. The cumulative operating values X.sub.Ri are shown in the diagram as the solid line 20.

(23) In addition, the reference values {circumflex over (X)}.sub.Ri(t) are accumulated over time. The cumulative reference values {circumflex over (X)}.sub.Ri are shown in the diagram 12 as the dashed line 22.

(24) A maximum cumulative deviation d.sub.i, also the maximum cumulative distance d.sub.i, between the accumulated operating values X.sub.Ri and the accumulated reference values {circumflex over (X)}.sub.Ri is determined as the comparison variable. The maximum cumulative deviation d.sub.i is determined using a so-called Kolmogorov-Smirnov test statistic. The maximum cumulative deviation d.sub.i is normalized to the number of measured values N in the respective time interval, wherein here, for example, N=min(N.sub.v,N.sub.{circumflex over (V)}). The comparison variable 14, here the maximum cumulative deviation d.sub.i, is calculated as follows:

(25) $d_i\text{:=}{d}_i\left(x_{R_i},{\hat{x}}_{R_i}\right)=\frac{\sup .Math.X_{R_i}\left(t\right)-{\hat{X}}_{R_i}\left(t\right).Math.}{N}$

(26) The comparison variable 14, here the maximum cumulative deviation d.sub.i, is indicated in FIG. 2 by an arrow 14.

(27) FIG. 3 shows a diagram 24 for determining a probability of occurrence of the comparison variable of FIG. 2. The graph 24 contains a predefined distribution 30 of a plurality of such comparison variables.

(28) On the x-axis 26 of the graph 24 the comparison variable 14, here the maximum cumulative deviation d.sub.i, is plotted. On the y-axis 28 of the graph 24 a frequency is plotted.

(29) The method described under FIGS. 2 and 3 has already been carried out for multiple components as the component to be investigated, in each case for multiple rail vehicles and for multiple time intervals. In doing so a plurality of such comparison variables has been determined. On the basis of these determined comparison variables, the distribution 30 of the plurality of these comparison variables is determined.

(30) The distribution 30 of the plurality of such distribution variables is determined by obtaining the frequencies of these determined distribution variables. In this way, an empirical frequency distribution can be determined. The empirical frequency distribution is shown in the diagram 24 in FIG. 3 as the histogram 32.

(31) Further, in this example to determine the distribution 30 of the plurality of such comparison variables, a distribution function of a given distribution type is defined as a parametrized distribution function. The distribution function can be a distribution density function or a cumulative distribution function. In this example, a distribution density function is specified, for example. For example, the specified distribution type can be a logarithmic normal distribution or an exponential function. The parameters of the parameterized distribution function are fitted (using known methods), so that a modified distribution function is determined. The modified distribution function is shown in the diagram 24 in FIG. 3 as a solid line 34.

(32) The modified distribution function is used to calculate the cumulative distribution function. The cumulative distribution function is shown in the diagram 24 in FIG. 3 as the dashed line 36.

(33) A probability of occurrence of the comparison variable d.sub.i with i=1, in short d.sub.1, is determined for the component R.sub.1 to be investigated using the specified distribution 30 of the plurality of such comparison variables.

(34) The probability of occurrence is then directly a measure of how anomalous the observed temperature profile is on the wheelset bearing R.sub.i to be investigated, here R.sub.1.

(35) For the probability of occurrence a limit value G is specified. In particular, the limit value G is specified before the calculation of the probability of occurrence. The limit value G can be, for example, 1% (0.01), 0.5% (0.005) or 0.1% (0.001).

(36) If the probability of occurrence of the comparison variable d.sub.1 is greater than the specified limit value G, a normal behavior of the component to be investigated is detected.

(37) If the probability of occurrence of the comparison variable d.sub.1 is less than the specified limit value G, an abnormal behavior of the component to be investigated is detected.

(38) If the probability of occurrence of the comparison variable d.sub.1 falls below the limit value G, for example, here e.g. 0.005, an alarm is generated which can be forwarded to a maintenance planning and/or maintenance control system, for example. In addition, an alarm can be generated when the probability of occurrence of the comparison variable d.sub.1 falls below the limit value G several times within a predefined time interval.

(39) Although the invention has been illustrated and described in greater detail by means of the preferred exemplary embodiments, the invention is not restricted by the examples disclosed and other variations can be derived therefrom by the person skilled in the art without departing from the scope of protection of the invention.