JP2022092311A - Abnormality detection system, abnormality detection device, abnormality detection method, program and learnt model - Google Patents

Abstract

To provide an abnormality detection system, an abnormality detection device, an abnormality detection method, a program, and a learnt model for securely detecting a clogging of a gate of a molding device.SOLUTION: An abnormality detection system includes a molding device and an abnormality detection device for detecting an abnormality of the molding device. The molding device includes a mold part provided with a plurality of gates and a mold cavity connecting with the plurality of gates, formed inside, an injection part for filling a molding material in a molten state to the mold cavity via the plurality of gates, a first sensor closest to a first gate among the plurality of gates for detecting a pressure or a temperature of the molding material in the mold cavity, and a second sensor closest to a second gate for detecting a pressure or a temperature of the molding material in the mold cavity. The abnormality detection device includes a data acquisition unit for acquiring a first evaluation value of the pressure or the temperature detected by the first sensor and a second evaluation value of the pressure or the temperature detected by the second sensor and an abnormality detection unit for detecting a clogging in the first gate or the second gate based on the first evaluation value and the second evaluation value.SELECTED DRAWING: Figure 1

Description

The present invention relates to anomaly detection systems, anomaly detection devices, anomaly detection methods, programs and trained models.

An injection molding apparatus is known in which a molten liquid in which a molding material is melted is supplied to a cavity formed between a plurality of molds to mold a molded product. For example, Patent Document 1 discloses a technique in which an injection speed is detected by an encoder attached to an injection servomotor and the injection operation is stopped when the injection speed does not reach the monitoring speed during the monitoring time of the injection process. It has been disclosed.

JP-A-2009-000929

In the molding apparatus, if the gate is clogged, the molding material may not flow into the cavity as designed, and an abnormality may occur in the molded product. In the technique of Patent Document 1, when the gate is clogged, the load pressure at the time of injection becomes higher than the load pressure at the time of molding a good product, and the injection speed becomes slower than the injection speed at the time of molding a good product. Detect clogging.

However, the inventors have discovered that even if the gate is clogged, the load pressure at the time of injection may hardly change from the time of molding a good product. In such a case, the technique of Patent Document 1 cannot detect the clogging of the gate.

Therefore, an object of the present invention is to provide an anomaly detection system, an anomaly detection device, an anomaly detection method, a program, and a trained model for more reliably detecting a gate blockage.

(1) The abnormality detection system of the present disclosure is an abnormality detection system including a molding device for molding a molded product and an abnormality detection device for detecting an abnormality in the molding device, and the molding device has a plurality of gates. A mold portion formed inside the cavity connected to the plurality of gates, an injection portion for filling the cavity with the molded material in a molten state via the plurality of gates, and the plurality of gates. Closest to the first sensor, which is provided in the region closest to the first gate and detects the pressure or temperature of the molding material in the cavity, and the second gate, which is different from the first gate among the plurality of gates. A second sensor provided in the region to detect the pressure or temperature of the molding material in the cavity, the anomaly detecting device is a first time series of pressure or temperature detected by the first sensor. A data acquisition unit that acquires a first evaluation value calculated based on the data and a second evaluation value calculated based on the second time-series data of the pressure or temperature detected by the second sensor. It is an abnormality detection system including an abnormality detection unit for detecting a clogging of the first gate or the second gate based on the first evaluation value and the second evaluation value.

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. This makes it possible to detect clogging of some gates that could not be detected in the past. As a result, the gate clogging can be detected more reliably.

(2) Preferably, the first distance through which the molding material passes from the first gate to the first sensor is equal to the second distance through which the molding material passes from the second gate to the second sensor.

With this configuration, when the gate is "not" clogged, the time points at which the molding material reaches the first sensor and the second sensor can be aligned. This makes it possible to make a difference between the first time series data and the second time series data only when some of the gates are clogged. As a result, the accuracy of clogging detection can be improved.

(3) Preferably, the first evaluation value includes the first rise time point of the first time series data or the first peak time point when the pressure or temperature of the first time series data becomes the maximum value. The second evaluation value includes the second rising point of the second time series data or the second peak time point when the pressure or temperature of the second time series data becomes the maximum value, and the abnormality detecting unit is the second time. When the time point of one rise is delayed from the time point of the second rise, or when the time point of the first peak is delayed from the time point of the second peak, the clogging of the first gate is detected.

(4) Preferably, the first sensor and the second sensor detect the pressure of the molding material, and the first evaluation value is the first peak value which is the maximum value of the pressure of the first time series data. The second evaluation value includes the second peak value which is the maximum value of the pressure of the second time series data, and the abnormality detection unit has the first peak value lower than the second peak value. In this case, the clogging of the first gate is detected.

(5) Preferably, the first sensor and the second sensor detect the pressure of the molding material, and the first evaluation value is the first rise point of the pressure of the first time series data. The second evaluation value includes the first integral value which is the time integration up to the first peak time when the pressure of the time series data becomes the maximum value, and the second evaluation value is the second rise point of the pressure of the second time series data. 2 When the abnormality detection unit includes the second integral value which is the time integral up to the time of the second peak when the pressure of the time series data becomes the maximum value, the first integral value is smaller than the second integral value. , The clogging of the first gate is detected.

According to the above configurations (3) to (5), the first evaluation value includes at least one of the first rise time point, the first peak time point, the first peak value, and the first integral value, and the second evaluation value. Includes at least one of a second rising time point, a second peak time point, a second peak value and a second integral value corresponding to the first evaluation value. Then, by comparing the first rising time and the second rising time, the first peak time and the second peak time, the first peak value and the second peak value, or the first integrated value and the second integrated value, the first Of the 1st gate and the 2nd gate, it can be specified that the 1st gate is clogged. That is, according to the abnormality detection system, it is possible not only to detect the clogging of some gates that could not be detected in the past, but also to specify which gate is clogged.

(6) Preferably, the abnormality detection unit applies the first evaluation value and the first evaluation value to a trained model in which the correlation between the first evaluation value and the second evaluation value and the clogging of the plurality of gates is machine-learned. By inputting the second evaluation value, clogging of at least one of the first gate and the second gate is detected.

With such a configuration, once the trained model is generated, it is possible to detect the clogging of a part of the gates from the first evaluation value and the second evaluation value. By using the trained model, it is possible to more accurately detect the clogging of some gates even when the molding conditions vary.

(7) The abnormality detecting device of the present disclosure is an abnormality detecting device for detecting an abnormality in a molding device for molding a molded product, and the molding device has a plurality of gates and a cavity connected to the plurality of gates. A mold portion formed inside, an injection portion for filling the cavity with a molded material in a molten state via the plurality of gates, and a region of the plurality of gates closest to the first gate are provided. A first sensor for detecting the pressure or temperature of the molding material in the cavity and a region of the plurality of gates closest to the second gate different from the first gate, the molding material in the cavity. It has a second sensor for detecting pressure or temperature, and the abnormality detecting device has a first evaluation value calculated based on the first time-series data of pressure or temperature detected by the first sensor. Based on the data acquisition unit that acquires the second evaluation value calculated based on the second time-series data of the pressure or temperature detected by the second sensor, the first evaluation value, and the second evaluation value. The abnormality detection device includes an abnormality detection unit for detecting clogging of at least one of the first gate and the second gate.

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. This makes it possible to detect clogging of some gates that could not be detected in the past. As a result, the gate clogging can be detected more reliably.

(8) In the abnormality detection method of the present disclosure, a mold portion in which a plurality of gates and cavities connected to the plurality of gates are formed inside, and molding of a molten state into the cavities via the plurality of gates. An abnormality detection method for detecting an abnormality in a molding apparatus including an injection portion for filling a material, wherein the cavity is detected by a first sensor provided in a region closest to the first gate among the plurality of gates. The first evaluation value calculated based on the first time-series data of the pressure or temperature of the molding material in the above, and the region closest to the second gate different from the first gate among the plurality of gates are provided. A data acquisition step for acquiring a second evaluation value calculated based on a second time-series data of the pressure or temperature of the molding material in the cavity detected by the second sensor, the first evaluation value, and the first evaluation value. It is an abnormality detection method including an abnormality detection step of detecting clogging of at least one of the first gate and the second gate based on the second evaluation value.

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. This makes it possible to detect clogging of some gates that could not be detected in the past. As a result, the gate clogging can be detected more reliably.

(9) In the program of the present disclosure, a mold portion in which a plurality of gates and cavities connected to the plurality of gates are formed therein, and a molding material in a molten state are provided into the cavities via the plurality of gates. A program for detecting an abnormality in a molding apparatus including an injection portion to be filled, and the inside of the cavity detected by a first sensor provided in a region closest to the first gate among the plurality of gates. The first evaluation value calculated based on the first time-series data of the pressure or temperature of the molding material, and the second of the plurality of gates provided in the region closest to the second gate different from the first gate. A data acquisition step for acquiring a second evaluation value calculated based on a second time-series data of the pressure or temperature of the molding material in the cavity detected by the sensor, the first evaluation value, and the first evaluation value. 2 It is a program that causes a computer device to execute an abnormality detection step of detecting clogging of at least one of the first gate and the second gate based on the evaluation value.

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. This makes it possible to detect clogging of some gates that could not be detected in the past. As a result, the gate clogging can be detected more reliably.

(10) In the trained model of the present disclosure, a mold portion in which a plurality of gates and a cavity connected to the plurality of gates are formed therein, and molding in a molten state into the cavity via the plurality of gates. A trained model for detecting anomalies in a molding apparatus comprising an injection section for filling a material, wherein the explanatory variable is a first sensor provided in the region closest to the first gate among the plurality of gates. The first evaluation value calculated based on the first time-series data of the pressure or temperature of the molding material in the cavity detected by the above, and the second gate different from the first gate among the plurality of gates are the most. The objective variable includes the second evaluation value calculated based on the second time series data of the pressure or temperature of the molding material in the cavity detected by the second sensor provided in the near region. A trained model that contains values for clogged gates out of multiple gates.

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. This makes it possible to detect clogging of some gates that could not be detected in the past. As a result, the gate clogging can be detected more reliably.

According to the present invention, clogging of the gate can be detected more reliably.

It is a block diagram which shows typically the abnormality detection system which concerns on 1st Embodiment. It is explanatory drawing which conceptually shows the molding apparatus which concerns on FIG. It is explanatory drawing which conceptually shows the molding apparatus which concerns on FIG. It is sectional drawing which shows the mold part of FIG. It is sectional drawing which shows the mold part cut by the cutting line of the arrow V of FIG. It is a figure explaining the clogging of the gate. It is an example of a graph showing time series data of pressure. It is an example of a graph showing time series data of temperature. It is a block diagram which shows the functional structure of the learning apparatus which concerns on 1st Embodiment. It is a block diagram which shows the functional structure of the abnormality detection apparatus which concerns on 1st Embodiment. It is a block diagram which shows typically the abnormality detection system which concerns on 2nd Embodiment. It is a block diagram which shows the functional structure of the abnormality detection apparatus which concerns on 2nd Embodiment.

<First Embodiment>
Hereinafter, the first embodiment of the present invention will be described with reference to the drawings.

<Overall configuration of anomaly detection system>
FIG. 1 is a block diagram schematically showing an abnormality detection system 10 according to the first embodiment. The abnormality detection system 10 includes a plurality of molding devices 20, a learning device 30, an abnormality detection device 40, an input unit 50, and a display unit 60.

The molding device 20, the learning device 30, the abnormality detection device 40, the input unit 50, and the display unit 60 are each provided so as to be able to communicate wirelessly or by wire. The learning device 30 and the abnormality detection device 40 are composed of an information processing device (computer device) having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.). The learning device 30 and the abnormality detection device 40 may be configured by the same information processing device or may be configured by different information processing devices.

In the present embodiment, a plurality of molding devices 20 are connected to one learning device 30 and one abnormality detecting device 40, and the learning device 30 and the abnormality detecting device 40 are used for various data transmitted from the plurality of molding devices 20. Based on this, learning and abnormality detection are performed. The molding device 20 may have a one-to-one correspondence with the learning device 30 and the abnormality detection device 40. That is, in the abnormality detection system 10, the number of molding devices 20 may be one, or a plurality of learning devices 30 and abnormality detection devices 40 may be provided.

The input unit 50 is, for example, a keyboard or a mouse, and receives various inputs from the operator. The display unit 60 is, for example, a display or a speaker, and displays various information in the abnormality detection system 10. The input unit 50 and the display unit 60 may be integrated, for example, a touch panel. Further, the input unit 50 and the display unit 60 may be provided as a portable terminal device so as to be movable to a place away from the molding device 20, the learning device 30, and the abnormality detection device 40.

<Outline configuration of molding equipment>
2 and 3 are explanatory views conceptually showing the molding apparatus 20. In FIGS. 2 and 3, the portion shown as a cross section is hatched. The molding apparatus 20 includes a bed 21, an injection unit 22, a mold clamping unit 23, a mold unit 24, a plurality of sensors 25, a temperature sensor 26, and a control panel 27. FIG. 2 shows a molding apparatus 20 in a state where the mold portion 24 is open, and FIG. 3 shows a molding apparatus 20 in a state where the mold portion 24 is combined. The molding apparatus 20 is a mold-clamping type injection molding apparatus.

The control panel 27 has a control unit 271 and a communication unit 272. The control unit 271 is electrically connected to each drive unit (motor 237, etc.) of the molding apparatus 20 and outputs an operation command to each drive unit. Further, the control unit 271 is electrically connected to each sensor (sensor 25 or the like) of the molding apparatus 20, and the signal detected by each sensor is input to the control unit 271. The control unit 271 is composed of an information processing device having a calculation unit (for example, CPU, GPU, etc.) and a storage unit (for example, HDD, SSD, etc.).

The communication unit 272 communicates with each other unit (learning device 30, etc.) of the abnormality detection system 10. The communication unit 272 transmits, for example, the signal detected by each of the sensors to the learning device 30 or the abnormality detection device 40. Further, the communication unit 272 receives the detection information described later from the abnormality detection device 40.

The mold clamping portion 23 includes a fixed plate 231, a movable plate 232, a tie bar 233, a ball screw 234, a support plate 235, a mold clamping force sensor 236, and a motor 237. The fixing plate 231 and the support plate 235 are fixed to the bed 21. The support plate 235 supports the ball screw 234. The ball screw 234 is connected to the motor 237. When the motor 237 is rotated by the operation command of the control unit 271, the ball screw 234 moves. A movable platen 232 is fixed to the end of the ball screw 234 opposite to the end connected to the motor 237.

Here, in the molding apparatus 20, the direction in which the ball screw 234 moves is referred to as "axial direction". The side where the motor 237 is located with respect to the ball screw 234 is referred to as "one side" in the axial direction, and the side where the movable platen 232 is located with respect to the ball screw 234 is referred to as "the other side" in the axial direction.

The movable platen 232 moves in the axial direction with the movement of the ball screw 234. The movable platen 232 is formed with a through hole 232a penetrating in the axial direction. The end of the tie bar 233 on one side in the axial direction is fixed to the support plate 235, and the end on the other side in the axial direction is fixed to the fixing plate 231. The tie bar 233 is inserted into the through hole 232a of the movable platen 232. As a result, the tie bar 233 guides the axial movement of the movable platen 232.

The mold clamping force sensor 236 detects the pressure (reaction force of the mold clamping force) applied from the ball screw 234 to the support plate 235. The mold clamping force sensor 236 outputs a detection signal regarding pressure to the control unit 271. The mold clamping force sensor 236 may be installed at another position as long as it can detect the mold clamping force in the mold portion 24 described later. The fixed plate 231 is formed with a through hole 231a having a diameter widened on the other side in the axial direction. A cylinder 222, which will be described later, is inserted into the through hole 231a.

The mold portion 24 has a plurality of molds 241 and 242. The mold 241 is fixed to the movable plate 232. When the ball screw 234 moves in the axial direction, the mold 241 moves in the axial direction together with the movable platen 232. That is, the mold 241 is a movable mold. The mold 242 is fixed to the fixing plate 231. That is, the mold 242 is a fixed mold.

See FIG. When the mold 241 is moved to the other side in the axial direction by the ball screw 234 and the mold 241 comes into contact with the mold 242 (that is, when a plurality of molds 241 and 242 are combined), between the molds 241 and 242. Cavity C1 is formed in. The cavity C1 of the present embodiment is an annular space filled with a molding material.

FIG. 4 is an enlarged cross-sectional view showing the mold portion 24 shown in FIG. The mold 242 has a spool 243, a plurality of (four in this embodiment) runners 244 branching from the spool 243, and a plurality of (four in this embodiment) gates connected to the plurality of runners 244, respectively. 245 and are formed inside. The cavity C1 is connected to a plurality of gates 245.

The spool 243 is a passage through which the molding material injected from the injection portion 22 to the mold portion 24 first flows. The end portion 243a on the injection portion 22 side of the spool 243 is open so as to communicate with the nozzle 224 described later. The spool 243 has the smallest inner diameter at the end 243a, and the inner diameter increases from the end 243a to the runner 244.

The plurality of runners 244 are slightly branched on the end portion 243a side of the end portion 243b on the cavity C1 side (that is, the opposite side of the end portion 243a) of the spool 243. The space on the end portion 243b side of the spool 243 from the position where the runner 244 branches is a resin reservoir for accumulating cold slug, and is also referred to as “cold slug well”.

FIG. 5 is a cross-sectional view showing a mold portion 24 cut by the cutting line shown by the arrow V in FIG.
The plurality of gates 245 have a one-to-one correspondence with the plurality of runners 244. Further, the inner diameters of the plurality of gates 245 decrease as they approach the cavity C1. Hereinafter, when a plurality of gates 245 are particularly distinguished, a predetermined one gate 245 (for example, a gate 245 located on the upper part of FIG. 5) is referred to as a gate GT1, and the remaining three gates are clockwise from the gate GT1. Gates 245 are referred to as gates GT2, GT3, and GT4, respectively.

Assuming that the position of the gate GT1 in the mold 24 is 0 degrees, the gate GT2 is at 90 degrees, the gate GT3 is at 180 degrees, and the gate GT4 is at 270 degrees. The gate GT3 is located on the opposite side of the gate GT1 and the gate GT4 is located on the opposite side of the gate GT2.

The plurality of sensors 25 are provided at positions facing the cavity C1 of the mold 241 and the mold 242. Each of the plurality of sensors 25 detects the pressure of the molding material in the cavity C1. The sensor 25 may be a sensor that detects the temperature. The sensor 25 outputs a detection signal regarding pressure (or temperature) to the control unit 271. In the present embodiment, as an example, a plurality of sensors 25 are provided in the mold 241. However, the sensor 25 may be provided in the mold 242.

The plurality of sensors 25 are provided at positions corresponding to the plurality of gates 245, respectively. In this embodiment, four sensors 25 correspond one-to-one to four gates 245. Hereinafter, when particularly distinguished, the sensor 25 corresponding to the gate GT1 among the plurality of sensors 25 is referred to as a sensor SN1. Similarly, the sensors 25 corresponding to the gates GT2, GT3, and GT4 are referred to as sensors SN2, SN3, and SN4, respectively.

The sensor SN1 (corresponding to the “first sensor” of the present disclosure) is provided in the region closest to the gate GT1 (corresponding to the “first gate” of the present disclosure) among the plurality of gates 245. That is, the distance from the sensor SN1 to the gate GT1 is shorter than the distance from the sensor SN1 to each of the other gates GT2, GT3, and GT4. In other words, the sensor 25 closest to the gate GT1 is the sensor SN1.

Similarly, the sensors SN2, SN3, and SN4 (all corresponding to the "second sensor" of the present disclosure) are the gates GT2, GT3, and GT4 (all of which correspond to the "second sensor" of the present disclosure) among the plurality of gates 245. It is provided in the area closest to).

The first distance d1 through which the molding material passes from the gate GT1 to the sensor SN1 is equal to the second distance d2 through which the molding material passes from the gate GT2 to the sensor SN2 (d1 = d2). Further, the first distance d1 is equal to the third distance d3 and the fourth distance d4 through which the molding material passes from the gate GT3 to the sensor SN3 and from the gate GT4 to the sensor SN4, respectively (d1 = d3 = d4). That is, the distances from the plurality of sensors 25 to the nearest gate 245 are equal.

See FIG. The temperature sensor 26 is built in the mold 242 and detects the temperature of the mold 242. The temperature sensor 26 outputs a temperature-related detection signal to the control unit 271. The temperature sensor 26 may be installed in the mold 241. Further, the temperature sensor 26 may be installed in the cylinder 222. That is, the temperature sensor 26 may be able to directly or indirectly detect the temperature of the molding material supplied into the cavity C1.

See FIG. The injection unit 22 includes a hopper 221, a cylinder 222, a screw 223, a ball screw 225, a motor 226, a pressurization sensor 227, a movement amount sensor 228, and a heater 229. The hopper 221 is connected to the cylinder 222 and supplies the molding material into the cylinder 222. The cylinder 222 is a member having a hollow cylindrical shape extending in the axial direction. The diameter of the one-sided end portion of the cylinder 222 in the axial direction becomes narrower as it approaches the one-sided end in the radial direction, and the nozzle 224 is provided at the one-sided end in the axial direction of the cylinder 222. The nozzle 224 is connected to the spool 243 of the mold 242.

The screw 223 is inserted into the cylinder 222 from the other end of the cylinder 222 in the axial direction. A ball screw 225 is connected to the other side of the screw 223 in the axial direction, and a motor 226 is connected to the other side of the ball screw 225 in the axial direction. When the motor 226 is rotated by the operation command of the control unit 271, the ball screw 225 moves in the axial direction. Along with this, the screw 223 also moves in the axial direction. At this time, the screw 223 rotates in the circumferential direction with the axial direction as the central axis.

The pressurization sensor 227 detects the pressure applied to the motor 226 from the ball screw 225 (the reaction force of the pushing force of the screw 223). That is, the pressurization sensor 227 detects the pressure that the screw 223 receives from the molding material. The pressurization sensor 227 outputs a detection signal regarding pressure to the control unit 271. The pressurization sensor 227 may be installed at another position as long as it can detect the pushing force of the screw 223.

The movement amount sensor 228 detects the movement amount of the ball screw 225 in the axial direction. The movement amount sensor 228 outputs a detection signal regarding the movement amount to the control unit 271. The movement amount sensor 228 may be installed at another position as long as it can detect the movement amount of the ball screw 225 in the axial direction.

The heater 229 is, for example, a resistance heating heater in which a resistance wire is wound in a coil shape. The heater 229 heats the inside of the cylinder 222 by the resistance heat when a current is passed through the resistance line according to the operation command of the control unit 271.

<Manufacturing method using molding equipment>
A method for manufacturing a molded product by the molding apparatus 20 will be described with reference to FIGS. 2 and 3 as appropriate. As for the method of manufacturing the molded product by the molding apparatus 20, the pre-process ST1, the mold clamping process ST2, the filling process ST3, the pressure holding process ST4, the pressure holding release process ST5, and the mold release process ST6 are performed in this order. Will be executed. In the present embodiment, the molded product is a resin cage used for rolling bearings. However, this is an example of a molded product, and the molded product molded by the molding apparatus according to the present invention may be a molded product having another shape and use.

See FIG. First, the pre-process ST1 is executed. In the previous step ST1, the screw 223 is rotated by the motor 226, and the pellets of the molding material are supplied from the hopper 221 into the cylinder 222 while the inside of the cylinder 222 is heated by the heater 229. The pellets of the molding material are melted in the cylinder 222 by the frictional heat accompanying the rotation of the screw 223 and the heating by the heater 229, and become the molding material in a molten state (melting step).

Next, by moving the screw 223 to the other side in the axial direction while rotating, a predetermined amount of molding material is stored in the cylinder 222 on the one side in the axial direction from the screw 223 (measurement step). As a result, the previous step ST1 is completed.

Next, when the mold clamping step ST2 is started, in the molding apparatus 20 in the state of FIG. 2, the ball screw 234 is moved to the other side in the axial direction by the operation command of the control unit 271, and the mold is as shown in FIG. The 241 is brought into contact with the mold 242. In the state where the mold 241 and the mold 242 are combined in this way, the ball screw 234 further presses the mold 241 to the mold 242 toward the other side in the axial direction by a predetermined mold tightening force. That is, the plurality of molds 241 and 242 are tightened. As a result, an annular cavity C1 is formed between the plurality of molds 241 and 242. As a result, the mold clamping step ST2 is completed.

Here, the mold clamping force is one of the molding conditions, and is determined according to other molding conditions such as the shape of the molds 241 and 242. The mold clamping force is detected by the mold clamping force sensor 236.

Subsequently, when the filling step ST3 is started, the ball screw 225 moves to one side in the axial direction while maintaining the above-mentioned mold clamping force. As a result, the screw 223 pushes the forming material to one side in the axial direction, and the forming material in a molten state is transferred from the nozzle 224 of the cylinder 222 to the cavity C1 via the spool 243, the plurality of runners 244, and the plurality of gates 245. Injected (filling operation).

When the cavity C1 is filled with the molding material, the filling step ST3 is completed. In the filling step ST3, the molten molding material is supplied to the cavity C1 while gradually solidifying from the vicinity of the surface of the molds 241 and 242.

Subsequently, when the pressure holding step ST4 is started, the screw 223 further pushes the molding material to one side in the axial direction, and the molding material is further ejected from the nozzle 224 of the cylinder 222 to the cavity C1. As a result, a predetermined pressure (for example, tens to hundreds of MPa) is applied to the molding material filled in the cavity C1. Then, by holding this state for a predetermined time, the screw 223 continues to apply a predetermined pressure to the molding material for a predetermined time (for example, several seconds) (pressure holding operation). The pressure (pressurization) at which the screw 223 pushes the molding material into the cavity C1 is detected by the pressurization sensor 227.

Subsequently, when the pressure holding release step ST5 is started, the screw 223 moves to the other side in the axial direction to release the pressure holding of the molding material (pressure holding release operation). When a predetermined time elapses after the holding pressure releasing operation and the pressure of the molding material in the cavity C1 becomes equal to or less than a predetermined value, the holding pressure releasing step ST5 ends. After that, when the mold release step ST6 is started, the mold portion 24 is cooled, so that the molding material in the cavity C1 is solidified and a molded product is formed. Then, the ball screw 234 moves to one side in the axial direction, and the mold 241 separates from the mold 242, so that the molded product is taken out. The cooling of the mold portion 24 may be started at the same time as the pressure holding release step ST5.

<About gate clogging>
FIG. 6 is a diagram illustrating clogging of the gate 245 detected in the present embodiment. FIG. 6 shows a state in which the gate GT1 is clogged with a foreign substance Fm1 as an example of clogging. FIG. 6A is a diagram showing a state in the middle of the filling step ST3, and FIG. 6B is a diagram showing a state in the middle of the pressure holding step ST4. 6 (a) and 6 (b) both show the same cross section as in FIG.

As shown in FIG. 6A, when the filling step ST3 is executed, the molding material L1 flows into the cavity C1 via the spool 243, the runner 244, and the gate 245. Here, since the gate GT1 is clogged with the foreign matter Fm1, the molding material L1 does not flow from the gate GT1 into the cavity C1, but the molding material L1 flows from the other gates GT2, GT3, and GT4 into the cavity C1.

Conventionally, the mold portion 24 is not provided with the sensor 25, and the pressure of the molding material L1 is monitored by the pressurization sensor 227 provided in the injection portion 22. For example, when all the gates 245 are clogged, the pressure rise of the pressurization sensor 227 in the filling step ST3 is clearly faster than in the normal state (when there is no clogging). It is possible to detect 245 clogging.

However, if some gates 245 (particularly only half or less of the gates 245 of the plurality of gates 245) are clogged, the cavities from the other unclogging gates GT2, GT3, GT4, as shown in FIG. 6 (a). Since the molding material L1 flows into C1, the pressure detected by the pressurization sensor 227 is almost the same as the normal pressure. Therefore, it is not possible to detect the clogging of a part of the gate 245 based on the pressure of the pressurization sensor 227.

Further, as shown in FIG. 6B, since the molding material L1 finally flowing from the other gates GT2, GT3, and GT4 flows into the cavity C1 near the gate GT1, the cavity C1 is filled with the molding material L1. As a result, even if an abnormality occurs in which a part of the gate 245 is clogged, a molded product will be completed for the time being.

However, since the molding material L1 does not flow into the cavity C1 near the clogged gate GT1 as usual, the dimensions (for example, roundness), weight, and quality (for example, strength) of the molded product may deviate from the design range. be. For example, a weld region may be formed in an unintended region (a region other than the designed region), and the strength of the molded product may decrease. Molded products that are out of the design range are defective. Further, since the density of the molding material L1 is low in the vicinity of the gate GT1, there is a possibility that abnormalities such as sink marks and voids may occur in the molded product.

Conventionally, such an abnormality of a molded product has been detected by measuring the dimensions, appearance, weight, etc. of the molded product. In the molding apparatus 20, molding of the molded product is continuously performed. Therefore, in the conventional detection method, an abnormality is detected between the molding of the molded product and the measurement of the molded product outside the molding apparatus 20 until the abnormality is detected. Molded products will continue to be molded in vain. Therefore, it is important to have a technique in which the molding device 20 senses the clogging of a part of the gate 245 and can determine the abnormality immediately after molding without measuring the molded product.

Therefore, the inventors have provided a plurality of sensors 25 at positions in the cavity C1 corresponding to each of the plurality of gates 245, and some of them are based on the difference in the time series data of the pressure detected by the plurality of sensors 25. The idea was to detect a blockage in the gate 245. Hereinafter, differences in time-series data due to clogging of some gates 245 will be described.

<Differences in time series data due to clogging of some gates>
FIG. 7 is an example of a graph showing time-series data of pressure detected by a plurality of sensors 25. In FIG. 7, the vertical axis is pressure and the horizontal axis is time.
In the present embodiment, the origin of time (t = 0) is the time when the control unit 271 commands the motor 226 to move the screw 223 to the other side in the axial direction. That is, it is the time when the injection unit 22 starts supplying the molding material to the mold unit 24 in the filling step ST3. The origin of time (t = 0) is not limited to this, and may be the time when the pressurization sensor 227 detects a predetermined pressure, or the time when the movement amount sensor 228 detects a predetermined movement amount. You may.

The graph line F11 in FIG. 7 shows the time series data of the pressure detected by the sensor SN1 (corresponding to the “first time series data” of the present disclosure). Similarly, the graph line F12 is the sensor SN2, the graph line F13 is the sensor SN3, and the graph line F14 is the time series data of the pressure detected by the sensor SN4 (all correspond to the "second time series data" of the present disclosure). Is shown. Further, the graph line AvF1 is time series data of the average value of the pressures of the four graph lines F11 to F14. The graph lines F11 to F14 may be plots of the pressures detected by the sensors SN1 to SN4 at predetermined time intervals as they are, or may be a moving average of the pressures detected at the predetermined time intervals (for example, 5). The point moving average) may be plotted.

Although the positions where the graph lines rise and the peak values are different, the overall tendencies of the graph lines F11 to F14 are the same. When the filling step ST3 is started, the pressures on the graph lines F11 to F14 rise and then monotonically increase. This is because the molding material L1 is continuously supplied from the injection unit 22, and the molding material L1 gradually and strongly presses the sensors SN1 to SN4.

The monotonous increase of the graph lines F11 to F14 continues until the middle of the pressure holding step ST4. The pressure on the graph lines F11 to F14 reaches a maximum value in the middle of the holding pressure step ST4, and then decreases monotonically until the end of the holding pressure releasing step ST5. The monotonous increase halfway through the pressure holding step ST4 is due to the additional pressure being applied from the injection portion 22 to the molding material L1 filled in the cavity C1. Further, the monotonous decrease in pressure starting from the middle of the pressure holding step ST4 is sealed by the molding material in which the gate 245 has cooled and hardened, the application of pressure from the injection portion 22 to the cavity C1 is stopped, and the pressure in the cavity C1 is stopped. This is due to the fact that the molding material L1 cools and shrinks.

Next, attention is paid to the individual tendency of the graph lines F11 to F14. First, the graph line F11 of the sensor SN1 corresponding to the clogged gate GT1 rises later than the other graph lines F12 to F14. As shown in FIG. 6, since the molding material L1 wraps around the positions of the sensors SN1 from the gates GT2 to GT4, the time when the molding material L1 reaches the sensor SN1 is the time when it reaches the other sensors SN2 to SN4. Due to being slower than.

Further, for the same reason, the peak time point Xt1 at which the pressure of the graph line F11 becomes the maximum value Pt1 is the peak time point Xt2 to Xt4 of the other graph lines F12 to F14 (in FIG. 7, the pressure of the graph line F13 is the maximum value). It is later than the peak time point Xt3 which becomes Pt3).

Here, the time point at which the pressure rises can be appropriately defined, but in the present embodiment, the time point at which the predetermined pressure Ps1 is reached is defined as the time point at which the pressure rises. The predetermined pressure Ps1 is, for example, a value determined based on the average value AvPt of the maximum value of the pressures of the graph lines F11 to F14, and more specifically, a predetermined% (for example, 10 to 20%) of the average value AvPt. ). In each of the graph lines F11 to F14, the time when the slope of the pressure reaches a predetermined value may be set as the respective rising time.

As shown in FIG. 7, the rising point Xs1 of the graph line F11 is larger than the rising time points Xs2 to Xs4 of the other graph lines F12 to F14 (in FIG. 7, the rising time point Xs3 of the graph line F13 is typically shown). slow. Further, the rising point Xs1 is later than the rising time AXs of the graph line AvF1.

The maximum pressure value Pt1 of the graph line F11 is the maximum value Pt2 to Pt4 of the pressure of the other graph lines F12 to F14 (in FIG. 7, the maximum value Pt3 of the pressure of the graph line F13 is typically shown) and the graph line AvF1. It is lower than the maximum value of AvPt. This is because the molding material L1 that has flowed into the cavity C1 shrinks as it is cooled by the mold portion 24. Since the gate GT1 is clogged, the time required for the molding material L1 to reach the sensor SN1 is longer than the time required for the molding material L1 to reach the other sensors SN2 to SN4, and the molding material L1 is molded by that time. The shrinkage of the material L1 will proceed. Since the force with which the molding material L1 pushes the sensors SN1 is weaker than the force with which the other sensors SN2 to SN4 are pushed by this shrinkage, the maximum pressure value Pt1 of the graph line F11 is the maximum of the other graph lines F12 to F14. It will be lower than the value.

From the above, when any of the following tendencies (1) to (3) appears, it is highly possible that the gate GT1 is clogged.
(1) The rising point Xs1 of the graph line F11 is behind the rising time Xs2 to Xs4 of the other graph lines F12 to F14. (2) The peak time point Xt1 of the graph line F11 is the other graph lines F12 to F14. (3) The maximum pressure value Pt1 on the graph line F11 is lower than the maximum pressure values Pt2 to Pt4 on the other graph lines F12 to F14.

As described above, by paying attention to the difference in the plurality of time series data detected by the plurality of sensors 25 corresponding to the plurality of gates 245, it is possible to detect that some of the gates 245 are clogged. 6 and 7 explain an example in which the gate GT1 is clogged, but the same tendency appears when any of the other gates GT2 to GT4 is clogged. For example, when only the gate GT2 is clogged, the rising time point Xs2 and the peak time point Xt2 of the graph line F12 become relatively slow, and the maximum value Pt2 of the pressure of the graph line F12 becomes relatively low.

Here, due to noise or the like, the pressure on the graph line F11 may suddenly exceed the predetermined pressure Ps1 at a time earlier than the rising time points Xs2 to Xs4. In this case, the rising point Xs1 of the graph line F11 becomes earlier than the rising time points Xs2 to Xs4. Similarly, the maximum value Pt1 may suddenly exceed the other maximum values Pt2 to Pt4 due to the influence of noise or the like. In this case, if the comparison is performed based only on the rising point or the maximum value, it is not possible to correctly detect the clogging of some gates 245.

Therefore, in order to reduce the influence of noise and the like, the time integral values of the graph lines F11 to F14 may be compared instead of (or in addition to) the comparison of the rising time point and the maximum value. As the time integral value, for example, the integral value IV1 which is the time integral from the rising time point Xs1 to the peak time point Xt1 of the pressure on the graph line F11 is calculated. Similarly, for the graph lines F12 to F14, the integrated values IV2 to IV4, which are time integrals from the rising time points Xs2 to Xs4 to the peak time points Xt2 to Xt4, are calculated. When the gate GT1 is clogged, the integrated value IV1 is smaller than the integrated values IV2 to IV4. Further, the integrated value IV1 is smaller than the average value AIV of the integrated values IV1 to IV4.

<Difference in time series data when the sensor detects the temperature>
FIG. 8 is an example of a graph showing time-series data of the temperature detected by the plurality of sensors 25. In FIG. 8, the vertical axis is temperature and the horizontal axis is time.

Here, FIG. 7 explains the difference in the time series data when the plurality of sensors 25 detect the pressure. However, the plurality of sensors 25 may each detect the temperature. When detecting the pressure, the surface of the sensor 25 (or an indirect member that presses the surface of the sensor 25) needs to face the cavity C1 and come into contact with the molding material L1. Therefore, if there is no space in the cavity C1 for exposing the sensor 25, it is difficult to provide the sensor 25 for detecting the pressure.

On the other hand, if the sensor 25 detects the temperature, even if the sensor 25 is not exposed in the cavity C1 (that is, even if it is buried in the mold 24 like the temperature sensor 26). The temperature of the molding material L1 can be detected by heat conduction. Then, even when the sensor 25 detects the temperature, it is possible to detect the clogging of a part of the gate 245 based on the difference in the time series data.

The graph line F21 in FIG. 8 shows the time series data of the temperature detected by the sensor SN1 (corresponding to the “first time series data” of the present disclosure). Similarly, the graph line F22 is the sensor SN2, the graph line F23 is the sensor SN3, and the graph line F24 is the time series data of the temperature detected by the sensor SN4 (all correspond to the "second time series data" of the present disclosure). Is shown. Further, the graph line AvF2 is time series data of the average value of the temperatures of the four graph lines F21 to F24. The graph lines F21 to F24 may be plots of the temperatures detected by the sensors SN1 to SN4 at predetermined time intervals as they are, or may be a moving average of the temperatures detected at each predetermined time (for example, 5 points). It may be a plot of moving average).

Although the positions where the graph lines rise and the peak values are different, the overall tendencies of the graph lines F21 to F24 are the same. First, the temperature of the graph lines F21 to F24 rises and then monotonically increases. This is because the molding material L1 supplied to the cavity C1 heats the sensors SN1 to SN4. The graph lines F21 to F24 reach their respective maximum values and then decrease monotonically. This is because the supply of the molding material L1 from the injection portion 22 to the cavity C1 is stopped and the molding material L1 in the cavity C1 cools down.

Next, attention is paid to the individual tendency of the graph lines F21 to F24. First, the rising time point Xs1 (time point when the predetermined temperature Ts1 is reached) and the peak time point Xt1 of the graph lines F21 are later than the rising time points Xs2 to Xs4 and the peak time points Xt2 to Xt4 of the other graph lines F22 to F24. Further, the rising point Xs1 is later than the rising time AXs of the graph line AvF2. This is because the gate GT1 is clogged, similar to the reason why the rise of the graph line F11 in FIG. 7 is delayed, so that the time when the molding material L1 reaches the sensor SN1 is higher than the time when it reaches the other sensors SN2 to SN4. Due to being late.

From the above, based on the difference in the time series data of the temperature, it is highly possible that the gate GT1 is clogged when the following tendency (1) or (2) appears.
(1) The rising point Xs1 of the graph line F21 is behind the rising time Xs2 to Xs4 of the other graph lines F22 to F24. (2) The peak time point Xt1 of the graph line F21 is the other graph lines F22 to F24. Being behind the peak time points Xt2 to Xt4

In this way, even when the sensor 25 detects the temperature, if attention is paid to the difference in the plurality of time series data detected by the plurality of sensors 25 corresponding to the plurality of gates 245, some of the gates 245 are clogged. Can be detected.

As described above, as a result of diligent research, when pressure or temperature is detected by a plurality of sensors 25 corresponding to a plurality of gates 245, if some of the gates 245 are clogged, the said case is concerned. Discovered that the time-series data detected by the sensor 25 (eg, sensor SN1) corresponding to the clogged gate 245 (eg, gate GT1) is different from the time-series data detected by other sensors 25 (eg, sensors SN2 to SN4). did.

In addition, the inventors have described that the differences in the time-series data are Xs1 to Xs4 at the rising point of pressure or temperature, Xt1 to Xt4 at the peak time at which the pressure or temperature reaches the maximum value, Pt1 to Pt4 at the maximum value of pressure, and pressure. It was found that it appears particularly prominently in the integrated values IV1 to IV4, which are time-integrated from the rise time to the peak time. Therefore, the inventors have conceived an invention in which these values are quantified as evaluation values and a clogging of a part of the gate 245 is detected based on the evaluation values.

The evaluation value is acquired for each of the plurality of sensors 25, for example. In the present embodiment, the four evaluation values R1 to R4 are calculated based on the time-series data detected by the four sensors 25, respectively. The evaluation value R1 (corresponding to the “first evaluation value” of the present disclosure) is a value calculated based on the time series data (graph line F11 or F21) of the pressure or temperature detected by the sensor SN1, and is, for example, rising. It is at least one of the time point Xs1, the peak time point Xt1, the maximum value Pt1, and the integrated value IV1.

The evaluation values R2 to R4 (both correspond to the "second evaluation value" of the present disclosure) are values calculated based on the time-series data of the pressure or temperature detected by the sensors SN2 to SN4. The evaluation values R1 to R4 are corresponding values. For example, when the evaluation value R1 includes the rise time, the evaluation values R2 to R4 also include the rise time. When the evaluation value R1 includes the peak time point, the evaluation values R2 to R4 also include the peak time point.

In the abnormality detection system 10 according to the present embodiment, the learning device 30 generates a trained model Tm1 in which the correlation between the plurality of evaluation values R1 to R4 and the clogging of the gate 245 is learned, and the learning device 40 has trained. The detection information for detecting the clogging of the gate 245 is acquired based on the model Tm1. Hereinafter, the learning device 30 and the abnormality detection device 40 will be described.

<Explanation of learning device>
FIG. 9 is a block diagram showing a functional configuration of the learning device 30 according to the present embodiment. The learning device 30 has a training data acquisition unit 31, a learning calculation unit 32, a molding information storage unit 33, and a learned model storage unit 34. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as an HDD.

Various molding information is stored in the molding information storage unit 33. The molding information is, for example, table-type information in which various types of first information and second information are associated with each other. For example, when the first information is a mold type, the second information includes various dimensions of the mold and the volume of the cavity C1. When the first information is the type or lot number of the molding material, the second information includes the physical properties (viscosity, water content, etc.) of the molding material.

The training data acquisition unit 31 acquires information on the training data from each unit of the abnormality detection system 10. The training data includes gate information, molding information corresponding to the gate information, and a plurality of evaluation values R1 to R4. The training data is acquired based on the data detected in each part (for example, the sensor 25) of the abnormality detection system 10 when, for example, the molded product to be learned is molded.

The gate information is a value relating to the clogging of the gate 245, and more specifically, a value obtained by quantifying how the gate 245 is clogged. For example, "0" is the state where all the gates 245 are not clogged, "1" is the state where only the gate GT1 is clogged, "2" is the state where only the gate GT2 is clogged, and only the gate GT3 is clogged. Is quantified as "3", and the state where only the gate GT4 is clogged is quantified as "4".

The training data acquisition unit 31 acquires gate information by input of an operator. For example, for learning of the learning device 30, the operator artificially clogs a predetermined gate 245, and inputs to the input unit 50 how the gate 245 is clogged. As a result, the training data acquisition unit 31 acquires the gate information.

Further, the training data acquisition unit 31 acquires molding information based on the information input by the operator to the input unit 50 and the molding information storage unit 33. For example, when molding a molded product to be learned, an operator inputs a lot number of a molding material related to the molded product. The training data acquisition unit 31 acquires molding information regarding the physical properties (for example, viscosity) of the molding material corresponding to the lot number from the molding information storage unit 33.

The molding apparatus 20 uses the unclogging gate 245 and the artificially predetermined clogged gate 245 to mold the molded product a plurality of times. Then, the training data acquisition unit 31 acquires a plurality of evaluation values R1 to R4 to which gate information is given as teacher data based on the time-series data of the pressure or temperature detected by the sensor 25. For example, "0" is given as gate information to a plurality of evaluation values R1 to R4 acquired when a molded product is molded using a gate 245 without clogging.

Further, the training data acquisition unit 31 further acquires a plurality of environmental values as training data. The environment value is, for example, a value related to the temperature detected by the temperature sensor 26 when the molded product to be learned is molded. The environmental values are the periphery of the molding device 20 and the molding device detected by the pressure sensor 227, the mold clamping force sensor 236 and other sensors (for example, humidity sensor) not shown when the molded product to be learned is molded. It may further include 20 internal environmental values.

For example, each time the training data acquisition unit 31 molds one molded product to be learned, one set of training data (gate information acquired at the time of molding the molded product, molding information, and a plurality of evaluation values R1 to R4 and a set of environment values) are acquired. The training data acquisition unit 31 acquires a plurality of sets of training data for a predetermined number of times by molding the molded product to be learned a predetermined number of times.

The learning calculation unit 32 models the correlation between molding information, a plurality of evaluation values R1 to R4, and environment values by performing supervised machine learning based on a plurality of sets of training data. Generate the trained model Tm1. In this embodiment, a support vector machine (SVM) is used as the machine learning model, but other models may be used. For example, a convolutional neural network (CCN) may be used, or a regression tree model which is a model for grouping data may be used.

Specifically, when the trained model Tm1 is generated, molding information, a plurality of evaluation values R1 to R4, and environmental values (these information are collectively referred to as "input information") are used as explanatory variables, and the purpose is gate information. By making it a variable, the correlation between the input information and the gate information is modeled. That is, when the input information is input, the learning calculation unit 32 generates a trained model Tm1 that constitutes an intermediate layer for outputting the gate information corresponding to the input information from the output layer.

Here, the tendency of the trained model Tm1 will be described. Consider a case where a plurality of evaluation values R1 to R4 are Xs1 to Xs4 at the time of rise of pressure or temperature, respectively. In this case, when the value of the evaluation value R1 input to the input layer is later than the other evaluation values R2 to R4 (the evaluation value R1 is larger than the other evaluation values R2 to R4), the output layer is used. The accuracy of the gate information (“1” in the above example) indicating that the gate GT1 is clogged is high. Further, for example, when the value of the evaluation value R4 input to the input layer is later than the other evaluation values R1 to R3, the gate information indicating that the gate GT4 is clogged in the output layer (in the above example). , "4") is more accurate.

Further, when the plurality of evaluation values R1 to R4 are the maximum values of the pressure respectively and the value of the evaluation value R1 input to the input layer is smaller than the other evaluation values R2 to R4, the gate GT1 is used in the output layer. The accuracy of the gate information (“1” in the above example) indicating that is clogged is high.

When generating the trained model Tm1, the input information may include a plurality of evaluation values R1 to R4, and may not include molding information and environmental values. Here, the higher the water content (molding information) of the molding material L1 or the higher the temperature (environmental value) detected by the temperature sensor 26, the lower the viscosity of the molding material L1 and the easier it is for the molding material L1 to flow. Become. Then, when the molding material L1 becomes easy to flow, the molding material L1 is filled in the entire cavity C1 earlier, so that the difference between the rise time Xs1 (evaluation value R1) and the rise time Xs3 (evaluation value R3) shown in FIG. Becomes smaller.

As described above, the plurality of evaluation values R1 to R4 also change depending on the molding information and the environmental value. Therefore, when the variation in the environmental value and the molding information is large, or when the variation has a large influence on the plurality of evaluation values R1 to R4, the gate information corresponding to the input information can be predicted more accurately. When generating the trained model Tm1, it is preferable to include the environment value and the molding information in the input information.

The trained model Tm1 generated by the learning calculation unit 32 is stored in the trained model storage unit 34. When new information is input to the learning device 30 and new training data is acquired by the training data acquisition unit 31, the trained model Tm1 stored in the trained model storage unit 34 responds to the content of the training data. Will be updated as appropriate. Further, the trained model Tm1 is transmitted from the learning device 30 to the abnormality detection device 40 described later, and is also stored in the trained model storage unit 45 of the abnormality detection device 40.

<How to generate a trained model>
Next, a method of generating the trained model Tm1 by the learning device 30 will be described. The method of generating the trained model Tm1 includes a training data acquisition process and a learning calculation process. These steps are realized by the computer device constituting the learning device 30 executing a predetermined program.

First, when the training data acquisition process is started, the training data acquisition unit 31 acquires a plurality of sets of training data. For example, the worker artificially clogs the predetermined gate 245 and inputs the clogged method of the gate 245 to the input unit 50, so that the training data acquisition unit 31 acquires the gate information. Then, a molded product is molded using the gate 245, and a plurality of evaluation values R1 to R4 (for example, for example) corresponding to the gate information are formed based on the time series data of the pressure or temperature detected by the sensor 25 at that time. Acquire the rising time points Xs1 to Xs4). This completes the training data acquisition process.

Next, when the learning calculation process is started, the learning calculation unit 32 learns the correspondence between the input information and the gate information based on a plurality of sets of training data, and generates a trained model Tm1. The trained model Tm1 is stored in the trained model storage units 34 and 45. With the above, the learning calculation process is completed.

<Explanation of abnormality detection device>
FIG. 10 is a block diagram showing a functional configuration of the abnormality detection device 40 according to the present embodiment. The abnormality detection device 40 includes a data acquisition unit 41, an abnormality detection unit 42, an output unit 43, a molding information storage unit 44, and a learned model storage unit 45. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as an HDD. The arithmetic unit executes the data acquisition process and the abnormality detection process described later based on the program stored in the storage unit.

Similar to the molding information storage unit 33, the molding information storage unit 44 stores table-type molding information in which various first information and the second information are associated with each other. The learned model Tm1 generated by the learning device 30 is stored in the learned model storage unit 45.

The molded information storage unit 44 and the trained model storage unit 45 may be realized by the same storage area as the molded information storage unit 33 and the trained model storage unit 34 of the learning device 30 among the computer devices, or may be stored separately. It may be realized by the area. That is, the learning device 30 and the abnormality detecting device 40 may be configured to share the same molding information storage unit 33 and the learned model storage unit 34, or the learning device 30 and the abnormality detecting device 40 may be independently molded. It may be configured to have information storage units 33, 44 and trained model storage units 34, 45.

The data acquisition unit 41 executes a data acquisition process for acquiring information for performing abnormality detection from each unit of the abnormality detection system 10. Information for performing abnormality detection includes, for example, a set of environmental values, molding information, and a plurality of evaluation values R1 to R4 (hereinafter, these information) acquired by the molding apparatus 20 when the molded product to be detected is molded. Collectively referred to as "detection data set").

The abnormality detection unit 42 inputs the detection data set acquired by the data acquisition unit 41 into the trained model Tm1. Based on these inputs, the trained model Tm1 outputs detection information D1 for detecting clogging of the gate 245.

The detection information D1 is information including the respective accuracy of the gate information. Specifically, the detection information D1 includes the first to fourth accuracy, which is the accuracy at which the gates GT1 to GT4 are clogged, and the fifth accuracy, which is the accuracy at which all the gates 245 are not clogged. include. As an example, when the detection data set is input to the trained model Tm1, the trained model Tm1 has a first accuracy of 60%, a second accuracy of 10%, a third accuracy of 10%, and a fourth accuracy of 10%. , The detection information D1 having a fifth accuracy of 10% is output.

The output unit 43 determines how the gate 245 is clogged (that is, the gate 245 having a high possibility of clogging among the plurality of gates 245) based on the detection information D1 acquired by the abnormality detection unit 42. For example, the output unit 43 acquires the most accurate clogging method among the detection information D1 as a determination result of the clogging method of the gate 245. For example, when the first accuracy is the highest among the first to fifth accuracy described above, the output unit acquires that "the gate GT1 is clogged" as a determination result.

The output unit 43 outputs the determination result to the display unit 60 and the control unit 271. The determination result is displayed on the display unit 60. When it is determined that any of the plurality of gates 245 is clogged, the determination result is displayed in a highlighted color such as red on the display of the display unit 60, and an alert is issued on the speaker. May be good.

Further, in this case, the molding apparatus 20 including the gate 245 for which the clogging is determined may be stopped by the operation command of the control unit 271 in a state where the mold unit 24 is open. In this case, the operator inspects the mold unit 24 based on the alert or the like by the display unit 60, and cleans or replaces the mold unit 24 as necessary.

In this embodiment, the detection information D1 obtained by the abnormality detection unit 42 may be displayed as it is on the display unit 60 without providing the output unit 43. In this case, the operator may determine whether or not the gate 245 is clogged based on the detection information D1 displayed on the display unit 60.

<Abnormality detection method using anomaly detection device>
Next, an abnormality detection method by the abnormality detection device 40 will be described. The abnormality detection method includes a data acquisition step and an abnormality detection step. These steps are realized by the computer device constituting the abnormality detection device 40 executing a predetermined program.

When the data acquisition process is started, the data acquisition unit 41 acquires a detection data set acquired when the molded product to be detected is molded. This completes the data acquisition process. Next, when the abnormality detection step is started, the abnormality detection unit 42 acquires the detection information D1 by inputting the detection data set into the trained model Tm1. Next, the output unit 43 acquires a determination result based on the detection information D1. Finally, the output unit 43 outputs the determination result to the display unit 60 and the control unit 271. As a result, the abnormality detection step is completed.

<Action / effect of anomaly detection system>
The abnormality detection system 10 according to the present embodiment acquires a plurality of evaluation values R1 to R4 based on the time series data of the pressure detected by the sensor 25, and uses the plurality of evaluation values R1 to R4 as the trained model Tm1. By inputting, the detection information D1 for detecting the clogging of a part of the gate 245 is acquired.

More specifically, the abnormality detection system 10 includes a molding device 20 and an abnormality detection device 40. The molding apparatus 20 is a molding material in a molten state into the cavity C1 via the mold portion 24 in which the plurality of gates 245 and the cavities C1 connected to the plurality of gates 245 are formed therein, and the plurality of gates 245. A first sensor (for example, a first sensor (for example) provided in the region closest to the first gate (for example, gate GT1) of the plurality of gates 245 and the injection unit 22 for filling L1 and detecting the pressure or temperature of the molding material L1 in the cavity C1. For example, the pressure of the molding material L1 in the cavity C1 provided in the region closest to the sensor SN1) and the second gate (for example, at least one of the gates GT2 to GT4) different from the first gate among the plurality of gates 245. Alternatively, it has a second sensor (for example, at least one of the sensors SN2 to SN4) that detects the temperature.

The abnormality detection device 40 includes a first evaluation value (for example, evaluation value R1) calculated based on the first time series data (for example, graph line F11 or F21) of the pressure or temperature detected by the first sensor. A second evaluation value (eg, evaluation value R2 to) calculated based on a second time series data of pressure or temperature detected by the second sensor (for example, at least one of graph lines F12 to F14 or F22 to F24). It has at least one of R4), a data acquisition unit 41 for acquiring data, and an abnormality detection unit 42 for detecting clogging of the first gate or the second gate based on the first evaluation value and the second evaluation value. ..

When the first gate or the second gate is clogged, a difference occurs between the first time series data and the second time series data. To make it easy to compare this difference, the first time-series data is quantified as the first evaluation value, the second time-series data is quantified as the second evaluation value, and based on these first evaluation value and second evaluation value. Detects a blockage in the first gate or the second gate. As a result, it is possible to detect the clogging of some gates 245, which could not be detected in the past.

In the present embodiment, the first distance (for example, the first distance d1) through which the molding material L1 passes from the first gate to the first sensor is the second distance (for example, the first distance d1) through which the molding material L1 passes from the second gate to the second sensor. For example, it is equal to at least one of the second to fourth distances d2 to d4). With this configuration, when the gate 245 is “not” clogged, the time points at which the molding material L1 reaches the first sensor and the second sensor (that is, the rising time and the peak time) can be aligned. As a result, it is possible to cause a delay at the rise time or a delay at the peak time only when a part of the gate 245 is clogged. As a result, the accuracy of clogging detection can be improved.

In the present embodiment, the first evaluation value is the first rise time point (for example, time point Xs1) of the first time series data, or the first peak time point (for example, the time point where the pressure or temperature of the first time series data becomes the maximum value). , Time point Xt1), and the second evaluation value is the maximum value at the second rise time point of the second time series data (for example, at least one of the time points Xs2 to Xs4), or the pressure or temperature of the second time series data. The abnormality detection unit 42 includes the second peak time point (for example, at least one of the time points Xt2 to Xt4), and the abnormality detection unit 42 is in the case where the first rise point time is later than the second rise point time, or the first peak time point is the first. When it is delayed from the time of two peaks, the clogging of the first gate is detected.

Further, in the present embodiment, the first sensor and the second sensor detect the pressure of the molding material L1, and the first evaluation value is the first peak value (for example, the maximum value of the pressure of the first time series data). The maximum value Pt1) is included, the second evaluation value includes the second peak value (for example, at least one of the maximum values Pt2 to Pt4) which is the maximum value of the pressure of the second time series data, and the abnormality detection unit 42 includes. , When the first peak value is lower than the second peak value, the clogging of the first gate is detected.

Further, in the present embodiment, the first sensor and the second sensor detect the pressure of the molding material L1, and the first evaluation value is the pressure of the first time-series data from the first rise time to the first time-series data. The second evaluation value includes the first integral value (for example, the integral value IV1) which is the time integration up to the time of the first peak when the pressure becomes the maximum value, and the second evaluation value is the second rising point of the pressure of the second time series data. The abnormality detection unit 42 includes the second integral value (for example, at least one of the integral values IV2 to IV4) which is the time integration up to the time of the second peak when the pressure of the two time series data becomes the maximum value, and the abnormality detection unit 42 is the first integral. When the value is smaller than the second integrated value, the clogging of the first gate is detected.

As described above, in the present embodiment, the first evaluation value includes at least one of the first rise time point, the first peak time point, the first peak value, and the first integral value, and the second evaluation value is the first evaluation value. Includes at least one of a second rising time point, a second peak time point, a second peak value and a second integral value corresponding to. Then, by comparing the first rising time and the second rising time, the first peak time and the second peak time, the first peak value and the second peak value, or the first integrated value and the second integrated value, the first Of the 1st gate and the 2nd gate, it can be specified that the 1st gate is clogged. That is, according to the abnormality detection system 10, it is possible not only to detect the clogging of some gates 245 that could not be detected in the past, but also to specify which gate 245 is clogged.

Further, the abnormality detection unit 42 inputs the first evaluation value and the second evaluation value into the trained model Tm1 in which the correlation between the first evaluation value and the second evaluation value and the clogging of the plurality of gates 245 is machine-learned. By doing so, clogging of at least one of the first gate and the second gate is detected.

With such a configuration, once the trained model Tm1 is generated, it becomes possible to detect the clogging of a part of the gate 245 from the first evaluation value and the second evaluation value. By using the trained model Tm1, it is possible to more accurately detect the clogging of a part of the gate 245 even when the molding conditions vary.

<Second Embodiment>
The abnormality detection system 10 according to the first embodiment has been described above. However, the implementation of the present invention is not limited to this, and various modifications can be made. Hereinafter, the abnormality detection system 11 according to the second embodiment of the present invention will be described. In the following description, the same reference numerals are given to the parts that are not changed from the first embodiment, and the description thereof will be omitted.

FIG. 11 is a block diagram schematically showing the abnormality detection system 11 according to the second embodiment. The abnormality detection system 11 includes a plurality of molding devices 20, an abnormality detection device 40a, an input unit 50, and a display unit 60.

In the present embodiment, the abnormality detection device 40a of the abnormality detection system 11 has an absolute value of the difference between the average value AvR of the plurality of evaluation values R1 to R4 and the evaluation value R1 (that is, the absolute value DIV1 of the deviation of the evaluation value R1). However, when it exceeds a predetermined reference value REF1, it is detected that the gate GT1 is clogged. That is, the abnormality detection system 11 differs from the abnormality detection system 10 according to the first embodiment in that the abnormality is detected by comparing the absolute value DIV1 of the deviation with the reference value REF1 without using the trained model Tm1. ..

FIG. 12 is a block diagram showing a functional configuration of the abnormality detection device 40a according to the present embodiment. The abnormality detection device 40a includes a data acquisition unit 41, an abnormality detection unit 42a, an output unit 43a, a molding information storage unit 44, and a reference value storage unit 46. Each of these units is realized by a computer device having a calculation unit such as a CPU and a storage unit such as an HDD. The data acquisition unit 41 acquires a plurality of evaluation values R1 to R4 in the same manner as the data acquisition unit 41 according to the first embodiment.

The reference value REF1 is stored in the reference value storage unit 46. The reference value REF1 is, for example, a value obtained by adding a predetermined margin to the standard deviations of the plurality of evaluation values R1 to R4 acquired when none of the plurality of gates 245 is clogged.

The abnormality detection unit 42a calculates the absolute value DIV1 of the deviation of the evaluation value R1 based on the plurality of evaluation values R1 to R4 acquired by the data acquisition unit 41. Then, by comparing the absolute value DIV1 of the deviation with the reference value REF1 stored in the reference value storage unit 46, the detection information D2 for detecting the clogging of the gate GT1 is acquired. The detection information D2 is, for example, the difference (DIV1-REF1) between the absolute value DIV1 and the reference value REF1 of the deviation.

The output unit 43a determines whether or not the gate GT1 is clogged based on the detection information D2 acquired by the abnormality detection unit 42a. For example, when the detection information D2 is the above difference (DIV1-REF1), the output unit 43a receives when the detection information D2 is a positive value (that is, when the absolute value DIV1 of the deviation exceeds the reference value REF1). ), It is determined that the gate GT1 is clogged. The output unit 43a outputs the determination result to the display unit 60 and the control unit 271.

Further, when determining whether or not the gate GT2 is clogged, the abnormality detection unit 42a determines the absolute value DIV2 of the deviation of the evaluation value R2 based on the plurality of evaluation values R1 to R4 acquired by the data acquisition unit 41. By calculating and comparing the absolute value DIV2 of the stitch difference with the reference value REF1, the detection information D3 for detecting the clogging of the gate GT2 is acquired. In this way, by comparing the absolute value of the deviation of the evaluation values R1 to R4 corresponding to the gates GT1 to GT4 for which clogging is to be detected with the reference value REF1, detection for detecting clogging of each gate GT1 to GT4 is performed. Information can be obtained.

In this embodiment, the detection information obtained by the abnormality detection unit 42a may be displayed as it is on the display unit 60 without providing the output unit 43a. In this case, the operator may determine that the gate 245 is clogged based on the detection information displayed on the display unit 60.

According to the abnormality detection system 11 according to the present embodiment, clogging of a part of the gate 245 can be easily detected by comparing the plurality of evaluation values R1 to R4 with the reference value REF1.

<Others>
The embodiments disclosed as described above are exemplary in all respects and are not restrictive. That is, the abnormality detection system of the present invention is not limited to the illustrated form, and may be another form within the scope of the present invention.

10 Abnormality detection system 11 Abnormality detection system 20 Molding device 21 Bed 22 Injection part 221 Hopper 222 Cylinder 223 Screw 224 Nozzle 225 Ball screw 226 Motor 227 Pressure sensor 228 Movement amount sensor 229 Heater 23 Type tightening part 231 Fixed plate 231a Through hole 232 Movable board 232a Through hole 233 Tie bar 234 Ball screw 235 Support board 236 Mold tightening force sensor 237 Motor 24 Mold part 241 Mold 242 Mold 243 Spool 243a End part 243b End part 244 Runner 245 Gate 25 Sensor 26 Temperature sensor 27 Control panel 271 Control unit 272 Communication unit 30 Learning device 31 Training data acquisition unit 32 Learning calculation unit 33 Molding information storage unit 34 Learned model storage unit 40 Abnormality detection device 40a Abnormality detection device 41 Data acquisition unit 42 Abnormality detection unit 42a Abnormality detection unit 43 Output unit 43a Output unit 44 Molding information storage unit 45 Learned model storage unit 46 Reference value storage unit 50 Input unit 60 Display unit C1 Cavity L1 Molding material Fm1 Foreign matter GT1 to GT4 Gate SN1 to SN4 Sensor d1 First distance d2 Second distance d3 3rd distance d4 4th distance R1 to R4 Evaluation value Tm1 Trained model D1 Detection information REF1 Reference value DIV1 Absolute value of deviation

Claims (10)

An abnormality detection system including a molding device for molding a molded product and an abnormality detection device for detecting an abnormality in the molding device.
The molding apparatus is
A mold portion formed inside a plurality of gates and a cavity connected to the plurality of gates.
An injection unit that fills the cavity with the molded material in a molten state via the plurality of gates,
A first sensor provided in the region closest to the first gate among the plurality of gates and detecting the pressure or temperature of the molding material in the cavity, and
A second sensor provided in the region closest to the second gate different from the first gate among the plurality of gates and detecting the pressure or temperature of the molding material in the cavity.
Have,
The abnormality detection device is
Calculated based on the first evaluation value calculated based on the first time series data of the pressure or temperature detected by the first sensor and the second time series data of the pressure or temperature detected by the second sensor. The second evaluation value to be obtained, the data acquisition unit to acquire, and
An abnormality detection unit that detects clogging of the first gate or the second gate based on the first evaluation value and the second evaluation value.
Anomaly detection system. The abnormality according to claim 1, wherein the first distance through which the molding material passes from the first gate to the first sensor is equal to the second distance through which the molding material passes from the second gate to the second sensor. Detection system. The first evaluation value is
At the time of the first rise of the first time series data, or
Includes the first peak time point when the pressure or temperature of the first time series data becomes the maximum value.
The second evaluation value is
At the time of the second rise of the second time series data, or
Includes the second peak time point when the pressure or temperature of the second time series data becomes the maximum value.
The abnormality detecting unit is clogged with the first gate when the first rising point is delayed from the second rising point or when the first peak time is delayed from the second peak point. 2. The abnormality detection system according to claim 2. The first sensor and the second sensor detect the pressure of the molding material, and the first sensor and the second sensor detect the pressure of the molding material.
The first evaluation value includes the first peak value which is the maximum value of the pressure of the first time series data.
The second evaluation value includes a second peak value which is the maximum value of the pressure of the second time series data.
The abnormality detection unit according to any one of claims 1 to 3, wherein the abnormality detection unit detects clogging of the first gate when the first peak value is lower than the second peak value. system. The first sensor and the second sensor detect the pressure of the molding material, and the first sensor and the second sensor detect the pressure of the molding material.
The first evaluation value includes a first integral value which is a time integral from the first rising time of the pressure of the first time series data to the first peak time when the pressure of the first time series data becomes the maximum value. ,
The second evaluation value includes a second integral value which is a time integral from the second rising time of the pressure of the second time series data to the second peak time when the pressure of the second time series data becomes the maximum value. ,
The abnormality detection unit according to any one of claims 1 to 4, wherein the abnormality detection unit detects clogging of the first gate when the first integral value is smaller than the second integral value. system. The abnormality detection unit transfers the first evaluation value and the second evaluation value to a trained model in which the correlation between the first evaluation value and the second evaluation value and the clogging of the plurality of gates is machine-learned. By inputting, clogging of at least one of the first gate and the second gate is detected.
The abnormality detection system according to any one of claims 1 to 5. An abnormality detection device that detects abnormalities in a molding device that molds a molded product.
The molding apparatus is
A mold portion formed inside a plurality of gates and a cavity connected to the plurality of gates.
An injection unit that fills the cavity with the molded material in a molten state via the plurality of gates,
A first sensor provided in the region closest to the first gate among the plurality of gates and detecting the pressure or temperature of the molding material in the cavity, and
A second sensor provided in the region closest to the second gate different from the first gate among the plurality of gates and detecting the pressure or temperature of the molding material in the cavity.
Have,
The abnormality detection device is
Calculated based on the first evaluation value calculated based on the first time series data of the pressure or temperature detected by the first sensor and the second time series data of the pressure or temperature detected by the second sensor. The second evaluation value to be obtained, the data acquisition unit to acquire, and
An abnormality detection unit that detects clogging of at least one of the first gate and the second gate based on the first evaluation value and the second evaluation value.
Anomaly detection device. Molding including a mold portion in which a plurality of gates and a cavity connected to the plurality of gates are formed therein, and an injection portion for filling the cavity with a molten molding material via the plurality of gates. It is an abnormality detection method that detects an abnormality in a device.
The first evaluation calculated based on the first time series data of the pressure or temperature of the molding material in the cavity detected by the first sensor provided in the region closest to the first gate among the plurality of gates. Second time series data of the value and the pressure or temperature of the molding material in the cavity detected by the second sensor provided in the region closest to the second gate different from the first gate among the plurality of gates. The second evaluation value calculated based on, the data acquisition process to acquire, and
An abnormality detection step of detecting clogging of at least one of the first gate and the second gate based on the first evaluation value and the second evaluation value.
Anomaly detection method. Molding including a mold portion in which a plurality of gates and a cavity connected to the plurality of gates are formed therein, and an injection portion for filling the cavity with a molten molding material via the plurality of gates. It is a program for detecting abnormalities in the device.
The first evaluation calculated based on the first time series data of the pressure or temperature of the molding material in the cavity detected by the first sensor provided in the region closest to the first gate among the plurality of gates. Second time series data of the value and the pressure or temperature of the molding material in the cavity detected by the second sensor provided in the region closest to the second gate different from the first gate among the plurality of gates. The second evaluation value calculated based on, the data acquisition process to acquire, and
An abnormality detection step of detecting clogging of at least one of the first gate and the second gate based on the first evaluation value and the second evaluation value.
A program that causes a computer device to execute. Molding including a mold portion in which a plurality of gates and a cavity connected to the plurality of gates are formed therein, and an injection portion for filling the cavity with a molten molding material via the plurality of gates. A trained model for detecting device anomalies,
The explanatory variable is
The first evaluation calculated based on the first time series data of the pressure or temperature of the molding material in the cavity detected by the first sensor provided in the region closest to the first gate among the plurality of gates. Value and
Based on the second time series data of the pressure or temperature of the molding material in the cavity detected by the second sensor provided in the region closest to the second gate different from the first gate among the plurality of gates. The calculated second evaluation value and
Including
The objective variable includes a value relating to a clogged gate among the plurality of gates.
Trained model.

Images