Method and Device Used for Providing and Evaulating a Sensor Model for Change Point Detection

Abstract

A method evaluates a data-based sensor model for determining a change-point time in a sensor signal time series. The method includes providing an evaluation signal time series within an evaluation time window of a sensor signal time series, and determining sensor signal extracts from the evaluation signal time series. The sensor signal extracts are (i) time-shifted with respect to one another, or (ii) respectively offset from one another by a number of sensing steps. The sensor signal extracts are shorter in length than the evaluation signal time series. The method further includes determining one or more frequency contributions from the sensor signal extracts using a fast Fourier transform (“FFT”) or a Goertzel algorithm, and evaluating the one or more frequency contributions in a trained data-based sensor model in order to determine a change-point time within the evaluation time window.

Claims

1. A method for evaluating a data-based sensor model for determining a change-point time in a sensor signal time series, the method comprising: providing an evaluation signal time series within an evaluation time window of a sensor signal time series; determining sensor signal extracts from the evaluation signal time series, the sensor signal extracts being (i) time-shifted with respect to one another, or (ii) respectively offset from one another by a number of sensing steps, the sensor signal extracts are shorter in length than the evaluation signal time series; determining one or more frequency contributions from the sensor signal extracts using a fast Fourier transform (“FFT”) or a Goertzel algorithm; and evaluating the one or more frequency contributions in a trained data-based sensor model in order to determine a change-point time within the evaluation time window.

2. The method according to claim 1, wherein the sensor model is trained to respectively associate a corresponding change-point time with the one or more frequency contributions from an evaluation-point time series.

3. The method according to claim 1, wherein the one or more frequency contributions are determined based on one or more predetermined frequencies or a phase state of an underlying sine or cosine signal.

4. The method according to claim 1, wherein the sensor model is configured as a single or multilayer neural network.

5. The method according to claim 1, wherein: the sensor model is configured to indicate the change-point time as a classification vector, and the change-point time is indicated as an argmax of the classification vector.

6. A device for carrying out the method according to claim 1.

7. A computer program product including instructions which, when executing the computer program product by a computer, cause the computer to execute the method according to claim 1.

8. A non-transitory machine-readable storage medium comprising instructions which, when executed by a computer, cause the computer to execute the method according to claim 1.

9. A method for training a data-based sensor model for evaluating an evaluation-point time series in order to determine a change-point time, comprising: providing training datasets which are in each case indicative of an evaluation-point time series and a label including a change-point time; determining sensor signal extracts from an evaluation-signal time series, which extracts are time-shifted with respect to one another or are respectively offset from one another by a number of sensing steps, the sensor signal extracts are shorter in length than the evaluation-signal time series; determining one or more frequency contributions from the sensor signal extracts using a fast Fourier transform (“FFT”) or a Goertzel algorithm; and training the data-based sensor model using the one or more frequency contributions and the change-point times associated therewith.

10. The method according to claim 9, wherein the data-based sensor model is configured as a deep neural network and is trained using a back propagation based training method.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Embodiments are explained in more detail in the following with reference to the accompanying drawings. Here:

[0031] FIG. 1 shows a schematic illustration of a sensor system for detecting sensor signal time series;

[0032] FIG. 2 shows a flow chart illustrating a method used for evaluating a data-based sensor model with previous frequency-based characteristic extraction;

[0033] FIG. 3 shows a representation of an evaluation signal time series with sensor signal extracts;

[0034] FIG. 4 shows a flow chart illustrating a method used for training a data-based sensor model with previous frequency-based characteristic extraction; and

[0035] FIG. 5 shows a schematic representation of an injection system for injecting fuel into the cylinder of an internal combustion engine using a control unit in which the sensor model is implemented.

DETAILED DESCRIPTION

[0036] In the following, the evaluation of a sensor model is described in greater details in reference to a block diagram in FIG. 1 and a flow chart in FIG. 2.

[0037] FIG. 1 shows a sensor system 1 having a sensor 2 configured to record and detect continuous measurement signals. For example, the sensor 2 can correspond to a pressure sensor, a mass flow sensor, a temperature sensor, an accelerometer, a vibration sensor, a radiation sensor, or the like, and it is sampled at a sampling rate in step S1 in order to obtain a continuous sensor signal time series S with respect to discrete sampling steps.

[0038] The sensor signal time series S can correspond to the detection of a varying physical variable that changes according to, e.g., a cyclic process. The cyclic process is detected and includes a cyclic state change that translates into a physical variable change.

[0039] The sensor signal time series S is fed in step S2 to a preprocessing block 3, which cyclically applies an evaluation time window to the sensor signal time series S in order to determine an evaluation signal time series A. The evaluation signal time series features a predetermined number of samples, which are generated from the sensor signal time series S. The preprocessing block 3, depending on a specification for the evaluation time window, creates the evaluation signal time series A as a vector of predetermined length.

[0040] The evaluation signal time series A is timed with respect to the sensor signal time series such that the former includes the repeating state change of the change point time to the extent possible.

[0041] The evaluation signal time series A is fed to a characteristic extraction block 4 in step S3. From the evaluation signal time series extracts A, the characteristic extraction block 4 extracts respective signal time series extracts, which in each case correspond to an extract from the evaluation signal time series A and are shorter in length, e.g., measuring between 30% and 70% of the length of the evaluation signal time series A. The signal time series extracts are offset with respect to one another by, e.g., one or a predetermined number of sample values. FIG. 3 illustrates, by way of example, the potential relationship between the signal time series extracts F1, F2, F3, F4 and the evaluation signal time series A.

[0042] In characteristic extraction block 4, a frequency analysis function is further applied during step S4 to each of the signal time series extracts F1, F2, F3, F4, e.g., in the form of an FFT (Fast Fourier Transformation), a DFT (Discrete Fourier Transformation) or a Goertzel algorithm. The Goertzel algorithm represents a particular form of discrete Fourier transformation by which discrete spectral fractions can be efficiently calculated.

[0043] Using the frequency analysis, a spectral fraction, i.e., a frequency contribution from one or more predetermined frequencies, can be determined for each of the signal time series extracts F1, F2, F3, F4. These predetermined frequencies correspond to predetermined hyperparameters of the data-based sensor model.

[0044] The one or more frequency contributions F for each of the signal time series extracts F1, F2, F3, F4 are then fed to a sensor model 5 in the form of a single or multilayered neural network during step S5. The neuronal functions of the neural network are defined in an inherently known manner as the sum of the initial values for the preceding neuron layer (which are weighted using a weighting factor), or rather the frequency contributions, and a corresponding bias value. This sum can be applied to a non-linear activation function. The results can be output as an output vector for further processing in a future layer of neurons, or as a classification result.

[0045] In step S6, the sensor model 5 can therefore output an output vector O corresponding to a classification output. As described above, the output vector O comprises elements whose index value is at a point in time or time period within the evaluation window and is permanently associated therewith.

[0046] FIG. 4 is a flow chart illustrating training of the data-based sensor model 5. Starting from training datasets provided in step S11, which in each case comprise one evaluation signal time series A and, optionally, one or more further state variables of the technical system, as well as an associated label in the form of a classification vector, said series are initially fed to the characteristic extraction block 4, which is also used for analysis of the evaluation signal time series A described hereinabove.

[0047] As illustrated in FIG. 3, in step S12 the characteristic extraction block 4 divides the evaluation signal time series A into the signal time series extracts F1, F2, F3, F4, for which a frequency contribution, or rather frequency contributions, are determined in the manner described above. In step S13, one or more frequency contributions for predetermined frequencies and phases, which are fed as input variables in the form of an input vector to the sensor model 5 that is configured as a neural network, result from the frequency analysis corresponding to signal time series extracts F1, F2, F3, F4.

[0048] The neural network of the sensor model 5 is then trained during step S14 according to the resulting frequency contributions. In other words, even during training does the evaluation signal time series A provided using a training dataset that is divided into several signal time series extracts F1, F2, F3, F4, which are offset from one another, in each case represent a temporal extract from the evaluation signal time series A. For example, the evaluation signal time series A is associated with a label in the form of a change point time, particularly in the form of a classification vector, the argmax of which indicates a change point time. The classification vector used for training can have an entry 1 at an index position corresponding to the change point time of the label, whereas a value of 0 is provided at the remaining positions.

[0049] The neural network is trained using inherently known gradient-based methods, e.g. back propagation, in order to appropriately adjust the model parameters, i.e., the weightings and bias values of the artificial neurons. The neural network preferably comprises two layers of neurons, wherein the starting layer can be designed to perform only a dimensional reduction based on the dimension of the classification vector in the form of an output vector O.

[0050] FIG. 5 shows, as an example of a sensor system 1, an injection system 10 for an internal combustion engine 12 of a motor vehicle, for which a cylinder 13 (of in particular several cylinders) is shown by way of example. The internal combustion engine 12 is preferably designed as a direct-injection diesel engine, but may also be provided as a gasoline engine.

[0051] The cylinder 13 comprises an intake valve 14 and an exhaust valve 15 for supplying fresh air and removing combustion exhaust.

[0052] Furthermore, fuel for operating the internal combustion engine 12 is injected into a combustion chamber 17 of the cylinder 13 via an injector valve 16. To this end, fuel is provided to the injector valve via a fuel supply 18, via which fuel is provided in an inherently known manner (e.g., a common rail) under high fuel pressure.

[0053] The injector valve 16 comprises an electromagnetically or piezoelectrically controllable actuator unit 21 coupled to a valve needle 22. In the closed state of the injector valve 6, the valve needle 22 is seated on a needle seat 23. By controlling the actuator unit 21, the valve needle 22 is moved longitudinally and exposes a portion of a valve opening in the needle seat 23 in order to inject the pressurized fuel into the combustion chamber 17 of the cylinder 13.

[0054] The injector valve 16 furthermore comprises a piezo sensor 25 arranged within the injector valve 6. The piezo sensor 25 is deformed by pressure changes in the fuel being conducted by the injector valve 6 and is generated by a voltage signal in the form of a sensor signal.

[0055] The injection is performed in a controlled manner by a control unit 30, which specifies a quantity of fuel to be injected by energizing the actuator unit 21. The sensor signal is sampled over time using an A/D converter 31 in the control unit 30, in particular at a sampling rate of 0.5 to 5 MHz. Doing so results in a sensor signal time series.

[0056] Furthermore, a pressure sensor 18 is provided in order to determine a fuel pressure upstream of the injector valve 16.

[0057] During operation of the internal combustion engine 12, the sensor signal is used to determine a correct opening or closing time of the injector valve 16. For this purpose, the sensor signal is, using the A/D converter 31 and via the indication from an evaluation time window, digitized into a corresponding sensor signal time series and evaluated by means of the above-described characteristic extraction and subsequent evaluation using the trained, data-based sensor model 5, whereby an opening duration for the injector valve 16 and, accordingly, an injected quantity of fuel can be determined, depending on the fuel pressure and further operating parameters. An opening time and a closing time are in particular needed in order to determine the opening duration, as the difference in time between these parameters.

[0058] In conjunction with the above sensor system 1, the sampled pressure signal corresponds to the sensor signal time series, wherein the controlling time for opening or closing the injector valve can be assumed as the change point time for the label. The evaluation time window arises as a result of the cyclic repetition of the injection process in an internal combustion engine with a temporal state that essentially begins substantially at a predetermined amount of time before the actuated opening time and can be determined as a crankshaft angle.