Method for detecting a clogged differential pressure line of a pressure-measuring arrangement

Abstract

The invention relates to a computer-implemented method for detecting a clogged differential pressure line of a pressure-measuring arrangement, comprising the following steps: - detecting a measured pressure signal by means of the pressure-measuring arrangement (1), preferably for a defined time period; - combining preferably successive measured values ​​of the measured pressure signal into preferably overlapping blocks (Bt-1, Bt) which preferably each have the same length; - generating in particular acoustic frequency spectra for each block, wherein each frequency spectrum has a plurality of frequency components (c1,t-1; cn,t); - entering the in particular acoustic frequency spectra of the blocks into a neural network which is configured to detect a blocked differential pressure line on the basis of the frequency spectra of the blocks; and - generating an output which indicates whether a blocked differential pressure line is present or not.

Description

[0001] Method for detecting a blocked differential pressure line of a pressure measuring arrangement

[0002] The invention relates to a computer-implemented method for detecting a blocked differential pressure line of a pressure measuring arrangement, a data processing device with means for carrying out the method, a computer program, a computer-readable medium and a pressure measuring arrangement.

[0003] European patent EP 1 840 549 B1 protects a device for detecting blockages in differential pressure lines, the device comprising two pressure sensors for detecting time series of the static pressure in each differential pressure line and a differential pressure sensor for detecting time series of the difference between the pressures in the two differential pressure lines and three calculation units for calculating the fluctuations of the two static pressures and the differential pressure based on the time series and the sums of squares of the fluctuations; correlation coefficient calculation units for determining the correlation coefficients between the time series of a static pressure and the differential pressure and an evaluation unit which detects blockages in the differential pressure lines based on the correlations and identifies which of the differential pressure lines is or are blocked.

[0004] The German patent application DE 10 2013 110 059 A1 describes a differential pressure measuring arrangement which detects blocked differential pressure lines based on a correlation of a differential pressure signal, which represents a pressure difference between a first medium pressure and a second medium pressure, and a temperature signal, which correlates with a temperature of a differential pressure line.

[0005] Furthermore, there are other approaches for detecting a blocked differential pressure line. In particular, there are approaches that are based on changing the frequency of the pressure signal fluctuations.

[0006] All approaches are effective in themselves, but each has specific weaknesses. Some methods fail to work, or only work to a limited extent, if the underlying process is too dynamic and subject to large fluctuations. Other methods only work within a specific pressure signal range, which dynamic processes regularly leave. Still other methods exhibit weaknesses when long differential pressure lines create resonances that affect the signal power and can sometimes artificially increase it.

[0007] The invention is based on the object of remedying this situation. This object is achieved according to the invention by the method for detecting a blocked differential pressure line according to claim 1, the method for training a neural network according to claim 2, the data processing device according to claim 14, the computer program according to claim 15, the computer-readable medium according to claim 16, and the pressure measuring arrangement according to claim 17.

[0008] The computer-implemented method according to the invention for detecting a blocked differential pressure line of a pressure measuring arrangement comprises the following steps:

[0009] Detecting a pressure measurement signal by means of the pressure measurement arrangement, preferably for a defined period of time;

[0010] Combining, preferably consecutive, measured values ​​of the pressure measurement signal into, preferably overlapping, blocks, each of which preferably has the same length;

[0011] Generating, in particular, acoustic frequency spectra for each block, wherein each frequency spectrum has several frequency components;

[0012] Input of the, in particular acoustic, frequency spectra of the blocks into a neural network which is configured, in particular trained, to detect a blocked differential pressure line based on the frequency spectra of the blocks;

[0013] Generate an output indicating whether or not a blocked differential pressure line is present.

[0014] The invention further relates to a computer-implemented method for training a neural network for detecting a blocked differential pressure line of a pressure measuring arrangement, comprising the following steps:

[0015] Detecting a pressure measurement signal by means of the pressure measurement arrangement, preferably for a defined period of time;

[0016] Combining, preferably consecutive, measured values ​​of the pressure measurement signal into, preferably overlapping, blocks, each of which preferably has the same length;

[0017] Generating, in particular, acoustic frequency spectra for each block, wherein each frequency spectrum has several frequency components;

[0018] Input of the, in particular acoustic, frequency spectra of the blocks into a neural network to train the neural network using the generated frequency spectra of the blocks.

[0019] According to the invention, the recording and evaluation of a frequency spectrum of a pressure measurement signal from a pressure measurement system is proposed. The frequency spectrum can preferably be an acoustic frequency spectrum that lies in a range below 20 kHz, in particular below 2.5 kHz, in particular in the range from 0 to 2 kHz, and most particularly in the range from 5 Hz to 200 Hz. The (acoustic) frequency spectrum characterizes the pressure measurement system and essentially represents a fingerprint of the pressure sensor. A change in the pressure measurement system affects the frequency spectrum, in particular the acoustic frequency spectrum. This is utilized according to the invention to train a neural network to detect blocked differential pressure lines.

[0020] An advantageous embodiment of the method according to the invention provides that, before the frequency spectra of the blocks are input into a neural network, the frequency spectra of several, preferably consecutive, blocks are first aggregated into a two-dimensional spectrogram, each of which has a time on the abscissa and an amplitude of the Fourier coefficients on the ordinate. The spectrograms are used as input for the neural network. In particular, the embodiment can provide that a power is calculated for each frequency component for the spectrograms.

[0021] A further advantageous embodiment of the method according to the invention provides that for the training of the neural network, each spectrogram is assigned a binary label which indicates whether a blocked differential pressure line is present or not.

[0022] A further advantageous embodiment of the method according to the invention provides that each spectrogram is processed by the neural network through several, preferably two, convolutional layers and / or reduced in dimension by a max-pooling layer.

[0023] A further advantageous embodiment of the method according to the invention provides that the neural network is a convolutional network configured to utilize convolutional layers, in particular depth-separable convolutions. In particular, the embodiment can provide for flattening to be performed in the convolutional network, in which the multidimensional output data structure of a respective convolutional layer of the convolutional network is converted into a one-dimensional output data structure.

[0024] A further advantageous embodiment of the method according to the invention provides that a multi-stage, preferably a three-stage classifier consisting of fully connected layers of decreasing size generates a scalar output, particularly from the output data structure, which is mapped, preferably by a sigmoid activation function, into a value range [0; 1] in order to generate the output indicating whether or not a blocked differential pressure line is present. A further advantageous embodiment of the method according to the invention provides that the frequency spectra are generated using a discrete Fourier transformation, particularly a fast Fourier transformation.

[0025] A further advantageous embodiment of the method according to the invention provides that the frequency spectra of the respective blocks are high-pass filtered so that frequency components that lie below a predefinable cutoff frequency are essentially filtered out or attenuated. In particular, the embodiment can provide that in order to filter out the idle phases of the process, at least one variance is calculated for each block, and based on the variance, a decision is made as to whether this is a idle phase of the process. In particular, the embodiment can provide that the variance is compared with a configurable or adjustable threshold value in order to decide whether this is a idle phase of the process, wherein the threshold value is preferably in a range < -70 dB, more preferably < -80 dB, most preferably in the range from -80 to -90 dB.

[0026] The invention further relates to a data processing device with means for carrying out the method for detecting a blocked differential pressure line and / or for carrying out the method for training a neural network according to one of the previously described embodiments.

[0027] The invention further relates to a computer program with instructions which, when the program is executed by a computer, cause the computer to carry out the method for detecting a blocked differential pressure line and / or the method for carrying out the method for training a neural network according to one of the previously described embodiments.

[0028] The invention further relates to a computer-readable medium containing instructions which, when executed by a computer, cause the computer to carry out the method for detecting a blocked differential pressure line and / or the method for carrying out the method for training a neural network according to one of the previously described embodiments.

[0029] The invention further relates to a pressure measuring arrangement, at least comprising: a pressure transducer for detecting at least one media pressure of a medium; an operating and / or evaluation circuit for providing a pressure measurement signal which depends on the at least one media pressure; at least one differential pressure line which is connected to a pressure input of the pressure transducer in order to apply the at least one media pressure to the pressure transducer; a detection unit which is preferably designed as part of the operating and / or evaluation circuit and is configured to carry out the method for detecting a blocked differential pressure line and / or the method for carrying out the method for training a neural network according to one of the previously described embodiments.

[0030] The invention is explained in more detail with reference to the following drawings. It shows:

[0031] Fig. 1 : an overall view of an embodiment of a pressure measuring arrangement according to the invention,

[0032] Fig. 2: a sensor module and an electronic module of the pressure measuring arrangement electrically connected to the sensor module in detail,

[0033] Fig. 3: a general block diagram of the computer-implemented method for detecting clogged differential pressure lines of a pressure measuring arrangement,

[0034] Fig. 4: another block diagram of the method for detecting a clogged differential pressure line of a pressure measuring arrangement, showing both the learning and the inference phases,

[0035] Fig. 5: another block diagram of the method for detecting a blocked differential pressure line of a pressure measuring arrangement using a neural network, and

[0036] Fig. 6: a spectrogram for chunks that have been mean-corrected and transformed into frequency domain using a discrete Fourier transform.

[0037] Figure 1 shows a highly simplified and schematic overall representation of a pressure measuring arrangement. Fig. 1 shows a differential pressure measuring arrangement. Although the invention is described below with reference to the differential pressure measuring arrangement, it is not limited to this, but can also be applied to any other pressure measuring arrangement, in particular an absolute and / or relative pressure measuring arrangement. The differential pressure measuring arrangement shown in Fig. 1 comprises a differential pressure measuring transducer 10, which has a sensor module 11, which is arranged between a first, high-pressure side process connection flange 12 and a second, low-pressure side process connection flange 13, and an electronics module 14. Both modules are arranged in a housing of the differential pressure measuring transducer.The electronic module can be connected to a process control system 18 via a two-wire cable 16, whereby the electronic module 14 can communicate and be supplied with power via the two-wire cable. The two-wire cable can be operated, in particular, as a fieldbus according to the Profibus or Foundation Fieldus standard or according to the HART standard. Such differential pressure transmitters are known per se and are manufactured and marketed by the applicant, for example, under the Deltabar brand.

[0038] The differential pressure measuring arrangement can further comprise a differential pressure transducer 20 for installation in a pipeline 21. The differential pressure transducer comprises an orifice plate 22, a first pressure tap channel 23 on a high-pressure side of the orifice plate 22, and a second pressure tap channel 24 on a low-pressure side of the orifice plate 22. The high-pressure-side process connection flange 12 is connected to the high-pressure-side pressure tap channel 23 via a high-pressure-side differential pressure line 25, and the low-pressure-side process connection flange is connected to the low-pressure-side pressure tap channel 24 via a low-pressure-side differential pressure line 26. The terms "high-pressure side" and "low-pressure side" refer to a pressure difference generated by a flow (from left to right in the drawing), which is proportional to the square of the flow velocity and, for example, is in the order of magnitude of a few tens to hundreds of mbar.The static pressure on which this flow-dependent pressure difference is superimposed can, for example, range from 1 bar to several 100 bar.

[0039] Fig. 2 shows a sensor module 11 and an electronics module 14 electrically connected to the sensor module 11 in detail. For reasons of simplicity, the arrangement of the two modules shown in Figure 2 is arranged side by side and thus differs from that in Figure 1. The sensor module in this case is a piezoresistive pressure sensor element 110 with a measuring diaphragm 112, wherein the pressure sensor element 110 has resistance elements in a bridge circuit in order to convert a pressure-dependent deformation or deflection of the measuring diaphragm into an electrical sensor module signal. Instead of the piezoresistive pressure sensor, a capacitive pressure sensor can also be provided, in which case the measuring diaphragm has an electrode whose capacitance relative to an electrode on a rigid counter-body is a measure of the pressure-dependent deformation or deflection of the measuring diaphragm.Details of electrical transducers are familiar to a person skilled in the field of pressure measurement technology and need not be explained in detail here. The sensor module outputs the sensor module signal, which is dependent on the detected pressure difference, to the electronic module 14. An operating and / or evaluation circuit 160 of the electronic module 14 uses the sensor module signal to generate a differential pressure measurement signal representing the pressure difference and can output it, for example, via the two-wire line 16, in particular to the process control system 18. For this purpose, the operating and / or evaluation circuit can have a computing unit, for example in the form of a microprocessor 162, for processing sensor module signals from the pressure sensor 110 digitized by an ADC 164.

[0040] The pressure measuring arrangement further comprises a detection unit 150 configured to execute the computer-implemented method described below. As shown in Fig. 2, the detection unit 150 can be part of the operating and / or evaluation circuit 160 or, alternatively, can be implemented as a separate unit, e.g., in a cloud.

[0041] The computer-implemented method is divided into several method steps: a preprocessor step executed in a preprocessor 152 (see Fig. 3), in which the pressure measurement signal or differential pressure measurement signal is transformed into a frequency spectrum; and a feature extractor step executed in a feature extractor 154 (see Fig. 4), in which features are derived from the frequency spectrum using a neural network. Furthermore, an optional classifier step can be executed in a classifier 156, in which a decision is made, based on the previously derived features, as to whether the differential pressure line 25, 26 is blocked or not. The method steps are executed sequentially. Fig. 3 shows a general block diagram of the computer-implemented method for detecting blocked differential pressure lines 25, 26 of a pressure measuring arrangement 1.

[0042] In the first step, the pressure measurement signal is first sampled at the highest possible sampling rate in order to obtain successive measured values, preferably digitized measured values, particularly preferably digitized measured values ​​by the ADC 164. The sampling rate is preferably selected such that it is at least a few tens of Hz, preferably at least 50 Hz, particularly preferably at least 100 Hz.

[0043] To prevent idle phases associated with the process, whose pressure values ​​are determined by means of the pressure measuring arrangement 1, from making it difficult to detect a clogged differential pressure line 25, 26, it is necessary to detect and filter out these idle phases as quickly as possible. This can generally be achieved in different ways. A simple subsequent method can provide for the preferably consecutive measured values ​​in the preprocessor step to be divided into preferably overlapping blocks B t-i, Bt (hereinafter also referred to as chunks), preferably of the same length. Chunks Bn, B t form inseparable units and all measured values ​​of a chunk Bn, B t are processed together. Subsequently, in the preprocessor step, for each chunk B t -i, B t A variance and optionally a mean are calculated for each. The variance is then compared with a configurable or adjustable threshold. If the variance of the chunk BM or B t below the threshold, the chunk BM or B tignored because it is most likely a rest phase in the process. The threshold value depends on the process and can vary from process to process. For example, the threshold value for a level process, in which a level is determined using a pressure measuring arrangement, may be different than for a flow process, in which a flow is determined using a differential pressure measuring arrangement. Typically, however, the threshold value is in a range < -70 dB, preferably < -80 dB, and particularly preferably in the range of -80 to -90 dB.

[0044] The remaining chunks B t -i, B tare then mean-corrected in the preprocessor step and transformed into the frequency domain using a discrete Fourier transformation, e.g. a Fast Fourier Transformation (FFT), in order to obtain the Fourier coefficients required for the subsequent feature extractor step for each chunk BM , B t Fig. 6 shows an example of such a spectrogram. After the Fourier transformation, the preprocessor step can optionally provide for the application of a high-pass filter 158, which operates at a predeterminable cutoff frequency in order to filter out the actual pressure value. The predeterminable cutoff frequency can, for example, be selected such that it lies in the range of a few Hz, preferably less than 5 Hz, particularly preferably in the range of 1-5 Hz. The output of the preprocessor after processing consists of the (possibly high-pass filtered) Fourier coefficients of a chunk.

[0045] In the feature extraction step following the preprocessor step, the features required to detect a clogged differential pressure line 25, 26 are determined. For this purpose, a variant is proposed here that is essentially based on the application of neural networks, in particular convolutional neural networks (hereinafter also referred to as convolutional neural networks or "CNNs" for short), from the field of machine learning, and which is essentially illustrated in Figs. 3 and 4.

[0046] This variant is divided into a learning or training phase, in which characteristics typical for the process are first recognized or learned, and an inference phase, in which the spectra are analyzed based on the learned patterns during the actual measurement operation of the pressure measuring arrangement.

[0047] The training is based on the (high-pass filtered) Fourier coefficients of several chunks, each composed of K frequency components (spectrograms) and provided by the preprocessor. However, to train the network, especially the convolutional network, several chunks must be aggregated into a single sample. Furthermore, it can be advantageous to calculate a power spectrum. The FFT aggregator is configured to group N consecutive preprocessor outputs into two-dimensional spectrograms. Furthermore, a power value for each frequency component can optionally be calculated for each output. After this step, the resulting two-dimensional spectrograms can be used to train the convolutional network.For training in the learning phase, it may initially be provided that the two-dimensional spectrograms are sent to an external computer via an unspecified data transmission, since the memory space and computing power of the microprocessor 162 may not be sufficient for training CNNs.

[0048] The two-dimensional spectrograms serve as inputs (samples) for the convolutional network. Each spectrogram is then assigned a binary label, where 0 represents a free impulse line (negative class) and 1 represents a clogged line (positive class). The degree of clogging at which a sample belongs to the positive class is determined in advance. The spectrogram in the transition region between the two classes is assigned to the one from which the most Fourier coefficients originate. In the event of a tie, the sample is preferably assigned to the negative class. For subsequent training, the spectrograms (samples) can be shuffled and randomly divided into a training, a validation, and a test data set. The training data set can preferably contain 60% of the samples, and the validation data set can contain another 30% of the samples, so that the remaining 10% can form the test data set.The test dataset is required as an independent dataset, as the validation data is used for adjusting hyperparameters, such as batch size, and for detecting under- and overfitting. It is preferable to scale the spectrograms in decibels when generating the dataset. The logarithmic scale provides better visual separation of small performance differences, which simplifies network training.

[0049] Fig. 5 shows an embodiment of a neural network that can be used for both the training and inference phases. The network can preferably be a convolutional network. As input, the network can receive, for example, a section of the spectrogram with dimensions T x K x 1, where T describes the number of consecutive time steps (represented by different chunks) and K the number of Fourier coefficients remaining after high-pass filtering. The spectrogram section can then be processed by two convolutional layers and reduced in dimension by max-pooling layers. After flattening, in which the multidimensional data structure is converted into a one-dimensional data structure, a three-stage classifier consisting of fully connected layers of decreasing size can be implemented.This ultimately results in a scalar output, which is mapped into a value range (0; 1) using a sigmoid activation function. A predefined threshold value q from this interval can then be used to set a limit for assigning a sample to the positive or negative class, so that a binary output can be generated that indicates whether a clogged differential pressure line is present or not.

[0050] For convolution, classic convolutions (convolutional layers) are preferably not used. Instead, convolutional layers, in particular so-called depthwise separable convolutions, are used. Depthwise separable convolutions are described, for example, in the publication "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications" by Andrew G. Howard, et al. or in the publication "Rigid Motion Scattering for Texture Classification" by Laurent Slfre, Stephane Mallat. For this purpose, each of the N input channels is first transformed by a separate convolution with a kernel size W x H x 1, where W and H denote the width and height of the kernels. The resulting results can then be combined to form a tensor of dimension T' x K' x N' before M convolutions with a kernel size 1 x 1 x N can be applied to the tensor.The final result is a tensor of dimension T' x K' x M with M feature maps of size T' x K', as would have been the case using a classical (W x H x M) convolution.

[0051] For example, both convolutional layers can have kernels of size 3 x 3, with padding and stride each set to one. Eight feature maps can be generated in the first convolutional layer, which can be expanded to twelve by the second convolution. After each layer, a (2 x 2) max pooling can halve the width and height of the feature maps. ReLu is preferably used as the activation function in all convolution and fully connected layers. This is the exception for the last layer, where, as described above, the sigmoid function is used.

[0052] For the inference phase, in which the spectra are analyzed using the learned patterns during the actual operation of the pressure measurement system, the trained model with model parameters 0, especially the weights of the neural network, is transferred back to the pressure measurement system's processing unit if the training was not performed on the pressure measurement system's processing unit. The model can then use the model parameters to directly classify the FFT aggregator sample at time t and generate an output based on the threshold value q indicating whether or not a clogged differential pressure line is present.

[0053] Basically, the output of the Feature Extractor can indicate whether a clogged

[0054] The detection of a clogged differential pressure line 25, 26 may be sufficient for detecting a clogged differential pressure line 25, 26. However, this may produce incorrect results. For this reason, it can optionally be provided that, in the classifier step following the feature extractor step, additional smoothing of the feature extractor output is performed using a mean filter. The mean filter aggregates the last M outputs resulting from the feature extractor step and, depending on the threshold, decides based on the mean value whether a clogged differential pressure line is actually present or not.

[0055] List of reference symbols

[0056] 1 pressure measuring arrangement

[0057] 10 pressure transducers

[0058] 11 Sensor module

[0059] 12, 13 Process connection flanges

[0060] 14 Electronic module

[0061] 16 two-wire cable

[0062] 18 Process control system

[0063] 20 differential pressure sensors

[0064] 21 Pipeline

[0065] 22 aperture

[0066] 23, 24 pressure tap channels

[0067] 25, 26 Differential pressure lines

[0068] 110 Pressure sensor element

[0069] 112 measuring membrane

[0070] 140 Pressure transmission medium

[0071] 150 detection units

[0072] 152 Preprocessor

[0073] 154 Feature Extractor

[0074] 156 Classifier

[0075] 158 high-pass filters

[0076] 160 Operating and / or evaluation circuit

[0077] 162 microprocessor

[0078] 164 ADC

[0079] Bt-i, Bt blocks or chunks

[0080] C1 ,t-1 ; C n ,t frequency components of the frequency spectra of the blocks

[0081] CNN Convolutional Neural Network

[0082] 0 Model parameter

Claims

Patent claims 1 . A computer-implemented method for detecting a clogged differential pressure line of a pressure measuring arrangement comprising the following steps: Detecting a pressure measurement signal by means of the pressure measurement arrangement (1), preferably for a defined period of time; Combining, preferably consecutive, measured values ​​of the pressure measurement signal into, preferably overlapping, blocks (B t -i, Bt), each of which preferably has the same length; Generating, in particular, acoustic frequency spectra for each block, where each frequency spectrum contains several frequency components (ci, t -i; c n ,t); Input of the, in particular acoustic, frequency spectra of the blocks into a neural network which is configured, in particular trained, to detect a blocked differential pressure line based on the frequency spectra of the blocks; Generate an output indicating whether or not a blocked differential pressure line is present.

2. A computer-implemented method for training a neural network for detecting a clogged differential pressure line of a pressure measuring arrangement, comprising the following steps: Detecting a pressure measurement signal by means of the pressure measurement arrangement (1), preferably for a defined period of time; Combining, preferably consecutive, measured values ​​of the pressure measurement signal into, preferably overlapping, blocks (B t -i, Bt), each of which preferably has the same length; Generating, in particular, acoustic frequency spectra for each block, where each frequency spectrum contains several frequency components (ci, t -i; c n ,t); Input of the, in particular acoustic, frequency spectra of the blocks into a neural network to train the neural network using the generated frequency spectra of the blocks.

3. Method according to claim 1 or 2, wherein, before the frequency spectra of the blocks are input into a neural network, the frequency spectra of several, preferably consecutive blocks are first aggregated to form a two-dimensional spectrogram, which has a time on the abscissa and an amplitude of the Fourier coefficients on the ordinate, and the spectrograms are used as input for the neural network.

4. Method according to the preceding claim, wherein a power is calculated for each frequency component of the spectrograms.

5. Method according to one or more of the preceding claims, wherein for training the neural network, each spectrogram is assigned a binary label which indicates whether a blocked differential pressure line is present or not.

6. Method according to one or more of the preceding claims, wherein each spectrogram is processed by the neural network through several, preferably two, convolutional layers and / or reduced in dimension by a max-pooling layer.

7. The method according to one or more of the preceding claims, wherein the neural network is a convolutional network which is configured to use convolutional layers, in particular depth-wise separable convolutions.

8. Method according to one of the preceding claims, wherein flattening is carried out in the convolutional network, in which a multidimensional output data structure of a respective convolutional layer of the convolutional network is converted into a one-dimensional output data structure.

9. Method according to one or more of the preceding claims, wherein a multi-stage, preferably a three-stage classifier made up of fully connected layers of decreasing size generates a scalar output, in particular from the output data structure, which is mapped, preferably by a sigmoid activation function, into a value range [0; 1] in order to generate the output indicating whether a clogged differential pressure line is present or not.

10. Method according to one or more of the preceding claims, wherein the frequency spectra are generated by means of a discrete Fourier transformation, in particular a fast Fourier transformation.

11. Method according to one or more of the preceding claims, wherein the frequency spectra of the respective blocks (B t -i, Bt) are high-pass filtered so that frequency components below a predefined cutoff frequency are essentially filtered out or attenuated.

12. Method according to the preceding claim, wherein for filtering out the idle phases of the process for the blocks (B t -i , Bt) at least one variance is calculated, and based on the variance it is decided whether it is a rest phase of the process.

13. Method according to the preceding claim, wherein the variance is compared with a configurable or adjustable threshold value to decide whether it is a rest phase of the process, wherein the threshold value is preferably in a range < -70 dB, particularly preferably < -80 dB, most preferably in the range from -80 to -90 dB.

14. Data processing device with means for carrying out the method for detecting a blocked differential pressure line according to claim 1 and / or for carrying out the method for training a neural network according to claim 2.

15. A computer program comprising instructions which, when executed by a computer, cause the computer to perform the method for detecting a blocked differential pressure line according to claim 1 and / or the method for performing the method for training a neural network according to claim 2.

16. A computer-readable medium containing instructions which, when executed by a computer, cause the computer to perform the method for detecting a blocked differential pressure line according to claim 1 and / or the method for performing the method for training a neural network according to claim 2.

17. A pressure measuring arrangement, comprising at least: a pressure transducer (10) for detecting at least one media pressure of a medium; an operating and / or evaluation circuit (14) for providing a pressure measurement signal which depends on the at least one media pressure; at least one differential pressure line (25) which is connected to a pressure input (12) of the pressure transducer (10) in order to apply the at least one media pressure to the pressure transducer (10); a detection unit (150), which is preferably designed as part of the operating and / or evaluation circuit and is configured to carry out the method for detecting a blocked differential pressure line according to claim 1 and / or the method for carrying out the method for training a neural network according to claim 2.

Images