Bonded state learning device, and bonded state estimation device

Abstract

To provide a device and the like capable enhancing estimation accuracy of bonded states of a plurality of workpieces.SOLUTION: On the basis of a plurality of pieces of learning data, a learning model is trained to learn a correlation between input data for the learning and bonded states of a plurality of workpieces. The input data for the learning includes partial spectrogram data, which is a part of a spectrogram of a vibration signal generated upon one time of bonding processing of a plurality of workpieces W1 and W2 using an ultrasonic bonding device, and which represents the partial spectrogram a range of which is identified by a time domain. Partial spectrogram data of when the workpieces W1 and W2 with unknown bonded states are bonded is input to the trained learning model as input data for estimation. This obtains an estimation result of the bonded states of the plurality of workpieces as an output from the trained learning model.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a technique for estimating the bonding state of a plurality of workpieces made of various materials such as metals using ultrasonic vibrations. [Background technology]

[0002] The present applicant has proposed a technique for joining workpieces using ultrasonic complex vibration (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7219495 Summary of the Invention [Problem to be solved by the invention]

[0004] It is important to estimate the bonding state of multiple workpieces.

[0005] Therefore, an object of the present invention is to provide an apparatus etc. that can improve the accuracy of estimating the bonding state of multiple workpieces. [Means for solving the problem]

[0006] The bonding state learning device of the present invention comprises: a learning data acquisition unit that acquires a learning data group consisting of a plurality of sets of learning data including, as learning input data, partial spectrogram data representing a partial spectrogram whose range is specified by a time domain, which is a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device; a model generation unit that generates a trained model by having a training model learn a correlation between the training input data and the bonding states of the plurality of workpieces based on the training data group acquired by the training data acquisition unit; and and a model evaluation unit that evaluates each of the plurality of trained models generated by the model generation unit based on each of a plurality of training data groups that include data representing each of the plurality of partial spectrograms having different ranges.

[0007] The bonding condition estimating device of the present invention comprises: an estimation input data acquisition unit that acquires, as estimation input data, partial spectrogram data representing a partial spectrogram whose range is specified by a time domain, which is a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device; and a bonding state estimation unit that inputs the input data acquired by the estimation input data acquisition unit into the trained model selected based on the evaluation results by the model evaluation unit of the bonding state learning device, thereby acquiring an estimation result of the bonding state of the plurality of workpieces as an output from the trained model. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating the configuration of an ultrasonic bonding device. [Figure 2] 1 is a diagram illustrating the configuration of a joint state estimating device according to an embodiment of the present invention; [Figure 3] 3 is a flowchart showing the function of the ultrasonic bonding device. [Figure 4] 1 is a flowchart showing the procedure of a bonding state learning method according to an embodiment of the present invention. [Figure 5] 1 is a flowchart showing the procedure of a bonding state estimation method according to an embodiment of the present invention. [Figure 6] FIG. 10 is an example of joint acoustic data. [Figure 7] FIG. 1 is an illustrative view of a spectrogram. DETAILED DESCRIPTION OF THE INVENTION

[0009] (Configuration of ultrasonic complex vibration device) The subject of estimation by the bonding condition estimation device in one embodiment of the present invention is, for example, the bonding condition of workpieces W1 and W2 bonded using ultrasonic complex vibration by an ultrasonic bonding device configured as shown in FIG. 1. The ultrasonic complex vibration device 10 shown in FIG. 1 is a component of the ultrasonic bonding device that bonds workpieces W1 and W2, which are objects to be bonded such as metal plates, using ultrasonic complex vibration, as described below. The ultrasonic bonding device is used, for example, to bond electrodes of lithium-ion batteries and / or semiconductor elements, or to bond metals of the same or different types. The ultrasonic bonding device may also be used to bond semiconductors, ceramics, and / or resins, etc.

[0010] 1, the ultrasonic complex vibration device 10 includes a first vibration element 111 having a substantially cylindrical shape, an intermediate vibration element 110 having a substantially cylindrical, cylindrical, or bottomed cylindrical shape, and a second vibration element 112 having a substantially cylindrical or bottomed cylindrical shape. The first vibration element 111, the intermediate vibration element 110, and the second vibration element 112 constitute a "vibration element (or vibrator)." The ultrasonic bonding device includes the ultrasonic complex vibration device 10, a horn tip 140 (ultrasonic bonding tip), and an anvil 118.

[0011] The first vibration element 111 and the intermediate vibration element 110 are coaxially connected by a mechanical connecting mechanism (such as a bolt and / or a clamp mechanism) at the middle or intermediate portion of the ultrasonic complex vibration device 10. The intermediate vibration element 110 and the second vibration element 112 are coaxially connected by a mechanical connecting mechanism at the middle portion of the ultrasonic complex vibration device 10. The first vibration element 111, the intermediate vibration element 110, and the second vibration element 112 may be integrally configured rather than being mechanically connected.

[0012] The intermediate vibration element 110 may be a component of the first vibration element 111. That is, the first vibration element 111 may be composed of two vibration elements. In this case, the first vibration element 111 and the intermediate vibration element 110 may be integrally formed rather than being mechanically connected. The intermediate vibration element 110 may be a component of the second vibration element 112. That is, the second vibration element 112 may be composed of two vibration elements. In this case, the second vibration element 112 and the intermediate vibration element 110 may be integrally formed rather than being mechanically connected.

[0013] As shown in FIG. 1, the first vibration element 111 is provided with a piezoelectric body 1112 whose axial direction is the piezoelectric polarization direction.

[0014] As shown in FIG. 1 , the intermediate vibration element 110 has a generally annular plate-shaped intermediate flange 102 formed at a central position in the axial direction, which extends radially around the entire circumference. The intermediate vibration element 110 is configured to be clamped or supported around the entire circumference by a clamping mechanism (not shown) at least at the intermediate flange 102. If it is ensured that the intermediate vibration element 110 is supported by a mechanical support mechanism, the intermediate flange 102 may be omitted. As shown in FIG. 1 , the intermediate vibration element 110 has a generally cylindrical shape with a generally constant outer diameter in the axial direction behind the intermediate flange 102 (leftward in FIG. 1 ). As shown in FIG. 1 , the intermediate vibration element 110 has a generally cylindrical shape (a shape in which a generally truncated conical shape and a generally cylindrical shape are coaxially connected) with a generally constant outer diameter after continuously tapering partway toward the tip of the intermediate vibration element 110 (rightward in FIG. 1 ).

[0015] 1, the second vibration element 112 is provided with a frequency adjustment element 1120 having a generally regular octagonal shape with rounded corners that protrudes radially around the entire circumference at a midpoint in the axial direction of the second vibration element 112. The frequency adjustment element 1120 adjusts the resonance frequencies of the longitudinal vibration component and the torsional vibration component of the ultrasonic vibration.

[0016] 1, the second vibration element 112 has a plurality of slits 1124 formed on its outer surface rearward of the frequency adjustment element 1120. A plurality of slits 1124 may be formed on the outer surface of the second vibration element 112 forward of the frequency adjustment element 1120. The slits 1124 extend obliquely in the second vibration element 112 when viewed from the side, or extend in the axial direction while being displaced in the circumferential direction in phase with each other. N (N=2, 3, ...) slits 1124 may be arranged to have N-fold rotational symmetry (e.g., N=8, 12, or 16) around the central axis of the second vibration element 112.

[0017] 1, the second vibration element 112 is provided with a tip portion 1126 of a generally regular octagonal shape with rounded corners that protrudes radially around the entire circumference at the tip position in the axial direction. Holes 1128 (or through holes) are formed in the tip portion 1126 at a plurality of locations spaced apart in the circumferential direction. The N (N=2, 3, ...) holes 1128 may be arranged to have N-fold rotational symmetry (e.g., N=4) around the central axis of the second vibration element 112. A female thread is provided on the inner surface of the hole 1128.

[0018] Horn tip 140 has a substantially truncated cone-shaped base portion and a tip portion that abuts against workpiece W1, the uppermost of workpieces W1 and W2. A male thread provided at the base end of horn tip 140 screws into a female thread provided in hole 1128 in tip portion 1126 of second vibration element 112, thereby removably fixing horn tip 140 to second vibration element 112. Horn tips 140 of various shapes are available, so that horn tips 140 can be appropriately replaced depending on the type of metal to be joined, etc.

[0019] The balancer for adjusting the phase difference between the longitudinal vibration and the torsional vibration at the tip 1126 of the second vibration element 112, and therefore at the horn tip 140, may be removably fixed to the tip 1126 of the second vibration element 112 by screwing the male thread of the balancer into the female thread of the hole 1128.

[0020] Anvil 118 is disposed so as to face the tip of horn tip 140 in the vertical direction. For example, approximately flat workpieces W1 and W2 are placed one on top of the other on the upper surface of anvil 118. Anvil 118 may be configured to passively or actively displace up and down in response to the pressure of horn tip 140 that it receives through workpieces W1 and W2.

[0021] As shown in FIG. 1, the ultrasonic bonding apparatus further includes a control device 120 , a high-frequency power supply device 121 , a pressure device 122 , a status sensor 124 , and an interface device 126 .

[0022] The high-frequency power supply device 121 is configured to apply a high-frequency AC voltage to the piezoelectric body 1112 of the first vibration element 111 in response to power supplied from a commercial power source (not shown), thereby exciting the first vibration element 111 in the axial direction. The pressure device 122 is equipped with a pressure block, and is configured to apply pressure to the workpieces W1 and W2 from the horn tip 140 by displacing a support mechanism, such as a clamp mechanism, that supports the intermediate vibration element 110 using the pressure block. The status sensor 124 includes a stroke sensor that outputs a signal corresponding to the amount of displacement of the pressure block that constitutes the pressure device 122, as well as an amplitude sensor that outputs a signal corresponding to the amplitude of the horn tip 140 (corresponding to a specified parameter). The amplitude sensor may be a sensor module configured with an imaging device and a device that calculates the amplitude by analyzing an image acquired through the imaging device.

[0023] As specified parameters that change depending on the progress of joining one workpiece W1 and the other workpiece W2, in addition to the amplitude A of the horn tip 40, the axial displacement, displacement speed and / or displacement acceleration of the vibration element (e.g., the second vibration element 12) may be measured as specified parameters.

[0024] The interface device 126 is, for example, configured as a display, and displays or outputs on the display the displacement amount of the pressure block and / or the time series of the pressure according to the output signal of the status sensor 124. The display may be configured as a touch panel display, and may be configured to accept a setting operation for allowing the user to directly or indirectly specify parameters, such as one of a plurality of bonding modes that defines a time series pattern of the target pressure.

[0025] The control device 120 is configured with a microcomputer, an arithmetic processing unit (CPU, microprocessor, processor core, etc.), and a storage device (memory such as ROM and RAM). The control device 120 is configured to control the displacement operation of the pressure block by the pressure device 122, for example, based on a time series of the displacement amount of the pressure block represented by an output signal of a stroke sensor constituting the state sensor 124. The control device 120 is configured to control the power supplied to the piezoelectric body 1112 based on the amplitude (corresponding to a specified parameter) of the horn tip 140 represented by an output signal of an amplitude sensor constituting the state sensor 124, and thereby to control the ultrasonic vibration power of the vibration elements (first vibration element 111, middle vibration element 110, and second vibration element 112) and the ultrasonic vibration power of the horn tip 140. As the status sensor 124, a pressure sensor is provided that outputs a signal corresponding to the pressure acting on the intermediate vibration element 110 from the pressure block of the pressure device 122 (~ the pressure that the horn tip 140 applies to the workpieces W1 and W2), and the control device 120 may control the time series of the pressure to be constant or in a specified manner based on the output signal of the pressure sensor.

[0026] (Configuration of the bonding condition estimation device) The bonding state estimation device 20 according to one embodiment of the present invention shown in FIG. 2 includes an input interface 21, a bonding state estimation model generation unit 22, a bonding state estimation model evaluation unit 222, a bonding state estimation unit 23, and an output interface 24. The bonding state estimation device 20 is configured by a computer such as a server computer (including a cloud server), a personal computer, a smartphone, or a tablet computer. Each component of the bonding state estimation device 20 includes an arithmetic processing unit (comprised of a CPU and a processor (core)), a storage device (comprised of memory (ROM, RAM, etc.)), an I / O circuit, and the like. The components of the bonding state estimation device 20 have a function in which the arithmetic processing unit reads necessary programs (bonding state learning program and / or bonding state estimation program (software)) and data from the storage device and executes designated arithmetic processing in accordance with the program based on the data.

[0027] The input interface 21 includes a learning input data acquisition unit 211, a learning output data acquisition unit 212, and an estimation input data acquisition unit 213. The input interface 21 may include touch panel buttons, a keyboard, and / or a voice input device.

[0028] The learning input data acquisition unit 211 is configured to acquire the i-th learning input data (i=1, 2, . . . N). A sound (vibration signal) generated during one ultrasonic welding operation (described later) of the workpieces W1 and W2 by the ultrasonic welding device (see FIG. 1) is collected by the microphone M. The microphone M may be a component of the input interface 21. The "i-th learning input data" may include, for example, a part of the spectrogram of the sound shown in FIG. 6, which is the i-th time domain [t i-1 ,t i7, the ith partial spectrogram data may be composed of split spectrogram data representing a plurality of split spectrograms obtained by splitting the ith partial spectrogram into a plurality of frequency regions.

[0029] The learning output data acquisition unit 212 is configured to acquire learning output data. The "learning output data" includes at least one of quantitative evaluation data (bonding strength, base material bond area, interface bond area, and their composition ratio, amount of contamination, amount of plastic deformation) regarding the bonding of the workpieces W1 and W2, or data representing at least one pair (or one of) non-quantitative evaluations ("bonding index value (label)"), such as comments indicating the bonding status (full base material fracture, partial base material fracture, interface fracture, tool escape, upper plate fracture, etc.). Quantitative evaluation data regarding the bonding of the workpieces W1 and W2 includes, for example, a tensile strength test (JIS L 1096) or a peel strength test (JIS Z 3140) in accordance with JIS standards. However, a quality evaluation method unique to each secondary battery manufacturer may also be used to evaluate the bonding condition of irregularly shaped parts.

[0030] The estimation input data acquisition unit 213 is configured to acquire the i-th estimation input data. The "estimation input data" includes the i-th partial spectrogram SP, which is a part of the spectrogram of the sound generated during one ultrasonic joining operation (described later) of the workpieces W1 and W2. i The i-th partial spectrogram data for

[0031] The bonding state estimation model generation unit 22 generates an i-th trained model by machine learning the correlation between the i-th training input data group and the bonding states of the workpieces W1 and W2 using a learning model, such as "supervised learning," "semi-supervised learning," or "unsupervised learning," based on the i-th training data group (consisting of multiple i-th training data sets) that includes at least the i-th training input data. The learning model is stored and held in a storage device, or is read from a database and then stored in a storage device, and is then read from the storage device.

[0032] The bonding condition estimation model evaluation unit 222 evaluates an index value representing the level of estimation accuracy of the bonding condition of the workpieces W1 and W2 by the i-th trained model. Based on the evaluation result, the bonding condition estimation model evaluation unit 222 selects an appropriate trained model from the perspective of estimating the bonding condition of the workpieces W1 and W2.

[0033] The bonding condition estimation unit 23 is configured to input the i-th estimation input data acquired by the estimation input data acquisition unit 213 into the i-th trained model generated by the bonding condition estimation model generation unit 22, thereby outputting an estimation result of the bonding condition of the workpieces W1 and W2.

[0034] The output interface 24 is configured to output the estimation results of the bonding state of the workpieces W1 and W2 output by the bonding state estimation unit 23, and is composed of an image output device or an image output device and an audio output device.

[0035] The "bonding state learning device" according to one embodiment of the present invention is configured with a learning input data acquisition unit 211 and a learning output data acquisition unit 212, which are included in the input interface 21, and a bonding state estimation model generation unit 22. The "bonding state estimation device" according to one embodiment of the present invention is configured with an estimation input data acquisition unit 213, which is included in the input interface 21, a bonding state estimation unit 23, and an output interface 24.

[0036] (Functions of ultrasonic bonding equipment) The following describes the steps of the ultrasonic bonding method for the workpieces W1 and W2 using the ultrasonic bonding device. A reference amplitude A0 is set (FIG. 3 / STEP 100). For example, the reference amplitude A0 may be directly or indirectly designated by the user via a touch panel display constituting the interface device 126, such as by selecting one of a plurality of bonding modes for which the reference amplitude A0 is determined.

[0037] Pressurizing device 122 moves first vibration element 111, intermediate vibration element 110, second vibration element 112, and horn tip 140 downward (FIG. 3 / STEP 102). Pressure P that the tip of horn tip 140 (free end of rod portion 44) receives from workpieces W1 and W2 is measured based on the output signal of the pressure sensor that constitutes state sensor 124 (FIG. 3 / STEP 104). When the tip of horn tip 140 is separated from workpiece W1, P=0, and when it abuts against workpiece W1, P>0.

[0038] It is determined whether pressure P received by horn tip 140 is equal to or greater than designated pressure P0 (FIG. 3 / STEP 106). If the determination result is negative (FIG. 3 / STEP 106...NO), pressure device 122 moves first vibration element 111, intermediate vibration element 110, second vibration element 112, and horn tip 140 downward (FIG. 3 / STEP 102). This adjusts the vertical position of horn tip 140, and therefore the static pressure applied from the tip of horn tip 140 to workpieces W1 and W2, to fall within a designated static pressure range (e.g., 200 N to 800 N).

[0039] If the determination result is affirmative (FIG. 3 / STEP 106...YES), ultrasonic vibrations are generated in the vibration elements (FIG. 3 / STEP 108). Specifically, in response to power being supplied to the high-frequency power supply device 121 from a commercial power source (not shown), the high-frequency power supply device 121 applies a high-frequency AC voltage to the piezoelectric body 1112 of the first vibration element 111. This causes the first vibration element 111 to vibrate in its axial direction at, for example, approximately 20 KHz, generating ultrasonic vibrations. The ultrasonic vibrations are transmitted from the first vibration element 111 to the intermediate vibration element 110 in its axial direction, and the amplitude of the ultrasonic vibrations is amplified. Furthermore, the ultrasonic vibrations with amplified amplitude are transmitted from the intermediate vibration element 110 to the second vibration element 112 in its axial direction.

[0040] In this way, a portion of the longitudinal vibration component (axial component of the second vibration element 112) of the ultrasonic vibration transmitted to the second vibration element 112 is converted into a torsional vibration component by the multiple slits 1124 formed on the outer surface of the second vibration element 112. Then, a composite vibration generated by combining the longitudinal vibration component and the torsional vibration component is transmitted to the horn tip 140 fixed to the tip portion 1126 of the second vibration element 112.

[0041] In response, the tip of the horn tip 140 displaces or vibrates horizontally, tracing a circular or elliptical orbit. As a result, as shown schematically in FIG. 4, the amplitude and ultrasonic vibration power of the horn tip 140 gradually increase from the vibration start time t=t0. During this process, impurities on the contact surfaces of the workpieces W1 and W2 are removed, further promoting plastic deformation of the contact surfaces of the workpieces W1 and W2. Then, as shown in FIG. 4, the rate of increase in the amplitude and ultrasonic vibration power of the horn tip 140 significantly decreases at time t=t1, after which the amplitude and ultrasonic vibration power of the horn tip 140 gradually increase. This is because the oxide coatings and other metal films constituting the joining surfaces of the workpieces W1 and W2 are removed, revealing clean, activated metal atoms on the joining surfaces. The temperature rise due to frictional heat activates the atomic movement, generating a mutual attraction between the atoms.

[0042] At this time, the amount of pressing of the workpieces W1 and W2 by the horn tip 140 and / or the static pressure applied to the workpieces W1 and W2 is adjusted, and a composite vibration is applied to one of the workpieces W1 and W2, thereby solid-state joining the workpieces W1 and W2.

[0043] An amplitude sensor constituting status sensor 124 optically measures amplitude A at a specified location (e.g., a location where the amplitude is relatively large) of horn tip 140 (FIG. 6 / STEP 10). It is determined whether the time derivative δA (= current amplitude A(k)−previous amplitude A(k−1)) of amplitude A of horn tip 140 is negative and whether amplitude A is equal to or less than reference amplitude A0 (FIG. 6 / STEP 12). Instead of this determination process, it may be determined whether amplitude A has decreased by reference amplitude A0 (or a reference ratio based on the maximum value) using the maximum value at the time when amplitude A started to decrease as a reference.

[0044] If the determination result is negative (FIG. 6 / STEP 12...NO), ultrasonic vibrations are continuously generated in the vibration element (FIG. 3 / STEP 108). On the other hand, if the determination result is positive (FIG. 6 / STEP 12...YES), the generation of ultrasonic vibrations in the vibration element is stopped (FIG. 6 / STEP 14). For example, as shown in FIG. 4, after the amplitude of horn tip 140 changes from increasing to decreasing and becomes equal to or less than reference amplitude A0 at time t=t2, the ultrasonic power of horn tip 140 is controlled to become 0 with a slight response delay.

[0045] (Method for estimating bonding condition (first embodiment)) In the first embodiment, the learning model is made to learn the correlation between the i-th learning input data and learning output data by "supervised learning."

[0046] The learning input data acquiring unit 211 acquires the ith learning input data (i=1, 2, . . . N). Specifically, joining acoustic data generated when the workpieces W1 and W2 are joined using the ultrasonic complex vibration device 10 is acquired through the microphone M (FIG. 4 / STEP 210). As a result, for example, as shown in FIG. 6, an acoustic signal with time on the horizontal axis and sound intensity on the vertical axis is acquired as the joining acoustic data.

[0047] A portion of the joint acoustic data is extracted in the time domain (FIG. 4 / STEP 211). For example, as shown in FIG. 6, the i-th time domain t=t i+1- ~t i+1+ At least one of the start time, end time, and length of each of the ith time domain and the jth time domain (i≠j) is changed, thereby extracting the ith joint acoustic data and the jth joint acoustic data included in the mutually different ith time domain and jth time domain, respectively. The ith time domain and the jth time domain may at least partially overlap in the time domain, or may be spaced apart.

[0048] The i-th learning output data is acquired by the learning output data acquisition unit 212. Specifically, the quality of the bonding state evaluated based on the bonding strength of the workpieces W1 and W2 is acquired as a label (FIG. 4 / STEP 212).

[0049] The training input data acquisition unit 211 generates a spectrogram by frequency analyzing the extracted or clipped joint acoustic data (FIG. 4 / STEP 213). As a result, an i-th spectrogram is generated based on the i-th joint acoustic data, with the horizontal axis representing time and the vertical axis representing frequency, and with brightness representing signal strength, as shown in FIG. 7. The i-th spectrogram group generated as described above is stored in a storage device and / or database as a joint acoustic data set (constituting training input data) (FIG. 4 / STEP 214).

[0050] Next, the i-th spectrogram is divided in the frequency domain (FIG. 4 / STEP 215). As a result, for example, as shown in FIG. 7, the i-th spectrogram is divided into a plurality of frequency domains (width Δf) to generate divided spectrograms. If the number of i-th spectrograms constituting the i-th spectrogram group is greater than the reference number, the division process of the i-th spectrogram in the frequency domain may be omitted.

[0051] Then, the prediction problem of the bonding state of the workpieces W1 and W2 is optimized by the machine learning model (FIG. 4 / STEP 216). Specifically, the i-th learning input data constituting the i-th learning data is input to the learning model, and model output data is output from the learning model. The "i-th learning input data" includes an i-th spectrogram (or multiple i-th split spectrograms obtained by splitting the i-th spectrogram) based on bonding acoustic data extracted from the i-th time domain in the time domain. Based on the comparison result between the model output data and the i-th learning output data constituting the i-th learning data, model parameters defining the learning model are identified. The "i-th learning output data" includes a label (a bonding index value indicating the quality of the bonding state of the workpieces W1 and W2).

[0052] When "supervised learning" is performed, for example, a neural network model is used as a learning model to generate the i-th trained model. According to the neural network model, the i-th training input data is input to the input layer, and the i-th model output data is output from the output layer as an estimation or inference result. It is determined whether the error between the i-th model output data and the i-th training output data (teacher data) included in the i-th training data is equal to or less than a reference value (FIG. 4 / STEP 217).

[0053] If the result of the determination is negative (FIG. 4 / STEP 217...NO), further optimization of the prediction problem of the bonding state of the workpieces W1 and W2 using a machine learning model is attempted (FIG. 4 / STEP 216).On the other hand, if the result of the determination is positive (FIG. 4 / STEP 217...YES), the i-th trained model defined by the model parameters identified at that stage is saved in a storage device and / or database (FIG. 4 / STEP 218).

[0054] The i-th trained model is generated for each combination of the specifications of the workpieces W1 and W2 (shape, size, material, surface roughness, etc.) and the specifications of the ultrasonic joining device (structure of the ultrasonic complex vibration device 10, shape, size, material, presence and type of uneven structure on the tip surface, reference amplitude A0, specified pressure P0, etc.).

[0055] Of the N trained models (first trained model, second trained model, ... i-th trained model, ... N-th trained model) generated using the above procedure, the "accuracy rate" is determined or evaluated as an index value representing the level of estimation accuracy of the trained models. In addition to or instead of the "accuracy rate," "precision," "recall," and / or "F-score" may be determined as index values ​​for the i-th trained model. Other index values ​​for the i-th trained model that may be evaluated include the "true positive rate," "false positive rate," and / or "specificity rate" using an ROC curve, or (in the case of a regression model) the "variance explained rate," "MAE (mean absolute error)," "MSE (mean squared error)," "RMSE (mean squared error)," and / or "coefficient of determination R2." Based on the accuracy rate, etc., an appropriate trained model (for example, the jth trained model (1≦j≦N)) is selected from the N trained models in terms of estimating the bonding state of the workpieces W1 and W2 and is stored in a storage device and / or database.

[0056] The jth estimation input data is acquired by the estimation input data acquisition unit 213. The jth estimation input data is acquired according to the same procedure as the jth training input data. Specifically, joining acoustic data generated when the workpieces W1 and W2 are joined using the ultrasonic complex vibration device 10 is acquired through the microphone M (FIG. 5 / STEP 220 (see FIG. 6)). A portion included in the jth time domain as part of the joining acoustic data is extracted in the time domain (FIG. 5 / STEP 221). The extracted or clipped joining acoustic data is subjected to frequency analysis to generate the jth spectrogram (FIG. 5 / STEP 223 (see FIG. 7)). The jth spectrogram group generated as described above is stored in a storage device and / or database as a joining acoustic data set (constituting the estimation input data) (FIG. 5 / STEP 224). Next, the jth spectrogram is divided in the frequency domain (FIG. 5 / STEP 225). The division process of the jth spectrogram in the frequency domain may be omitted.

[0057] Next, the jth input data for estimation is input to the jth trained model, and a joint index value representing the quality of the joint state of the workpieces W1 and W2 is estimated (FIG. 5 / STEP 226). The "jth input data for estimation" includes the jth spectrogram (or a plurality of jth split spectrograms obtained by splitting the jth spectrogram) based on the joint acoustic data extracted from the jth time domain in the time domain.

[0058] It is determined whether the bonding index value is equal to or greater than the standard value (FIG. 5 / STEP 227). If the determination result is affirmative (FIG. 5 / STEP 227...YES), the bonding condition of the workpieces W1 and W2 is determined to be "good" (FIG. 5 / STEP 228). If the determination result is negative (FIG. 5 / STEP 227...NO), the bonding condition of the workpieces W1 and W2 is determined to be "poor" (FIG. 5 / STEP 229).

[0059] The estimation result of the bonding state of the workpieces W1 and W2 as the model output data is output through the output interface 24.

[0060] (Method for estimating bonding condition (second embodiment)) The second embodiment differs from the first embodiment in that "unsupervised learning" is adopted as the machine learning method instead of "supervised learning." The following mainly describes the differences from the first embodiment.

[0061] The ith learning data group including the ith learning input data acquired by the learning input data acquisition unit 211 is acquired (see FIG. 4 / STEP 210, STEP 211, STEP 213, STEP 214, and STEP 215). That is, the learning data acquisition unit is not configured with both the learning input data acquisition unit 211 and the learning output data acquisition unit 212, but is configured with only the learning input data acquisition unit 211. Label acquisition processing is omitted (see FIG. 4 / STEP 212).

[0062] The bonding condition estimation model generation unit 22 inputs the ith learning data into the learning model, causing the unsupervised learning model to learn the correlation between the ith learning input data included in the ith learning data and data indicating that the bonding condition is normal or good, and generates the ith bonding condition estimation model (ith trained model) (see Figure 4 / STEP 216, STEP 217, STEP 218).

[0063] For example, an autoencoder (AE) model is employed as the learning model. According to the autoencoder model, for example, i-th learning input data relating to the normal or good bonding state of workpieces W1 and W2 is input to the input layer as learning data. Subsequently, in the intermediate layer, scores are assigned (weighted) according to the importance of the data, and data with low scores are eliminated (encoded). Weighting is then applied again when moving to the output layer, and the sum of data received from multiple edges is output (decoded) as the final value. Through this procedure, patterns and / or trends possessed by the i-th learning input data are learned so that the model output data matches the i-th learning input data. This series of steps is repeatedly executed, and when specified conditions are met, learning by the learning model is completed, and the i-th trained model is generated.

[0064] As the autoencoder (AE), a stacked autoencoder (SAE), a convolutional autoencoder (CAE), a variational autoencoder (VAE) or a conditional autoencoder may be employed.

[0065] Thereafter, the estimation input data acquisition unit 213 acquires the jth partial spectrogram data derived from the joining sound of the workpieces W1 and W2, whose joining state is unknown, as estimation input data (see FIG. 5 / STEP 220, STEP 221, STEP 223, STEP 224, STEP 225).

[0066] Next, the bonding condition estimation unit 23 inputs the input data for estimation to the input layer of the jth trained model selected as described above. In response to this, an estimation result of the bonding index value indicating the quality of the bonding condition of the workpieces W1 and W2 is acquired as model output data output from the output layer of the jth trained model (see FIG. 5 / STEP 226). A deviation is calculated between the input data for estimation input to the bonding condition estimation model or a feature value based thereon and the model output data output from the jth bonding condition estimation model or a feature value based thereon. Furthermore, if the deviation is less than a threshold, the bonding condition is determined to be "good," and if the deviation is equal to or greater than the threshold, the bonding condition is determined to be "poor" (see FIG. 5 / STEP 227, STEP 228, and STEP 229).

[0067] Then, the estimated results of the joint state of the target soil as the model output data are output through the output interface 24.

[0068] (Other embodiments) In the above-described embodiments, a neural network (first embodiment) or an autoencoder (second embodiment) is used as the learning method, but any other machine learning method may be used. For example, tree-type models such as decision trees and regression trees, ensemble learning such as bagging and boosting, neural network-type models (including deep learning) such as recurrent neural networks and convolutional neural networks, clustering-type models such as hierarchical clustering, non-hierarchical clustering, k-nearest neighbors, and k-means, multivariate analysis such as principal component analysis (PCA), factor analysis, and logistic regression, and support vector machines (SVM) may be used as learning models. In addition to supervised learning and unsupervised learning, semi-supervised learning may also be used.

[0069] In the above embodiment, the microphone M collects "sound" as a vibration signal generated during one ultrasonic bonding operation of the workpieces W1 and W2 by the ultrasonic bonding device (see FIG. 1). In another embodiment, the vibration propagating to the vibration element, the anvil 118, or a structure connected to or in contact with the vibration element or the anvil 118 may be collected as the vibration signal by the vibration sensor. From the spectrogram of the vibration signal detected by the vibration sensor, a part of it, the i-th partial spectrogram SP i The i-th partial spectrogram data for is acquired.

[0070] The vibration sensor may be a contact type (such as a frequency change type, piezoelectric type, electrodynamic type, or servo type) vibration sensor and / or a non-contact type (such as an eddy current type, capacitance type, or optical type) vibration sensor. The vibration sensor may be a component of the input interface 21. The optical vibration sensor may be a single laser Doppler vibration sensor, a scanning laser Doppler vibration sensor, or a microscope type laser Doppler vibration sensor. [Explanation of symbols]

[0071] 10. Ultrasonic complex vibration device 102...Intermediate flange 110. Intermediate vibration element 111...First vibration element 1112...Piezoelectric material 112...Second vibration element 1120...Frequency adjustment element 1121...Cylindrical part 1122...Cylindrical part 1124...Slit 1126‥Tip 1128...hole 118. Anvil 120. Control device 121‥High frequency power supply equipment 122. Pressure device 124. Status sensor 126. Interface device 140...Horn tip 20. Bonding condition estimation device 21. Input interface 211...Learning input data acquisition unit 212...Learning output data acquisition unit 213...Estimation input data acquisition unit 22. Bonding condition estimation model generation unit 222... Bonding condition estimation model evaluation section 23. Bonding condition estimation section 24. Output interface W1: One of the workpieces W2: The other work.

Claims

1. a learning data acquisition unit that acquires a learning data group consisting of a plurality of sets of learning data including, as learning input data, partial spectrogram data representing a partial spectrogram whose range is specified by a time domain, which is a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device; a model generation unit that generates a trained model by having a training model learn a correlation between the training input data and the bonding states of the plurality of workpieces based on the training data group acquired by the training data acquisition unit; and a model evaluation unit that evaluates each of the plurality of trained models generated by the model generation unit based on each of a plurality of training data groups including data representing each of the plurality of partial spectrograms having different ranges. Joint state learning device.

2. The bonding state learning device according to claim 1, The learning data acquisition unit acquires a learning data group consisting of a plurality of sets of learning data including, as learning input data, partial spectrogram data representing a plurality of divided spectrograms obtained by dividing the partial spectrogram in the frequency domain. Joint state learning device.

3. The bonding state learning device according to claim 1, the learning data acquisition unit acquires the plurality of sets of learning data including data representing the joining states of the plurality of workpieces as learning output data; The model generation unit generates the trained model by having the training model learn the correlation between the training input data and the training output data by supervised learning or semi-supervised learning. Joint state learning device.

4. The bonding state learning device according to claim 1, The model generation unit generates the trained model by having the training model learn features of the plurality of training input data included in each of the plurality of sets of training data through unsupervised training. Joint state learning device.

5. an estimation input data acquisition unit that acquires, as estimation input data, partial spectrogram data representing a partial spectrogram whose range is specified by a time domain, which is a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device; a bonding state estimation unit that inputs the input data acquired by the estimation input data acquisition unit into the trained model selected based on the evaluation result by the model evaluation unit of the bonding state learning device according to any one of claims 1 to 4, and thereby acquires an estimation result of the bonding states of the plurality of workpieces as an output from the trained model. Bonding condition estimation device.

6. a learning data acquisition step of acquiring a learning data group consisting of a plurality of sets of learning data including, as learning input data, data representing a partial spectrogram, the partial spectrogram being a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device, the range of which is specified by a time domain; a model generation step of generating a trained model by having a training model learn a correlation between the training input data and the bonding states of the plurality of workpieces based on the training data group acquired in the training data acquisition step; and a model evaluation step of evaluating each of the plurality of trained models generated in the model generation step based on each of a plurality of training data groups including data representing each of the plurality of partial spectrograms having different ranges. A bonding state learning program for causing a computer to execute a bonding state learning method.

7. an estimation input data acquisition step of acquiring, as estimation input data, partial spectrogram data representing a partial spectrogram, the partial spectrogram being a part of a spectrogram of a vibration signal generated during a joining process of a plurality of workpieces by the ultrasonic joining device, the range of which is specified by a time domain; and a bonding state estimation step of inputting the input data acquired in the estimation input data acquisition step into the trained model selected based on the evaluation result in the model evaluation step of the bonding state learning method according to claim 6, thereby acquiring an estimation result of the bonding states of the plurality of workpieces as an output from the trained model. A program for estimating junctional states.

Images