JP6587989B2 - Measuring device, measuring method, program - Google Patents

Description

  The present invention relates to a measuring device, a measuring method, and a program for monitoring mass production such as an injection molding device.

2. Description of the Related Art A measurement system that measures a detection signal of a sensor installed in an injection molding apparatus with a measurement device is known. This measuring system uses a temperature sensor, a pressure sensor, etc. provided in an injection molding machine, a mold, or a molding peripheral device (for example, a temperature controller for cooling, a vacuuming device, etc.) to determine the behavior of a molding material such as a resin. It can be detected and output as a waveform to an information processing apparatus such as a personal computer in real time.
The measurement data can be used for various purposes such as setting of optimum molding conditions, automatic selection of defective products, quality control, and die evaluation.
In the measurement system, it is also possible to monitor the measurement value based on the detection signal of the sensor and output an alarm in response to the occurrence of an abnormality. This alarm output makes it possible to stop the injection molding apparatus and identify defective products.

  Patent Document 1 listed below discloses a technique in which a sensor provided in an injection molding apparatus detects the pressure of a resin in a cavity, and a detection signal of the sensor is sampled by an amplifier device.

JP 2008-36975 A

  Here, as the abnormality determination based on the detection signal of the sensor, for example, when the injection molding conditions are changed, such as changing a mold or a resin material used in the injection molding apparatus, it may be necessary to review the determination criteria. . In that case, it is common that an operator or the like is formulated in consideration of a sensor detection value and a finished product at the stage of trial production before mass production under new injection molding conditions.

However, it takes a considerable amount of man-hours to formulate the judgment criteria manually, which helps to increase the cost of manufacturing the product.
In order to prevent such an increase in cost, for example, a threshold value for the sensor detection value is fixedly set as a numerical value with a certain margin so as to be compatible with a wide range of injection molding conditions. However, in that case, there is a possibility that the accuracy of abnormality determination is reduced.

  Accordingly, an object of the present invention is to achieve both reduction in abnormality determination accuracy and cost reduction.

The measuring device according to the present invention includes an acquisition unit that acquires a detection value of a sensor provided in an injection molding device, and the detection value that the acquisition unit has acquired in the past in each molding cycle of the injection molding device. And generating a threshold value for each time point, adaptively updating the threshold value, a detection value acquired by the acquisition unit, and the threshold value corresponding to the acquisition time point of the detection value, based on the results the threshold generating part is compared with the adaptively updated the threshold, obtain Preparations a determining unit for determining the abnormality determination result of the injection molding conditions, a.
And the said threshold value generation part produces | generates the said threshold value for every said time based on the said detected value for every said time of each shaping | molding cycle in the threshold value production | generation cycle which is a past n shaping | molding cycle (n is a natural number of 2 or more). And obtaining the average value for each detected time point for the detected value by averaging the detected value for each time point acquired by the acquisition unit over a plurality of molding cycles, and for each time point The difference between the detected value and the average value is calculated, and the number of molding cycles of the threshold generation target cycle is adjusted based on the difference value .
Another measuring device according to the present invention includes an acquisition unit that acquires a detection value of a sensor provided in the injection molding device, and the acquisition unit acquires in the past each time point in one molding cycle of the injection molding device. Based on the detected value, a threshold value for each time point is generated, the threshold value generating unit adaptively updating the threshold value, the detection value acquired by the acquisition unit, and the detection value corresponding to the acquisition time point A determination unit that obtains an abnormality determination result of the injection molding situation based on a result of comparing the threshold value and the threshold value adaptively updated by the threshold value generation unit.
And the said threshold value generation part obtains the average value for every said time point about the said detected value by averaging the said detected value for every said time point which the said acquisition part acquired over the some shaping | molding cycle, The difference between the detected value and the average value is calculated for each time point, and the threshold update period is adjusted based on the difference value.

According to the measurement apparatus described above, the abnormality determination of the injection molding situation is performed based on the threshold value generated based on the actual detection value, so that the abnormality determination can be performed based on the determination criterion adapted to the change in the injection molding conditions. The Therefore, it is possible to suppress a decrease in abnormality determination accuracy due to a change in injection molding conditions.
In addition, since it is not necessary to search for a determination criterion every time the injection molding condition is changed in order to ensure determination accuracy, man-hours related to the manufacture of a molded product can be reduced.

  In the measurement apparatus according to the present invention described above, the threshold value generation unit averages the detection value for each time point acquired by the acquisition unit over a plurality of molding cycles. It is conceivable to obtain an average value for each time point and generate a threshold value for each time point based on the average value for each time point.

  Thereby, a threshold value for abnormality determination is generated based on the average waveform of the plurality of detection value waveforms obtained in the past plural molding cycles.

  In the measurement apparatus according to the present invention described above, the threshold value generation unit calculates a standard deviation of the detection value for each time point based on the detection value for each time point acquired by the acquisition unit over a plurality of molding cycles. A value obtained by offsetting the average value for each time point based on the standard deviation at the corresponding time point may be generated as the threshold value for each time point.

  Thereby, the determination which considered the variation degree of the past detected value as an abnormality determination is performed.

  In the measuring apparatus according to the present invention described above, the threshold value generation unit may generate two types of threshold values, an upper limit value and a lower limit value, as the threshold value.

  As a result, it is possible to perform abnormality determination with a certain monitoring width for the detected value.

The measurement method according to the present invention includes an acquisition step of acquiring a detection value of a sensor provided in an injection molding device, and the acquisition step acquired in the past for each time point in one molding cycle of the injection molding device. Based on a detection value, a threshold value generation step for generating a threshold value for each time point and adaptively updating the threshold value, a detection value acquired by the acquisition step, and the threshold value corresponding to the acquisition time point of the detection value And a determination step for obtaining an abnormality determination result of the injection molding condition based on a result of comparison with the threshold value adaptively updated by the threshold generation step .
In the threshold value generation step, the threshold value for each time point is generated based on the detected value for each time point of each molding cycle in a threshold generation cycle that is a past n molding cycle (n is a natural number of 2 or more). And obtaining an average value for each detected time point for the detected value by averaging the detected value for each time point acquired by the acquiring step over a plurality of molding cycles. The difference between the detected value and the average value is calculated, and the number of molding cycles of the threshold generation target cycle is adjusted based on the difference value .
Further, another measurement method according to the present invention includes an acquisition step of acquiring a detection value of a sensor disposed in an injection molding device, and the acquisition step is acquired at each time point in the molding cycle of the injection molding device in the past. Based on the detected value, a threshold value for each time point is generated, and a threshold value generating step for adaptively updating the threshold value, a detection value acquired in the acquisition step, and the detection value corresponding to the acquisition time point A determination step of obtaining an abnormality determination result of the injection molding condition based on a result of comparison with the threshold value that is a threshold value and is adaptively updated by the threshold value generation step.
And, in the threshold generation step, the average value for each detection point for the detection value is obtained by averaging the detection value for each point in time acquired in the acquisition step over a plurality of molding cycles, The difference between the detected value and the average value is calculated for each time point, and the threshold update period is adjusted based on the difference value.
The program according to the present invention is a program that causes an arithmetic processing unit to execute the processing of each step described above.

  According to the present invention, it is possible to achieve both reduction in abnormality determination accuracy and cost reduction.

It is the block diagram which showed the structure of the injection molding measurement system of embodiment. It is explanatory drawing of the display screen by the management software of embodiment. It is a block diagram of the measuring device of an embodiment. It is the figure which showed the example of the detected value waveform in 1 shaping | molding cycle. It is the figure which showed the image of the threshold value for every time. It is the flowchart which showed the process for producing | generating the threshold value for every time. It is the flowchart which showed the process for abnormality determination using the produced | generated threshold value.

<Configuration of measurement system>
Embodiments according to the present invention will be described below. First, an injection molding measuring system 100 (also simply referred to as “measuring system 100”) including a measuring device 1 and an injection molding device 2 according to an embodiment of the present invention will be described.
FIG. 1 is a diagram showing an outline of the configuration of the measurement system 100.
As illustrated, the measurement system 100 includes a measurement device 1, an injection molding device 2, a dedicated amplifier 3, and a personal computer 4.

  As generally known, the injection molding apparatus 2 includes a mold 10 disposed at a predetermined position, an injection unit 11 having a mechanism for injecting and filling a resin material into the mold 10, and an injection unit 11. The molding control unit 12 controls the injection operation and the opening / closing operation of the mold 10 to execute and control a series of injection molding operations.

For example, an upper mold and a lower mold are arranged in the mold 10, and the upper mold is opened and closed by a mechanism provided in the injection unit 11 with respect to the lower mold arranged in the molding stage, for example. In a state where the upper mold is closed with respect to the lower mold, for example, a resin material is injected into the gate provided in the upper mold by the injection cylinder of the injection unit 11 and the cavity in the mold 10 is filled with the resin material. The Then, after filling, when the required time elapses, the upper mold is opened, and the resin molded product is taken out from the cavity.
An in-mold sensor 31 is disposed in the mold 10. For example, there are a temperature sensor that detects the temperature of the filled resin material, a pressure sensor that detects the pressure of the resin material, and the like.
The structure and type of the mold 10 are not particularly limited, and various types are assumed.

The injection unit 11 is provided with mechanisms necessary for injection molding, such as a resin material injection mechanism, a mold clamping mechanism, an injection cylinder mechanism, and an injection motor for the mold 10.
The injection unit 11 is provided with an in-injection sensor 32 and a sensor amplifier 33. Examples of the in-injection sensor 32 include a temperature sensor that detects the temperature of the resin material during the injection process, a pressure sensor that detects pressure, and a position sensor that calculates the injection speed.
In the present embodiment, the mechanism and structure of the injection unit 11, such as a cylinder structure, a mold clamping mechanism structure, a runner structure, a nozzle structure, a heater arrangement, a motor arrangement, a material input mechanism, and the like are not particularly limited. It may be of a type.

The molding control unit 12 includes, for example, a microcomputer having a ROM (Read Only Memory), a RAM (Random Access Memory), and a CPU (Central Processing Unit).
The molding control unit 12 performs drive control of each unit by the injection unit 11. For example, injection motor control, mold stage operation control, mold opening / closing mechanism operation control, nozzle opening / closing mechanism operation control, heater control, material charging operation control, and the like are performed. This causes a series of injection molding operations to be executed.

The detection signal S1 of the in-mold sensor 31 is converted into a voltage value by a dedicated amplifier 3 disposed separately from the injection molding device 2, for example. Then, it is supplied to the measuring device 1 as a detection signal Vs1 converted into a voltage signal.
The detection signal S2 of the in-injection unit sensor 32 is converted into a voltage value by a sensor amplifier 33 provided in the injection unit 11, for example. Then, it is supplied to the measuring device 1 as a detection signal Vs2 converted into a voltage signal.

Here, although two detection signals are shown as the detection signals Vs1 and Vs2, the detection signal Vs1 is a generic name of the detection signals from the in-mold sensor 31, and the detection signal Vs2 is a detection signal from the in-injection section sensor 32. It is a generic name. Of course, a case where a plurality of sensors are arranged as the in-mold sensor 31 and a case where a plurality of sensors are arranged as the in-mold sensor 32 are also assumed.
Therefore, the detection signals Vs1 and Vs2 do not indicate detection signals of only two systems, but merely indicate that any detection signal of the in-mold sensor 31 and the in-injection unit sensor 32 can be input to the measuring apparatus 1. Absent.
The measuring apparatus 1 is provided with an n-channel input system and can simultaneously input n detection signals. Therefore, the detection signal Vs1 of n sensors as the in-mold sensor 31 may be supplied to the measuring device 1, or the detection signal Vs2 of n sensors as the in-injection sensor 32 may be supplied to the measuring device 1. Good. Further, one or a plurality of detection signals Vs1 and Vs2 as the in-mold sensor 31 and the in-injection section sensor 32 may be distributed to n channels and supplied to the measuring device 1.
The type of detection signal input to the measuring device 1 depends on the actual structure, type, molded product, number of mounted sensors, contents of measurement / monitoring to be executed, etc. What is necessary is just to be decided suitably.
Although not shown in the drawings, various sensors may be provided in peripheral devices of the injection molding apparatus 2, such as a temperature controller for cooling and a vacuuming apparatus, and the detection signals of these sensors are supplied to the measuring apparatus 1. It is also assumed that

Various communications are possible between the measuring apparatus 1 and the molding control unit 12. In FIG. 1, as one of the communications, various timing signals STM are transmitted from the molding control unit 12 to the measuring device 1, and a notification signal SI is transmitted from the measuring device 1 to the molding control unit 12. Which indicates that.
As one of the timing signals STM, for example, there is a signal for notifying the start / end timing of one cycle of injection molding. The measuring device 1 can detect a one-cycle molding period by one-shot resin injection using the timing signal STM, and perform logging and determination of various detection signals during that period.
Other timing signals STM include a signal indicating the start / end timing of the mold clamping period, a signal indicating the transition timing of the process, or a signal indicating the switching timing of the control method (speed control, pressure control). Conceivable.

  Hereinafter, one cycle of injection molding by the injection molding apparatus 2 is referred to as “one molding cycle”.

  The notification signal SI from the measuring device 1 is a signal for notifying the results of various types of detection information and determination information. For example, it is a signal such as an alarm notification at the time of abnormality determination in which a molding defect or the like is estimated, or a notification of the rising timing / falling timing of the detection signal waveform. The molding control unit 12 can perform various operation controls according to the notification signal SI having these contents.

Measurement results such as temperature and pressure by the measuring device 1 can be viewed by the computer device 4 connected to the measuring device 1 through a wired or wireless communication path US. The communication path US is realized by, for example, a LAN (Local Area Network) cable.
Management software for managing the measurement of various detection signals by the measuring device 1 is installed in the computer device 4. With this management software, an operator or the like can view the measurement result obtained by the measurement device 1 via the display of the computer device 4.
Moreover, the operator etc. can perform various numerical value settings by the setting using management software.
Further, the measurement result can be recorded in a predetermined storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Disk) in the computer device 4.

  FIG. 2 shows a display content example of the management screen 90 presented on the screen of the computer device 4 by the management software. As shown in the figure, on the management screen 90, the measurement results of detection signals from various sensors can be displayed as waveforms, and predetermined numerical values (for example, peak value, integral value, rise timing value, rise time, etc.) of each detection signal can be displayed. Falling timing value, etc.). Operators are provided for the operator to input various settings.

  In FIG. 2, the display of the measurement waveform is such that the drawing direction is from left to right along the time axis, but it can be switched so that the drawing is drawn from right to left. That is, the waveform drawing direction can be switched in accordance with the operator's operation. For example, in an injection molding apparatus, an injection mechanism is arranged on the right side, and there are many apparatuses that inject toward the left. As described above, when the injection direction of the resin material is from right to left, if the waveform to be displayed also proceeds from right to left, the measurement contents can be easily understood by the operator.

<Configuration of measuring device>
FIG. 3 shows the internal configuration of the measuring apparatus 1.
The measuring device 1 is provided with a calculation unit 20, an input unit 21, an A / D converter 22, a buffer and IF unit 23, and a memory unit 24.

The input unit 21 can input n channels for the detection signals Vs1 and Vs2. In the example shown in the figure, an 8-channel input is assumed, and the input channels are indicated as I1 to I8.
The detection signals Vs1 and Vs2 input to the input channels I1 to I8 are signals obtained by converting the detection information into voltage levels by the dedicated amplifier 3 or the sensor amplifier 33 as described above.
The detection signal Vs1 or Vs2 is input to all or part of the channels I1 to I8. That is, detection signals of one or a plurality of sensors arranged in the injection molding apparatus 2 as the in-mold sensor 31 and the in-injection part sensor 32 can be simultaneously input to the required channels.

The A / D converter 22 can input the same number as the number of input channels. Accordingly, in the example shown in the figure, an 8-channel input A / D converter is used.
The A / D converter 22 converts the input detection signals of the channels I <b> 1 to I <b> 8 into digital data corresponding to the voltage value, and supplies the digital data to the buffer and IF unit 23.

The buffer and IF unit 23 is a part for passing detection signals of the channels I1 to I8 to the calculation unit 20 and transmitting / receiving communication data between the calculation unit 20 and external devices (the computer device 4 and the molding control unit 12). Summarized.
For example, digital data (detection value Ddet described later) of the detection signals of a plurality of channels input simultaneously from the A / D converter 22 is temporarily buffered by the buffer and IF unit 23 and detected at each time point. Are sequentially transferred to the arithmetic unit 20 together with time information at the time of sampling of the detection signal.
The notification signal SI from the arithmetic unit 20 is transmitted from the terminal TM2 to the molding control unit 12 by the buffer and IF unit 23. Various timing signals STM from the molding control unit 12 are once taken into the buffer and IF unit 23 from the terminal TM1 and sequentially transferred to the calculation unit 20 together with time information.
Various kinds of information communication between the arithmetic unit 20 and the computer apparatus 4 are executed by the communication path US connected to the terminal TM3 (for example, a LAN connector terminal) via the buffer and IF unit 23.

The arithmetic unit 20 is constituted by a microcomputer having, for example, a ROM, a RAM, and a CPU.
In the present embodiment, the arithmetic unit 20 particularly has functions as an acquisition processing unit 20a, a data log processing unit 20b, a threshold generation processing unit 20c, and a determination processing unit 20d.

The acquisition processing unit 20a acquires a detection value of a sensor provided in the injection molding apparatus. That is, this corresponds to the function of acquiring the sampling value of the detection signal transferred from the A / D converter 22 via the buffer and IF unit 23.
Hereinafter, a value obtained by digital sampling of the detection signal of the in-mold sensor 31 or the injection unit sensor 32 is referred to as “detection value Ddet”.

  The data log processing unit 20b performs a process of storing the detection value Ddet at each time point acquired by the acquisition processing unit 20a as log data. For example, for each of the detection signals of the channels I1 to I8 converted into digital values by the A / D converter 22, a value for each sample (detection value Ddet) is stored together with time information at the time of sampling.

The threshold value generation processing unit 20c generates a threshold value for each time point based on the detection value Ddet acquired by the acquisition processing unit 20a for each time point in the molding cycle of the injection molding device 2 in the past.
The determination processing unit 20d obtains an abnormality determination result of the injection molding situation based on the result of comparing the detection value Ddet acquired by the acquisition processing unit 20a with the threshold value generated corresponding to the acquisition time point of the detection value Ddet. .
A specific processing example by the calculation unit 20 for realizing the functions as the threshold generation processing unit 20c and the determination processing unit 20d will be described later.

The memory unit 24 collectively indicates a memory area that can be used by the arithmetic unit 20 as, for example, a ROM, a work memory, a nonvolatile memory, or the like.
The memory unit 24 is used, for example, as a storage area for log data by the processing of the data log processing unit 20b. The memory unit 24 is used as a work area for various arithmetic processes. The memory unit 24 is also used as a storage area for programs for realizing the processing of the calculation unit 20, particularly the processing of the acquisition processing unit 20a, the data log processing unit 20b, the threshold value generation processing unit 20c, and the determination processing unit 20d. .

<Example of detected value waveform in one molding cycle>
FIG. 4 shows an example of the waveform of the detection value Ddet obtained from the detection signal from the sensor as the in-mold sensor 31 or the injection unit sensor 32. The vertical axis represents the detection value Ddet, and the horizontal axis represents time. For example, the waveform shown in this figure is the detection value Ddet of the pressure sensor.

  In FIG. 4, time points T0 to T1 are periods of one molding cycle. In this one molding cycle, for example, the upper and lower molds of the mold 10 are closed, the resin material is injected into the mold 10 by the cylinder of the injection unit 11, the pressure after filling, and the molding solidification. Each process such as weighing / cooling, mold opening, and extrusion of the molded product is included.

<Abnormality determination method as embodiment>
Here, as described above, the abnormality determination based on the detection signal of the sensor is determined when the injection molding conditions are changed, for example, by changing the mold 10 or the resin material used in the injection molding apparatus 2. Standards may need to be reviewed. However, it takes a considerable amount of man-hours to formulate the judgment criteria manually, which helps to increase the cost of manufacturing the product.
For this reason, in the present embodiment, a determination criterion, that is, a threshold value used for the determination is generated by the arithmetic processing of the measuring apparatus 1, thereby reducing the number of man-hours.

  At this time, in the embodiment, a method of generating a threshold value based on the actual detection value Ddet is adopted so that the determination threshold value adaptively changes in response to a change in injection molding conditions such as a change in the mold 10.

Further, in the present embodiment, the abnormality determination is performed not on only the local part of the detected value waveform in one molding cycle but on a relatively wide range including at least a plurality of steps. For example, in this example, abnormality determination using the detection value Ddet at each time point and the determination threshold is performed for the entire one molding cycle. In this case, the threshold value is generated for each time point in one molding cycle.
By adopting such a method, even if it is not possible to specify at which timing (for example, process) the waveform difference appears in one molding cycle between when a good product is produced and when a defective product is produced, an abnormality is determined. Can be performed appropriately.

In addition, when performing abnormality determination based on the detection signal, it should be noted that the detected value waveform may change while the injection molding is repeated even after the injection molding conditions such as the type of the mold 10 are fixed. Should. For example, if the shot is repeated after the injection molding apparatus 2 is started, the temperature of the mold 10 may rise and the resin may enter the mold 10 better. As the detected value Ddet for pressure, the waveform shifts with time from the time of activation (for example, the peak portion shown in FIG. 4 arrives at an earlier timing).
Since such a waveform change is a change unrelated to the quality of the molded product, it is not desirable to determine that it is abnormal. In other words, it is desirable that the threshold for determination can follow such a change with time of the waveform.
Therefore, in the present embodiment, the detection value Ddet for each time point acquired by the acquisition processing unit 20a over a plurality of molding cycles is averaged for each time point, thereby obtaining an average value Va for each detection time value Ddet, A threshold value for each time point is generated based on the average value Va for each time point.

In this example, two types of upper limit value Lu and lower limit value Ll are generated as threshold values for each time point.
FIG. 5 shows an image of the upper limit value Lu and the lower limit value Ll for each time point.
In FIG. 5, a waveform A indicated by a broken line represents an average value Va for each time point as a waveform. A waveform U and a waveform L indicated by solid lines represent the upper limit value Lu for each time point and the lower limit value Ll for each time point, respectively.
In this example, the upper limit value Lu and the lower limit value Ll are generated as values obtained by giving an offset as indicated by an arrow in the drawing to the average value Va at the corresponding time point.
Specifically, the offset value given to the average value Va is a value based on the standard deviation σ of the detection value Ddet acquired over a plurality of past molding cycles at the corresponding time point. In this example, the upper limit value Lu and the lower limit value Ll are generated with the offset value = 3σ. That is, as the upper limit value Lu and the lower limit value Ll in this case, the average value Va and the standard deviation σ calculated for each corresponding time point are used, and “upper limit value Lu = average value Va + 3σ” “lower limit value Ll = average value Va”. −3σ ”.
In the abnormality determination of this example, it is determined whether or not the condition “Lu ≦ Ddet ≦ Ll” is satisfied for the detection value Ddet, the upper limit value Lu, and the lower limit value Ll for each time point. That is, if the condition is not satisfied, a determination result of abnormality is obtained.

  By generating the threshold value based on the standard deviation σ as described above, it is possible to perform a determination in consideration of the degree of variation in the past detection value Ddet as the abnormality determination. In particular, by generating a threshold value using 3σ as described above, it is possible to realize a determination that only a situation that is presumed to be peculiar from a past tendency is an abnormality target. That is, it is possible to prevent the abnormality determination sensitivity from becoming excessively high.

<Processing procedure>
A processing procedure for realizing the abnormality determination method according to the embodiment described above will be described with reference to the flowcharts of FIGS.
FIG. 6 shows processing for generating threshold values (upper limit Lu and lower limit value Ll) for each time point, and FIG. 7 shows processing for abnormality determination using the generated threshold values.
Here, it is assumed that threshold generation and abnormality determination are performed in real time during injection molding by the injection molding apparatus 2. In this case, the processing unit 20 performs the processing illustrated in FIGS. 6 and 7 as parallel processing.
The processing shown in FIGS. 6 and 7 can be performed in parallel by the arithmetic unit 20 for a plurality of types of detection values Ddet (actual processing may be time division).

First, in FIG. 6, the calculation unit 20 stands by in step S <b> 101 until the total number of shots from the start reference time becomes a predetermined value n or more. Note that n is a natural number of 2 or more.
Here, the start reference time represents a reference time point at which the abnormality determination process shown in FIG. 7 is started.
For example, the process for determining an abnormality shown in FIG. 7 can be started in response to a predetermined operation input to the computer apparatus 4. In this case, the start reference time may be the time when the predetermined operation input is detected. Alternatively, if the process shown in FIG. 7 is started in response to the start of the injection molding device 2, for example, the time point at which the start is detected may be set as the start reference time. The activation of the injection molding device 2 can be determined based on whether or not the timing signal STM is detected when a signal indicating the activation timing is output as the timing signal STM described above.
The total number of shots is synonymous with the number of injection molding operations for one molding cycle. The total number of shots is specified by, for example, the number of receptions of the start timing signal of one molding cycle input from the injection molding apparatus 2 side as one of the timing signals STM.

When the total number of shots from the start reference time is equal to or greater than the predetermined value n, the calculation unit 20 proceeds to step S102 and waits until the cycle end timing is reached. This corresponds to waiting for the end timing of the nth molding cycle from the start reference time. The end timing standby process can be realized as a process of waiting for an end timing signal of one molding cycle input from the injection molding apparatus 2 side as one of the timing signals STM.
Alternatively, the end timing may be managed by the arithmetic unit 20 using an internal timer count as a predetermined time has elapsed from the latest start timing.

  When the cycle end timing is confirmed, the calculation unit 20 sets the time point identification value x to “0” in step S103. In this example, the time point identification value x is an identification value for identifying each time point (each sampling timing of the detection value Ddet) in one molding cycle. For example, if the total number of samplings of the detection value Ddet in one molding cycle is “α”, the time point identification value x can take a value from “0” to “α−1”.

  In step S104 following step S103, the arithmetic unit 20 calculates an average value Va for the detected value Ddet at the xth time point in the past n cycles. In this example, the acquired detection value Ddet is stored as a log in a predetermined area of the memory unit 24 by the function as the data log processing unit 20b of the calculation unit 20 (stored together with time information at the time of sampling). In step S104, for example, the detection value Ddet at the corresponding time in the corresponding cycle is specified from the detection values Ddet of the past n cycles stored by the function, and the average value Va is calculated for the specified detection value Ddet.

In step S <b> 105 following step S <b> 104, the calculation unit 20 performs processing for storing the calculated average value Va as the threshold reference value at the x-th time point.
Further, in the subsequent step S106, the arithmetic unit 20 calculates the standard deviation σ at the xth time point in the past n cycles. That is, the standard deviation σ is calculated for the detected value Ddet at the xth time point in each of the past n cycles specified in the previous step S104.

  In response to the calculation of the standard deviation σ in step S106, the calculation unit 20 proceeds to step S107 and performs a process of calculating “threshold reference value + 3σ” and storing it as the upper limit value Lu at the xth time point. Further, in the subsequent step S <b> 108, the calculation unit 20 performs a process of calculating “threshold reference value−3σ” and storing it as the x-th time point lower limit value Ll.

In step S109 following step S108, the calculation unit 20 determines whether or not the time point identification value x is equal to or greater than the maximum value xMAX. Here, the maximum value xMAX is the same value as “α−1” described above.
If the time point identification value x is not equal to or greater than the maximum value xMAX, the calculation unit 20 proceeds to step S110, increments the time point identification value x by 1, and then returns to step S104. Thus, the upper limit Lu and the lower limit Ll are generated for each remaining time point in one molding cycle.

  On the other hand, if the time point identification value x is equal to or greater than the maximum value xMAX, the calculation unit 20 proceeds to step S111 and determines whether or not the end condition is satisfied. That is, it is determined whether or not a predetermined condition that the process for abnormality determination shown in FIG. Examples of the termination condition include a predetermined operation input using the operator's computer device 4, completion of processing for a specified number of molding cycles, or a case where abnormality is determined by the processing of FIG. 7. it can.

If the end condition is not satisfied, the calculation unit 20 returns to step S102. Thus, the upper limit value Lu and the lower limit value Ll at each time point are generated for each molding cycle until the end condition is satisfied.
On the other hand, when the end condition is satisfied, the arithmetic unit 20 finishes the process shown in FIG.

  In steps S107 and S108, the upper limit value Lu and the lower limit value Ll stored in the past can be overwritten and stored by the latest upper limit value Lu and lower limit value Ll, respectively.

With the processing in FIG. 6, threshold values adaptively generated from the detection values Ddet for the past n shots are sequentially obtained for each molding cycle after the (n + 1) th shot from the start reference time.
Hereinafter, the threshold value generated in the process of FIG. 6 is referred to as an “adaptive threshold value” for convenience.

Here, in the process of FIG. 6, the adaptive threshold value cannot be obtained in each molding cycle from the start reference time to the n-th shot. For this reason, the “initial threshold value” described below is used.
As the initial threshold value, for example, the injection molding apparatus 2 is operated before the start reference time, and the detection value Ddet acquired in a plurality of molding cycles at that time is used in the same method as the method described in FIG. 6 (S104 to S108). Generate in advance.
For example, an operator or the like causes the injection molding apparatus 2 to perform an operation for n molding cycles as an injection molding operation before the start reference time. At this time, the workers check the performance of the molded product obtained in each molding cycle in those n molding cycles, and if all the products are determined to be non-defective products, the threshold generation using the detection value Ddet for these n molding cycles is measured. 1 (arithmetic unit 20). For example, the initial threshold value is obtained by such a method.
Alternatively, as an injection molding operation before the start reference time, the injection molding device 2 is caused to execute an operation for a molding cycle larger than the n molding cycles, and the measurement device 1 stores log data of the detection value Ddet obtained in each molding cycle. Remember. An operator or the like displays the waveform of the detection value Ddet based on the log data on the computer device 4 and selects a molding cycle at which a good waveform is determined to be obtained from each molding cycle. The computer apparatus 4 accepts the selection, and instructs the measuring apparatus 1 to execute threshold generation using the detection value Ddet of each selected molding cycle. The initial threshold can be obtained also by such a method.
For example, the measurement apparatus 1 (calculation unit 20) generates an initial threshold value at a stage before the start reference time by the above method. In this case, the arithmetic unit 20 stores the generated initial threshold value in the memory unit 24, for example.

Subsequently, processing for determining an abnormality will be described.
In FIG. 7, the arithmetic unit 20 first waits until the cycle start timing is reached in step S201. Specifically, a process of waiting for the above-described start timing signal as one of the timing signals STM is performed.
If the start timing signal confirms that the cycle start timing is reached, the calculation unit 20 determines in step S202 whether the total number of shots from the start reference time is “n + 1” or more.

If the total number of shots from the start reference time is not “n + 1” or more, the calculation unit 20 proceeds to step S203 and performs a process using the reference threshold as an initial threshold. That is, a setting process (for example, a flag setting process) is performed to set a threshold referred to in an abnormality determination process (S208) described later, that is, a threshold to be acquired in step S207 as the initial threshold described above.
On the other hand, if the total number of shots from the start reference time is equal to or greater than “n + 1”, the calculation unit 20 proceeds to step S204 and performs a process using the reference threshold as an adaptive threshold.

  In response to the execution of the processing in step S203 or S204, the calculation unit 20 sets the time point identification value x to “0” in step S205, and waits until the detection value Ddet at the xth time point is acquired in step S206. That is, in this example, the process waits until the detection value Ddet transferred from the A / D converter 22 via the buffer and IF unit 23 is acquired.

  When the detection value Ddet at the xth time point is acquired, the calculation unit 20 proceeds to step S207 and acquires the upper and lower limit values at the xth time point. That is, if the setting process described in step S203 is performed, the calculation unit 20 acquires the upper limit value Lu and the lower limit value Ll at the xth time point stored as the initial threshold value. On the other hand, if the setting process in the previous step S204 has been performed, the arithmetic unit 20 calculates the upper limit value Lu and the lower limit value Ll at the xth time point stored for each molding cycle by the processes in steps S107 and S108 in FIG. , Get the latest value respectively. In other words, the adaptive threshold value generated in the molding cycle immediately before the molding cycle currently being processed is acquired.

  In response to the execution of the acquisition process in step S207, the calculation unit 20 determines in step S208 whether or not the condition of “lower limit L1 ≦ detected value Ddet ≦ upper limit Lu” is satisfied. That is, it is determined whether or not the detection value Ddet acquired in step S206 is not less than the lower limit value L1 and not more than the upper limit value Lu acquired in step S207.

If the condition of “lower limit value L1 ≦ detection value Ddet ≦ upper limit value Lu” is satisfied, the calculation unit 20 proceeds to step S209, determines whether or not the time point identification value x is greater than or equal to the maximum value xMAX. Is not equal to or greater than the maximum value xMAX, the time point identification value x is incremented by 1 in step S210, and the process returns to step S206.
As a result, until the condition is not satisfied (determined as abnormal) in step S208, the abnormality determination is repeated for the remaining detection values Ddet at each time point within one molding cycle to be processed.

On the other hand, if the condition of “lower limit L1 ≦ detected value Ddet ≦ upper limit Lu” is not satisfied, the arithmetic unit 20 executes the abnormality handling process of step S211. As the abnormality handling process, it is conceivable to perform a process of transmitting a notification signal SI (alarm signal) for notifying an abnormality to the molding control unit 12 and notifying the computer device 4 of the abnormality.
At this stage, the determination result information indicating abnormality (molding failure) may be stored as log data together with the identification information of the current molding cycle.

When the time point identification value x is greater than or equal to the maximum value xMAX in the previous step S209, or in response to the execution of the abnormality handling process in step S211, the calculation unit 20 proceeds to step S212, and whether or not the end condition is satisfied. Determine. Since the determination process in step S212 is the same as the determination process in the previous step S111, redundant description is avoided.
If the end condition is not satisfied, the calculation unit 20 returns to step S201. That is, the above-described processing is executed again for a new molding cycle.
On the other hand, if the end condition is satisfied, the calculation unit 20 ends the process shown in FIG.

  In the above description, an example in which the initial threshold value (common value) is used in each molding cycle from the start reference time to the n molding cycle has been described. However, the processing in FIG. 6 was acquired in each molding cycle before the start reference time. The threshold value generation based on the detection value Ddet of the past n molding cycles including the detection value Ddet may be performed. Thus, the initial threshold value can be used only in one molding cycle immediately after the start reference time, and the adaptive threshold value can be used in each subsequent molding cycle.

  In the above, an example is given in which the threshold value generation and the abnormality determination process using the threshold value are performed in real time while logging the detection value Ddet during execution of one molding cycle. Log data can also be obtained at a later time. That is, during the execution of one molding cycle, the detection values Ddet of the channels I1 to I8 input and acquired from the input unit 21 at each time point or the detection values Ddet stored as log data by the data log processing unit 20b are used. Thus, threshold generation and abnormality determination processing can be performed at various timings.

  In the above description, the threshold generation cycle is set to one molding cycle. However, the threshold generation cycle may be set to a plurality of molding cycles. Specifically, for example, by performing the processing of steps S103 to S110 in FIG. 6 for each N (N is a natural number of 2 or more) molding cycle, a threshold value is generated for each N molding cycle. The threshold value used every time is changed to a newly generated threshold value.

At this time, the threshold generation cycle can be made variable based on the evaluation index, using the difference Δd of the detected value Ddet with respect to the average value Va as an evaluation index. The difference Δd can be regarded as an index of the stability of the injection molding operation (the difference Δd is small = stable).
For example, if the average value of the difference Δd at each time point or the average value in the time axis direction (referred to as “average value Vat”) is less than or equal to a predetermined value, the threshold value used for abnormality determination is generated for each molding cycle Instead, it is conceivable to carry out every plural molding cycles. Alternatively, if the maximum value or average value of the difference Δd at each time point is continuously equal to or less than a predetermined value for a predetermined number of times (or a frequency that is equal to or lower than the predetermined value is equal to or higher than a predetermined frequency), a plurality of threshold values used for abnormality determination are formed It can also be done every cycle.
Thereby, at the time of stability, the threshold value calculated once can be used over a plurality of molding cycles. It is not necessary to perform generation of the threshold value every molding cycle, and the calculation processing load of the calculation unit 20 can be reduced. At this time, if it is determined that it is not stable, the threshold generation cycle is returned for each molding cycle.
Note that the threshold generation cycle is not adjusted in two steps as in the above example, but may be adjusted in many steps according to the magnitude of the average value Vat.
It is also possible to accept an operation input from an operator or the like and make the threshold generation cycle variable according to the operation input.

In the above description, the threshold value is generated based on the detection value Ddet acquired over the past n molding cycles. However, the value of “n” may be variable instead of fixed. At this time, if the value of “n” is reduced, the calculation processing burden related to the generation of the threshold can be reduced.
For example, the value of “n” can be made variable according to the operation input.
Alternatively, the stability of the injection molding operation is determined for the difference Δd by the same method as described above. If it is determined that the difference is stable, the value of “n” is set to a small value or the average value Vat is set. Thus, the value of “n” can be adjusted in multiple stages.

In the above description, the average value Va is exemplified as the threshold reference value (the reference value for generating the threshold value). However, the threshold reference value is, for example, the center of the detection value Ddet acquired over the past plural molding cycles for the target time point. It is not limited to the average value Va, such as a value.
Further, the offset value given to the threshold reference value is not limited to a value based on the standard deviation σ, and is not limited to a value based on the standard deviation σ, such as giving an offset based on a fixed value.
Further, the offset value given to the threshold reference value can be made variable according to the operation input. For example, when an offset based on the standard deviation σ is given, it is conceivable that a selection operation of 2σ / 3σ can be performed. Thereby, it is possible to perform abnormality determination based on an appropriate standard according to the purpose.

  Note that the measurement device 1 functions as a data logger for detection values Ddet of various sensors provided in the injection molding device 2, for example, a pressure sensor and a temperature sensor, by the function as the data log processing unit 20b described above. This makes it possible to perform an ex-post evaluation on the detection value Ddet. That is, by storing the detection value Ddet during the molding cycle as log data, it is possible to obtain information useful for various evaluations, operation analysis, adjustment, etc., such as analysis of the detection value Ddet in a plurality of cycles. it can.

  In addition, storing the abnormality determination result as log data is also useful for evaluation and adjustment at a later time. In particular, it also means that the determination process once executed is not wasted.

<Summary and modification>
As described above, the measurement device 1 according to the embodiment includes an acquisition unit (acquisition processing unit 20a) that acquires a detection value of a sensor provided in the injection molding device 2, and one acquisition cycle of the injection molding device in the past. Corresponding to the threshold value generation unit (threshold generation processing unit 20c) that generates a threshold value for each time point based on the detection value acquired for each time point in the network, the detection value acquired by the acquisition unit, and the acquisition time point of the detection value And a determination unit (determination processing unit 20d) for obtaining an abnormality determination result of the injection molding situation based on a result of comparison with the generated threshold value.

According to the measurement apparatus 1 described above, the abnormality determination of the injection molding situation is performed based on the threshold value generated based on the actual detection value, and therefore it is possible to perform the abnormality determination based on the determination criterion adapted to the change in the injection molding conditions. Is done. Therefore, it is possible to suppress a decrease in abnormality determination accuracy due to a change in injection molding conditions.
In addition, since it is not necessary to search for a determination criterion every time the injection molding condition is changed in order to ensure determination accuracy, man-hours related to the manufacture of a molded product can be reduced.
From these points, according to the present embodiment, it is possible to achieve both reduction in abnormality determination accuracy and cost reduction.

  Moreover, in the measuring device 1 of the embodiment, the threshold generation unit averages the detection values for each time point acquired by the acquisition unit over a plurality of molding cycles for each time point, thereby averaging the detection values for each time point. And a threshold value for each time point is generated based on the average value for each time point.

Thereby, a threshold value for abnormality determination is generated based on the average waveform of the plurality of detection value waveforms obtained in the past plural molding cycles.
Therefore, for example, a threshold value that follows a change in waveform over time after the start of the injection molding operation such as a waveform shift due to a temperature change after the start of the injection molding operation can be generated, and the accuracy of abnormality determination can be improved. it can.

  Furthermore, in the measuring apparatus 1 of the embodiment, the threshold value generation unit calculates a standard deviation for each time point of the detection value based on the detection value for each time point acquired by the acquisition unit over a plurality of molding cycles, A value obtained by offsetting the average value for each time point based on the standard deviation at the corresponding time point is generated as a threshold value for each time point.

Thereby, the determination which considered the variation degree of the past detected value as an abnormality determination is performed.
Therefore, it is possible to realize the determination that only the situation that is presumed to be unusual from the past tendency is an abnormality target, and it is possible to prevent the sensitivity of the abnormality determination from becoming excessively high. Reduction suppression is achieved.

  Furthermore, in the measurement apparatus 1 according to the embodiment, the threshold value generation unit generates two types of threshold values, that is, an upper limit value and a lower limit value as threshold values.

  Thereby, abnormality determination with a certain monitoring width can be performed on the detected value.

The present invention should not be limited to the specific examples described above, and various modifications can be considered.
For example, various configurations of the injection molding apparatus 2 are conceivable. The configuration of the measuring device 1 is the same.
Further, the processing examples of the calculation unit 20 shown in FIGS. 6 and 7 are merely examples, and various specific processing examples can be considered.
Various sensors (in-mold sensor 31 and in-injection sensor 32) mounted on the injection molding apparatus 2 can be considered. In other words, the measuring device 1 can measure various detection signals other than the pressure measurement of the resin material in the injection unit 11 and the mold 10 by the pressure sensor and the measurement of the molding material and the mold surface temperature based on the detection signal of the temperature sensor. Applicable to measurement. For example, flow rate measurement of molding material based on detection signal of optical sensor, flow front measurement based on detection signal of infrared sensor, etc. (for example, measurement of time until molding resin reaches predetermined position in cavity), position sensor, etc. The present invention can also be suitably applied to detection signals of various sensors when performing other measurements related to injection molding, such as measurement of misalignment between molds when the mold is closed based on the detection signal (measurement of mold opening amount).

<Program and storage medium>
The program according to the embodiment of the present invention causes the calculation unit 20 (an arithmetic processing device such as a microcomputer) in the measurement apparatus 1 to execute functions as the acquisition processing unit 20a, the threshold generation processing unit 20c, and the determination processing unit 20d. It is a program.

  The program according to the embodiment includes an acquisition step of acquiring a detection value of a sensor arranged in the injection molding device, and the detection value acquired at each time point in the molding cycle of the injection molding device in the past by the acquisition step. Based on the result of comparing the threshold value generation step for generating the threshold value for each time point, the detection value acquired in the acquisition step, and the threshold value generated corresponding to the acquisition time point of the detection value, This is a program for causing the arithmetic processing unit to execute a determination step (S115) for obtaining a determination result of the abnormality of the situation. That is, it is a program for executing the processing of FIG. 6 and FIG.

Such a program facilitates the manufacture of the measuring apparatus 1 of the present embodiment.
Such a program can be stored in advance in a storage medium built in a device such as a computer device or a ROM in a microcomputer having a CPU. Alternatively, it can be stored (stored) temporarily or permanently in a removable storage medium such as a semiconductor memory, memory card, optical disk, magneto-optical disk, or magnetic disk. Such a removable storage medium can be provided as so-called package software.
Further, such a program can be installed from a removable storage medium to a personal computer or the like, or can be downloaded from a download site via a network such as a LAN or the Internet.

Moreover, the computer apparatus 4 can also be provided with the function of the measuring device 1 by installing the program of such embodiment in the computer apparatus 4. FIG.
For example, the dedicated amplifier 3 and the computer device 4 are directly connected by a connector. A detection signal of a single or a plurality of input channels is supplied simultaneously to the computer apparatus 4 via the dedicated amplifier 3. And when the software containing the said program is started in the computer apparatus 4, the process of FIG. 6 and FIG. That is, processing for obtaining a detection value for the detection signals of the sensors (31, 32), generating a threshold value based on the detection value, and obtaining an abnormality determination result of the injection molding situation based on the detection value and the threshold value is performed. Thereby, the measuring apparatus 1 is realizable using computer apparatuses 4, such as a personal computer.

  DESCRIPTION OF SYMBOLS 1 Measurement apparatus, 2 Injection molding apparatus, 3 Dedicated amplifier, 4 Computer apparatus, 10 Mold, 11 Injection part, 12 Molding control part, 20 Calculation part, 20a Acquisition process part, 20b Data log process part, 20c Threshold generation process part 20d determination processing unit, 21 input unit, 22 A / D converter, 23 buffer and IF unit, 24 memory unit, 31 sensor in mold, 32 sensor in injection unit, 33 sensor amplifier, 100 measurement system

Claims (9)

An acquisition unit for acquiring a detection value of a sensor provided in the injection molding apparatus;
Based on the detection value acquired for each time point in the molding cycle of the injection molding apparatus in the past by the acquisition unit, a threshold value generation unit that adaptively updates the threshold value while generating a threshold value for each time point;
Based on the result of comparing the detection value acquired by the acquisition unit with the threshold value corresponding to the acquisition time point of the detection value and the threshold value generation unit adaptively updated, the abnormality determination of the injection molding situation a determining unit for determining a result, the Bei example,
The threshold generation unit
The threshold value for each time point is generated based on the detected value for each time point of each molding cycle in a threshold generation target cycle that is a past n molding cycle (n is a natural number of 2 or more),
The acquisition unit obtains an average value for each detection point by averaging the detection values for each time point acquired over a plurality of molding cycles, and the detection value for each time point. A measuring device that calculates a difference from the average value and adjusts the number of molding cycles of the threshold generation target cycle based on the difference value .   An acquisition unit for acquiring a detection value of a sensor provided in the injection molding apparatus;
  Based on the detection value acquired for each time point in the molding cycle of the injection molding apparatus in the past by the acquisition unit, a threshold value generation unit that adaptively updates the threshold value while generating a threshold value for each time point;
  Based on the result of comparing the detection value acquired by the acquisition unit with the threshold value corresponding to the acquisition time point of the detection value and the threshold value generation unit adaptively updated, the abnormality determination of the injection molding situation A determination unit for obtaining a result,
  The threshold generation unit
  The acquisition unit obtains an average value for each detection point by averaging the detection values for each time point acquired over a plurality of molding cycles, and the detection value for each time point. The difference from the average value is calculated, and the threshold update period is adjusted based on the difference value.
  Measuring device. The threshold generation unit
By averaging the detection values for each time point acquired by the acquisition unit over a plurality of molding cycles for each time point, an average value for each time point for the detection value is obtained, and based on the average value for each time point The measurement device according to claim 1 or 2 , wherein a threshold value for each time point is generated. The threshold generation unit
Based on the detection value for each time point acquired by the acquisition unit over a plurality of molding cycles, the standard deviation for each time point of the detection value is calculated, and the average value for each time point corresponds to the standard deviation at the corresponding time point. Generates a value offset based on the threshold value for each time point
The measuring device according to claim 3 . The threshold generation unit
Measurement apparatus according to any one of claims 1 to 4 to produce two kinds of threshold upper limit and the lower limit value as the threshold value. An acquisition step of acquiring a detection value of a sensor provided in the injection molding apparatus;
Based on the detection value acquired for each time point in one molding cycle of the injection molding apparatus in the past, the threshold value generating step for adaptively updating the threshold value and generating the threshold value for each time point;
Based on the result of comparing the detection value acquired by the acquisition step and the threshold value corresponding to the acquisition time point of the detection value and adaptively updated by the threshold value generation step, the abnormality determination of the injection molding situation A determination step for obtaining a result ,
In the threshold generation step,
The threshold value for each time point is generated based on the detected value for each time point of each molding cycle in a threshold generation target cycle that is a past n molding cycle (n is a natural number of 2 or more),
The acquisition step obtains an average value for each time point for the detection value by averaging the detection value for each time point acquired over a plurality of molding cycles, and the detection value for each time point. A measurement method of a measurement apparatus that calculates a difference from the average value and adjusts the number of molding cycles of the threshold generation target cycle based on the difference value .   An acquisition step of acquiring a detection value of a sensor provided in the injection molding apparatus;
  Based on the detection value acquired for each time point in one molding cycle of the injection molding apparatus in the past, the threshold value generating step for adaptively updating the threshold value and generating the threshold value for each time point;
  Based on the result of comparing the detection value acquired by the acquisition step and the threshold value corresponding to the acquisition time point of the detection value and adaptively updated by the threshold value generation step, the abnormality determination of the injection molding situation A determination step for obtaining a result,
  In the threshold generation step,
  The acquisition step obtains an average value for each time point for the detection value by averaging the detection value for each time point acquired over a plurality of molding cycles, and the detection value for each time point. The difference from the average value is calculated, and the threshold update period is adjusted based on the difference value.
  Measuring method of the measuring device. An acquisition step of acquiring a detection value of a sensor provided in the injection molding apparatus;
Based on the detection value acquired for each time point in one molding cycle of the injection molding apparatus in the past, the threshold value generating step for adaptively updating the threshold value and generating the threshold value for each time point;
Based on the result of comparing the detection value acquired by the acquisition step and the threshold value corresponding to the acquisition time point of the detection value and adaptively updated by the threshold value generation step, the abnormality determination of the injection molding situation a determining step of determining the result, to be executed by the arithmetic processing unit,
In the threshold generation step,
The threshold value for each time point is generated based on the detected value for each time point of each molding cycle in a threshold generation target cycle that is a past n molding cycle (n is a natural number of 2 or more),
The acquisition step obtains an average value for each time point for the detection value by averaging the detection value for each time point acquired over a plurality of molding cycles, and the detection value for each time point. The program which makes the said arithmetic processing unit perform the process which calculates the difference with the said average value, and adjusts the molding cycle number of the said threshold value production | generation object cycle based on the value of this difference .   An acquisition step of acquiring a detection value of a sensor provided in the injection molding apparatus;
  Based on the detection value acquired for each time point in one molding cycle of the injection molding apparatus in the past, the threshold value generating step for adaptively updating the threshold value and generating the threshold value for each time point;
  Based on the result of comparing the detection value acquired by the acquisition step and the threshold value corresponding to the acquisition time point of the detection value and adaptively updated by the threshold value generation step, the abnormality determination of the injection molding situation A determination step for obtaining a result, and causing the arithmetic processing unit to execute,
  In the threshold generation step,
  The acquisition step obtains an average value for each time point for the detection value by averaging the detection value for each time point acquired over a plurality of molding cycles, and the detection value for each time point. A process of calculating a difference from the average value and adjusting an update period of the threshold based on the difference value is executed by the arithmetic processing unit.
  program.

Images