METHODS AND SYSTEMS FOR WORKPIECE QUALITY CONTROL

Abstract

A computer-implemented method for providing a trained function for performing a workpiece quality control includes receiving a plurality of training machining datasets, wherein different training high-frequency machining datasets are representative for the quality of different workpieces, transforming the plurality of training machining datasets into the time-frequency domain to generate a plurality of training time-frequency domain datasets, and training a function based on the plurality of training time-frequency domain datasets, wherein the function is based on an autoencoder. The autoencoder has input layers, output layers and a hidden layer. The plurality of training time-frequency domain datasets are provided to the input layers and the output layers during training, and a trained autoencoder function is outputted.

Claims

1.-15. (canceled)

16. A computer-implemented method for providing a trained function for performing a workpiece quality control, said computer-implemented method comprising: receiving a plurality of training machining datasets, wherein different training machining datasets are representative of different workpieces having an acceptable quality based on prior tests; transforming the plurality of training machining datasets into a time-frequency domain to produce a plurality of training time-frequency domain datasets; training, based on the plurality of the training time-frequency domain datasets, a function based on an autoencoder, which comprises an input layer, an output layer and a hidden layer, by providing the input layer and the output layer with each of the plurality of training time-frequency domain datasets; and providing the trained function as a trained autoencoder function.

17. The computer-implemented method of claim 16, wherein the autoencoder is a variational autoencoder and the training comprises specifying a statistical distribution for the hidden layer.

18. The computer-implemented method of claim 17, wherein the statistical distribution is a normal distribution, in particular a normal distribution with mean value equal to zero and variance equal to one, or uniform distribution.

19. The computer-implemented method of claim 16, wherein the training machining datasets are based on time-series CNC machining data.

20. The computer-implemented method of claim 16, wherein the transforming is performed by a wavelet transformation.

21. The computer-implemented method of claim 16, wherein the autoencoder is a recurrent neural network, in particular a long short-term memory network.

22. A computer-implemented method for performing a workpiece quality control, said computer-implemented method comprising: receiving a machining dataset based on time-series CNC machining data, wherein the machining dataset is representative for a quality of a workpiece; transforming the machining dataset into a time-frequency domain to produce a time-frequency domain dataset; applying the trained autoencoder function of claim 16 to the time-frequency domain dataset and generate data in the hidden layer; analyzing the quality of the workpiece based on the data generated in the hidden layer of the trained autoencoder function; and providing results of the analysis.

23. The computer-implemented method of claim 22, wherein the autoencoder is a variational autoencoder with a specified statistical distribution for the hidden layer, and wherein the analyzing the quality of the at least one workpiece comprises comparing the data generated in the hidden layer of the variational autoencoder function with the specified statistical distribution.

24. The computer-implemented method of claim 23, wherein the statistical distribution is a normal distribution, in particular a normal distribution with mean value equal to zero and variance equal to one, or uniform distribution.

25. The computer-implemented method of claim 23, wherein the transforming is performed by means of the wavelet transformation.

26. The computer-implemented method of claim 23, wherein the autoencoder is a recurrent neural network, in particular a long short-term memory network.

27. A computer program for providing a trained function, the computer program embodied on a non-transitory computer-readable medium and comprising instructions which, when the program is loaded into a memory of a computing platform and executed by a processor of the computing platform, causes the computing platform to carry out the method of claim 16.

28. A computer program for performing a workpiece quality control, the computer program embodied on a non-transitory computer-readable medium and comprising instructions which, when the program is loaded into a memory of a computing platform and executed by a processor of the computing platform, causes the computing platform to carry out the method of claim 22.

29. A system for providing a trained function for performing a workpiece quality control, said system comprising: a first training interface configured to receive a plurality of training machining datasets, wherein different training machining datasets are representative of different workpieces having an acceptable quality based on prior tests: a training computation apparatus configured to transform the plurality of the training machining datasets into a time-frequency domain to generate a plurality of training time-frequency domain datasets, train, based on the plurality of training time-frequency domain datasets, a function based on an autoencoder, which comprises an input layer, an output layer and a hidden layer, provide the training time-frequency domain datasets for the input layer and for the output layer; and a second training interface configured to provide the trained function.

30. A system for performing a workpiece quality control, said system comprising: a first interface receiving a machining dataset based on time-series CNC machining data, wherein the machining dataset is representative for a quality of a workpiece, a computation apparatus configured to transform the machining dataset into a time-frequency domain to generate a time-frequency domain dataset, apply a trained autoencoder function provided by the method of claim 16 to the time-frequency domain dataset and generate data in the hidden layer, analyze the quality of the workpiece based on the data generated in the hidden layer of the trained autoencoder function, and a second interface providing results of the analysis of the quality of the workpiece.

Description

[0057] The above and other objects and advantages of the invention will be apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings, in which like reference characters refer to like parts throughout, and in which:

[0058] FIG. 1 is a flow diagram of an example of a computer-implemented method for providing a trained function for performing a workpiece quality control,

[0059] FIG. 2 is an example of the structure of an autoencoder,

[0060] FIG. 3 is a flow diagram of an example of a computer-implemented method for performing a workpiece quality control,

[0061] FIG. 4 is an example of results of the analysis performed according to the method of FIG. 3,

[0062] FIG. 5 an example of a training system, and

[0063] FIG. 6 an example of an analyzing system.

[0064] Turning to FIG. 1, a flow diagram of an example of a computer-implemented method for providing a trained function for performing a workpiece quality control is illustrated.

[0065] In the first step a training high-frequency machining dataset is received 100. The training high-frequency machining data is representative of machining data, e.g. machining signals, from good quality workpieces or reference workpieces.

[0066] Receiving 100 may be performed by means of a first interface, which for example can be designed as an edge computing device. In an embodiment of the invention the training high-frequency machining dataset can be based on time-series CNC (Computerized Numerical Control) machining data. In an embodiment the machining data can comprise a plurality of time-series machining signals 101, where on the x-axis the time in seconds is depicted and the y-axis represents spindle torque. The time series machining signal 101 comprises measured values of a measured quantity (bold black line), e.g. of the spindle torque, and an acceptable range of the measured value (grey area surrounding the bold black line). To obtain the machining signal 101 the data is sampled from corresponding machine sensors every 2 milliseconds. It will be appreciated that other sampling intervals are possible, e.g. 5, 10, 15 ms etc., as well as other parameters can be measured, e.g. spindle motor current, control deviation signal etc.

[0067] The training dataset is then transformed 200 into a dataset on a time-frequency domain to achieve a training time-frequency domain dataset. In an embodiment the 1D high-frequency time-series machining signals from the training dataset can be transformed by means of a wavelet transformation. An example of a transformed signal 201 in the time-frequency domain (2D signal) is illustrated in FIG. 1. The x-label is time in seconds. The y-label is frequency in Hertz (Hz). It will be appreciated that the time axis can be transferred for example to path axis, where path is the distance which tool travels over the workpiece to perform the operation.

[0068] The reference signals on the time-frequency domain 201 are used for training 300 an autoencoder function 310. The (2D) training signals are given for an input 311 and an output layer 312 of the autoencoder 310 (see FIG. 2) in order to train it.

[0069] Transforming 200 of the high-frequency time-series data (1D signals) and training 300 of the encoder can be performed by means of a computation apparatus, a computation unit, a computing platform. In an embodiment the (untrained or trained) autoencoder can be stored on the computing platform or, e.g. in a memory of the computation apparatus.

[0070] After training the autoencoder, the trained autoencoder is provided 400. Providing 400 the trained autoencoder can be performed by means of a second interface. Providing the trained autoencoder can comprise for example notifying an operator of the machine, e.g. of the CNC machine, that the autoencoder has been trained and is ready for use. The notifying can be performed in various ways. For example, the operator can receive a notification via email or as a push message on his\her mobile device or as an alarm on a CNC control panel.

[0071] Turning to FIG. 2, an example of the structure of the autoencoder 310 is illustrated. The autoencoder 310 comprises an input layer 311, an output layer 312, and a hidden layer 314. The variables x.sub.1, . . . x.sub.5 can be fed into neurons of the input layer 311 and the variables x.sup.1.sub.1, . . . x.sup.1.sub.5 are fed interneurons of the output layer 312. A reduced representation of the input data can be generated in one of the (intermediate) layers 313, which comprise the hidden layer 314 and optionally one or more convolutional layers. The hidden layer 314 can have the minimal number of neurons.

[0072] The variables x.sub.1, . . . , x.sub.5 can represent single signals. As explained above, in the prior art solutions threshold for acceptable range had to be found for each single signal x.sub.1, . . . , x.sub.5. This needs an exact preprocessing of the signals (example given filtering, etc.) which requires domain expert knowledge.

[0073] Using the autoencoder 310 reduces the complexity of the problem, since the thresholds are to be found only for the (data at the neurons of the) hidden layer 314 with the reduced representation of the signals.

[0074] Good results can be achieved if the autoencoder is a recurrent neural network (RNN), more particularly a long short-term memory (LSTM) network.

[0075] In an embodiment the autoencoder 310 can be a variational autoencoder (VAE). In this case a statistical distribution for the hidden layer 314 is specified during the training 300 of the encoder.

[0076] In an embodiment, the autoencoder 310 can comprise an encoder, wherein the encoder can comprise one or more layers, and a decoder, wherein the decoder can comprise one or more layers. The hidden layer can comprise one or more layers as well. In an embodiment the layer(s) of the hidden layer can comprise two neurons. The input layer 311 and the output layer 312 can comprise 32 neurons.

[0077] In an embodiment the statistical distribution can be a normal distribution. Good results can be achieved, if the normal distribution's mean value is equal to zero and its variance is equal to one—N(0,1). The statistical distribution can be for example a uniform distribution.

[0078] In one embodiment a (training) computer program is provided. The computer program comprises instructions which, when the program is executed by a computing platform, cause the computing platform to carry out the method for providing the trained autoencoder as explained above. The computing platform can comprise an edge computing device, cloud computing device, smartphone, PC etc. The computer program can be designed as an app running on an edge device or on a cloud server.

[0079] Turning to FIG. 3, a flow diagram of an example of a computer-implemented method for performing a workpiece quality control is illustrated.

[0080] In the first step a high-frequency machining dataset from at least one workpiece is received 120. For example, the high-frequency machining dataset can be collected for a workpiece, which is currently under processing. In particular, the high-frequency machining dataset can be based on time-series CNC machining data collected, while machining the workpiece.

[0081] In the second step the high-frequency machining dataset is transformed 220 into a dataset on a time-frequency domain to achieve a time-frequency domain dataset. This transformation 220 can be performed, when the processing of the workpiece is finished, the corresponding data is collected, and the high-frequency machining dataset is received 120. In an embodiment, the transformation 220 can be performed by a wavelet transformation. Using the wavelet transformation may reduce the need of filtering and preprocessing of the data.

[0082] In the third step a trained autoencoder function is applied 320 to the time-frequency domain dataset. The trained autoencoder function can be an autoencoder function 310 provided by the training method described regarding FIG. 1 and FIG. 2.

[0083] The trained autoencoder function comprises a hidden layer. When the trained autoencoder function is applied to the time-frequency domain data set, data in the hidden layer 314 of the autoencoder function is generated.

[0084] Further on, analysis of a quality of the at least one workpiece based on the data generated in the hidden layer 314 of the autoencoder function 310 is performed 420.

[0085] For example, threshold values for the representation of the time frequency domain signal in the hidden layer can be determined based on the domain expert knowledge.

[0086] In an embodiment the autoencoder can be a variational autoencoder VAE with a specified statistical distribution for the hidden layer. In this case the analysis of the quality of the at least one workpiece can be carried out by comparing the data generated in the hidden layer 314 of the autoencoder function 310 (factual statistical distribution) with the specified statistical distribution. All the good quality workpieces will generate a signal in the reduced dimension (in the hidden layer 314), which follows the specified statistical distribution, whereas the bad quality workpieces will have deviation compared to the reference (specified) distribution.

[0087] In an embodiment the specified statistical distribution can be a normal distribution, in particular a normal distribution with mean value equal to zero and variance equal to one, or the uniform distribution.

[0088] Surprisingly good results can be achieved, if the autoencoder 310 is a recurrent neural network (RNN), in particular a long short-term memory (LSTM) network.

[0089] In the next step the results of the analysis are provided 520. In an embodiment the providing 520 can comprise alarming an operator of the machine, for example of the CNC machine, if the analyzed workpiece cannot be verified to be a workpiece of a good quality.

[0090] In this case the operator may receive an alarm (on a CNC control panel), an email (on his/her mobile device, e.g. smartphone), a push message (on his/her mobile device, e.g. smartphone) etc. that notifies him/her that the analyzed workpiece seems to be of bad quality and needs to be further investigated.

[0091] In an embodiment the method can comprise a further step of repeating the steps 120 to 520 for other workpieces.

[0092] The result of the analysis can be an assessment that the work piece is good or otherwise needed to be further investigated.

[0093] In one embodiment a computer program is provided. The computer program comprises instructions which, when the program is executed by a computing platform, cause the computing platform to carry out the method for performing the workpiece quality control as explained above. The computing platform can comprise an edge computing device, cloud computing device, smartphone, PC etc. The computer program can be designed as an app running on an edge device or on a cloud server.

[0094] FIG. 4 illustrates results of such analysis. The desired distribution is set here to bivariate Gaussian distribution with zero mean and unit variance. Data from 20 workpieces was used to train the variational autoencoder function. For monitoring/performing a quality control a set of data (containing data from good and bad quality workpieces) is used. The result shows that the VAE can distinguish between the good and bad quality workpieces with a good performance/precision. This shows that the method performs well with only few (30, 20 or less) training samples.

[0095] Turning to FIG. 5, an example of a (training) system 1 for providing a trained function for performing a workpiece quality control is illustrated. In an embodiment the system 1 can comprise an industrial machine 10, e.g. for milling, boring etc. The industrial machine 10 may carry a frequency converter apparatus 11. In an embodiment the frequency converter apparatus 11 may comprise two or more separate units (FIG. 5 and FIG. 6 show for example two separate units). In an embodiment the frequency converter apparatus 11 can be arranged on the industrial machine 10 or located apart from the industrial machine 10. The system may also comprise a computing device 12, e.g. an edge computing device, and a training computing system 13, which can be based on a computing platform, more particularly on a cloud computing platform. The industrial machine 10, the frequency converter apparatus 11, the computing device 12 and the training computing system 13 are configured to be connected to each other via data links. An exemplary data flow is shown with arrows.

[0096] High-frequency time-series training data 14 is send from the frequency converter apparatus 11 to the computing device 12. The transformation of the high-frequency time-series training data 14 into to time-frequency domain can be performed for example by the computing device 12 or by the training computing system 13. The above-mentioned training computer program can be stored on the training computing system 13. The training computing system 13 can be located in the same industrial plant as the industrial machine 10 or be partially (it can be connected to a cloud server) or entirely in a cloud.

[0097] After the autoencoder 310 is trained by the training computing system 13 a trained autoencoder 15 can be deployed on the computing device 12.

[0098] Turning to FIG. 6, an example of a system 2 for performing a workpiece quality control is illustrated.

[0099] The system 2 can comprise the industrial machine 10, the frequency converter apparatus 11, the computing device 12 of FIG. 1 and an analyzing computing system 23, which are configured to be connected to each other via data links. An exemplary data flow is shown with arrows. The analyzing computing system 23 can be based on a computing platform, more particularly on a cloud computing platform.

[0100] High-frequency time-series data 24 from a workpiece, which is being currently analyzed, is send from the frequency converter apparatus 11 to the computing device 12. The transformation of the high-frequency time-series data 24 into to time-frequency domain can be performed for example by the computing device 12 or by the analyzing computing system 23.

[0101] The above-mentioned computer program for carrying out the method for performing a workpiece quality control can be stored on the analyzing computing system 23 or on the computing device 12 and be remotely accessible by the analyzing computing system 23. The analyzing computing system 23 can be located in the same industrial plant as the industrial machine 10 or be partially (it can be connected to a cloud server) or entirely in a cloud. The analyzing computing system 23 can comprise a man-machine interface or user interface for operating the computer program.

[0102] The trained autoencoder 15 can be stored on the computing device 12.

[0103] The results of analyzing workpiece(s) can be displayed by means of the analyzing computing system 23. The analyzing computing system 23 can comprise a mobile device for displaying the results of the performed quality analysis. The operator may be notified of the state of the analyzed workpiece via an alarm, an email, a push message etc. The notification can be designed in a traffic light manner (see FIG. 6). Each color can correspond to a certain quality of the workpiece.

[0104] As mentioned above, the type of parameters which are measured in order to provide training data or data for the analysis of the quality of workpiece(s) can be chosen during the training process or before or during the analysis.

[0105] For example, the choice of the parameters can be performed by an operator in the app on his smartphone. The app can comprise both a training mode (for training and Autoencoder function) and an analyzing mode (for analyzing the workpiece under machining).

[0106] The above described embodiments of the present disclosure are presented for purposes of illustration and not of limitation.

[0107] In particular, the embodiments described with regard to figures are only few examples of the embodiments described in the introductory part.

[0108] The solution according to the invention is described above with respect to the claimed providing systems as well as with respect to the claimed methods. Features, advantages or alternative embodiments herein can be assigned to the other claimed objects and vice versa. In other words, claims for the providing systems can be improved with features described or claimed in the context of the methods. In this case, the functional features of the method are embodied by objective units of the providing system.

[0109] Furthermore, in the following the solution according to the invention is described with respect to methods and systems for training an autoencoder function as well as with respect to methods and systems for performing a workpiece quality control. Features, advantages or alternative embodiments herein can be assigned to the other claimed objects and vice versa. In other words, claims for methods and systems for training an autoencoder function can be improved with features described or claimed in context of the methods and systems for performing the quality analysis, and vice versa.

[0110] In particular, the trained autoencoder function (algorithm) of the methods and systems for checking the quality of the workpiece can be adapted by the methods and systems for training. Furthermore, the input data can comprise advantageous features and embodiments of the training input data, and vice versa. Furthermore, the output data can comprise advantageous features and embodiments of the output training data, and vice versa.

[0111] The reference signs in the claims used only for clarity purposes and shall not be considered to be a limiting part of the claims.