Characteristic prediction device and characteristic prediction method

Abstract

To provide a characteristic prediction device and a characteristic prediction method that can accurately predict the characteristic of a resin to be extruded and molded.SOLUTION: A characteristic prediction device 1 that predicts characteristics of a resin extruded and molded by an extruder 60 includes: a vibration sensor 11 that is attached to a crosshead 62 of the extruder 60 to detect vibrations; a preprocessing part 22 that performs frequency analysis on outputs of the vibration sensor 11 to obtain an intensity spectrum; and a prediction processing part 24 that predicts characteristics based on a relationship between characteristics and the spectrum total value in a predetermined frequency range of the intensity spectrum obtained in advance and the intensity spectrum obtained by the preprocessing part 22.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a characteristics prediction device and a characteristics prediction method. [Background technology]

[0002] Previously, techniques have been proposed to predict the properties of molded products obtained by extrusion molding. In particular, in recent years, many techniques related to abnormality detection using AE (acoustic emission) sensors have been proposed, and efforts are being made to apply these to abnormality detection during extrusion molding.

[0003] For example, Patent Document 1 discloses a technology for acquiring the output of an AE sensor installed on the surface of the housing of an extruder while the extruder is in operation, and determining whether an abnormality has occurred based on the acquired AE sensor output. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 8-216230 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the above-mentioned Patent Document 1, the AE sensor is attached to the housing of the extruder, which makes it susceptible to the effects of crushing sounds and the like that occur immediately after the resin material is added, making it difficult to accurately predict the characteristics of the resin being extruded.

[0006] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a property prediction device and a property prediction method that are capable of accurately predicting the properties of a resin to be extruded. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems, the present invention provides an apparatus for predicting the properties of a resin extruded by an extruder, the apparatus comprising: a vibration sensor attached to a crosshead of the extruder for detecting vibrations; a pre-processing unit for performing frequency analysis on the output of the vibration sensor to obtain an intensity spectrum; and a prediction processing unit for predicting the properties based on the relationship between the properties and a spectral sum value in a predetermined frequency range of the intensity spectrum obtained in advance, and the intensity spectrum obtained by the pre-processing unit.

[0008] Furthermore, in order to solve the above-mentioned problems, the present invention provides a method for predicting the properties of a resin extruded by an extruder, comprising: a preprocessing step of using a vibration sensor attached to the crosshead of the extruder to detect vibrations and performing frequency analysis on the output of the vibration sensor to obtain an intensity spectrum; and a prediction processing step of predicting the properties based on the relationship between the properties and the spectral sum value of a predetermined frequency range of the intensity spectrum obtained in advance, and the intensity spectrum obtained in the preprocessing step. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a property prediction device and a property prediction method that can accurately predict the properties of a resin to be extruded. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a schematic configuration diagram of a characteristic prediction device according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating how to obtain an intensity spectrum. [Figure 3] FIG. 10 is a diagram illustrating an example of preprocessed data. [Figure 4] 10(a) is an example of relational data used when determining the viscosity of a resin, and FIG. 10(b) is an example of relational data used when determining the shrinkage rate of a molded body. [Figure 5] FIG. 10 is a flow chart for creating relationship data. [Figure 6] FIG. 10 is a flow diagram of preprocessing. [Figure 7] FIG. 1 is a flow diagram of a method for predicting properties of a composite material according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0011] [Embodiment Mode] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

[0012] 1 is a schematic diagram of a property prediction device 1 according to this embodiment. The property prediction device 1 is a device that predicts the properties of a resin extrusion molded by an extruder 60, and includes a vibration sensor 11 that detects vibrations, and a computing device 12. The "properties of the resin to be extruded" referred to here include both the properties of the molded product obtained by extrusion molding and the properties of the resin in a molten state during molding.

[0013] The properties of a molded product obtained by extrusion molding are thought to be significantly affected by the viscosity of the resin immediately before it is discharged. Therefore, in this embodiment, a case will be described in which two properties are predicted: the viscosity of the resin immediately before it is discharged (near the die 62a in the crosshead 62) (hereinafter simply referred to as the viscosity of the resin) and the shrinkage rate of the molded product after extrusion molding (hereinafter simply referred to as the shrinkage rate). However, this is not limited to this, and the method can also be applied to other properties of the molded product, such as the tensile strength and elongation.

[0014] (Extruder 60) Before describing the characteristics prediction device 1 in detail, the extruder 60 will be described. Raw material pellets, which are the raw material for the molded body (here, the coating material), are fed into a hopper 61a. The pellets are mixed by a screw in the extruder body 61, and molten resin is extruded from a die 62a via a crosshead 62. The extruded resin coats the surface of a conductor 65 traveling along a travel line. The resin coating the surface of the conductor 65 is then air-cooled immediately after being extruded from the die 62a, and then water-cooled in a water bath 63. The resin extruded from the die 62a solidifies through the air- and water-cooling processes, becoming a molded body, which is a coating material. This results in an electric wire 66 having a coating material disposed around the conductor 65. The electric wire 66 that has passed through the water bath 63 is wound around a drum (not shown). The conductor 65 is, for example, a stranded conductor formed by twisting together multiple wires made of copper, a copper alloy, or the like. In this embodiment, PVC (polyvinyl chloride) appropriately mixed with fillers, additives, etc. is used as the resin material to be molded.

[0015] (Vibration sensor 11) The vibration sensor 11 is attached to the crosshead 62 of the extruder 60 and detects vibrations. By attaching the vibration sensor 11 to the crosshead 62, the device is less susceptible to the effects of crushing noises and the like that occur in the early stages of mixing (immediately after the resin material is added), and it becomes possible to accurately detect the behavior of the resin immediately before it is discharged.

[0016] In this embodiment, an acoustic emission sensor (hereinafter referred to as an AE sensor) 111 is used as the vibration sensor 11. The AE sensor 111 is a sensor that detects elastic waves generated by the release of elastic energy due to localized fractures inside the material during kneading. Here, the sampling frequency is set to 2 MHz and the gain is set to 20 dB. Note that, although the AE sensor 111 is used as the vibration sensor 11 in this embodiment, this is not limiting, and an acceleration sensor can be used, for example, when sensing at a lower frequency is desired. Also, various sensors may be used in combination, such as a combination of the AE sensor 111 and an acceleration sensor.

[0017] The AE sensor 111 needs to be attached to a flat surface, but the crosshead 62 generally does not have a flat surface on which the AE sensor 111 can be attached. Furthermore, the AE sensor 111 has a low heat resistance temperature of, for example, approximately 80°C, and therefore cannot be directly attached to the crosshead 62, which can reach high temperatures of 200°C or higher. Therefore, in this embodiment, a sensor installation jig 41 is provided to separate the AE sensor 111 from the crosshead 62, and the AE sensor 111 is attached to the crosshead 62 via the sensor installation jig 41. The sensor installation jig 41 integrally includes a cylindrical rod portion 42 that is threadedly fixed into a bolt hole (not shown) formed in the crosshead 62, and a flat base portion 43 provided at the tip of the rod portion 42. Here, the rod portion 42 and the base portion 43 are made of steel. The AE sensor 111 is attached to the base portion 43. Vibrations (sound waves, elastic waves) generated in the crosshead 62 are transmitted through the rod portion 42 and base portion 43 of the sensor installation jig 41 and are detected by the AE sensor 111. The length of the rod portion 42 may be set appropriately so that the temperature at the position of the AE sensor 111 is lower than the heat resistance temperature of the AE sensor 111. Although not shown, a fan or the like may be provided to cool the AE sensor 111.

[0018] (Arithmetic unit 12) The arithmetic device 12 has a control unit 2 and a storage unit 3. A display unit 4 is connected to the arithmetic device 12, and is configured to be able to display the estimation results of the physical properties of the composite material, various data, etc. on the display unit 4. The arithmetic device 12 is also provided with an input device 5 such as a keyboard, and various settings and the display contents of the display unit 4 can be operated by inputting data into the input device 5. The display unit 4 may be configured as a touch panel display so that it also serves as the input device 5. Furthermore, the display unit 4 and the input device 5 do not have to be connected to the arithmetic device 12 by wire, and may be connected wirelessly. In this case, the display unit 4 and the input device 5 may be, for example, a smartphone or a tablet.

[0019] The control unit 2 is equipped with a data acquisition processing unit 21, a preprocessing unit 22, a relationship derivation processing unit 23, a prediction processing unit 24, and a prediction result presentation processing unit 25. The data acquisition processing unit 21, the preprocessing unit 22, the relationship derivation processing unit 23, the prediction processing unit 24, and the prediction result presentation processing unit 25 are realized by appropriately combining a processing element such as a CPU, memories such as RAM and ROM, software, an interface, a storage device, etc. Details of each unit will be described later. The storage unit 3 is realized by a predetermined storage area of ​​a memory or storage device.

[0020] (Data acquisition processing unit 21) The data acquisition processing unit 21 receives the output from the vibration sensor 11 (AE sensor 111) during kneading, and performs data acquisition processing to store the received data in the storage unit 3 as measurement data 31. The input device 5 or the like may be configured to allow input of blending information of the materials being kneaded.

[0021] (Pre-processing unit 22) The pre-processing unit 22 performs pre-processing (pre-processing step) to convert the measurement data 31 into a spectrum. More specifically, the pre-processing unit 22 extracts the output of the vibration sensor 11 (AE sensor 111) for a predetermined period. In this embodiment, the output of the AE sensor 111 is extracted for 300 seconds, but the period for extracting the output of the AE sensor 111 can be set appropriately. However, if the period is too short, the periodicity is lost and the estimation accuracy decreases, so it is necessary to set the period to a certain length (a period that does not impair the periodicity). More specifically, for example, it is preferable to set the period to a length longer than the period required for one rotation of the screw of the extruder 60. In this embodiment, the screw is rotated at 7 rpm. In this case, it is desirable to set the period for extracting the output of the AE sensor 111 to at least 8.6 seconds (60 / 7 seconds) or more. Furthermore, it is desirable not to use the period from the start of kneading until the kneading becomes stable as the output of the AE sensor 111 is not stable during this period.

[0022] The pre-processing unit 22 performs frequency analysis on the extracted output of the AE sensor 111 to obtain an intensity spectrum. More specifically, as shown in FIG. 2, the pre-processing unit 22 performs fast Fourier transform (FFT) analysis on the extracted output of the AE sensor 111 (upper left in FIG. 2) to convert it into intensity data for each frequency (upper right in FIG. 2). Because the amount of data is still too large when only performing the fast Fourier transform, the pre-processing unit 22 divides the frequency into predetermined intervals, integrates the intensity distribution of the output of the AE sensor 111 for each interval, and creates a histogram to obtain an intensity spectrum (lower left in FIG. 2). In this embodiment, the number of data is further reduced by extracting only a predetermined frequency range from the obtained intensity spectrum and deleting data of unnecessary frequencies (lower right in FIG. 2). The number of data after reduction is preferably, for example, between 20 and 100. The frequency range to be used should be set based on a preliminary experiment to allow easy determination of whether the kneading state is normal or abnormal. This can be set appropriately depending on the material to be kneaded, the type of extruder 60, etc. The preprocessing unit 22 stores the determined intensity spectrum in the storage unit 3 as preprocessed data 32 .

[0023] In this embodiment, FFT analysis is performed every second to determine signal strength in 0.5 kHz increments for a 1 MHz frequency range, and the sum of signal strengths in 10 kHz increments (0 kHz or more and less than 10 kHz, 10 kHz or more and less than 20 kHz, ...) is calculated. This is performed 300 times (i.e., 300 seconds) to calculate the conditional average of the signal strength for each frequency band. The obtained conditional average value is used as the signal strength P for each frequency band in the intensity spectrum in the relationship derivation process and prediction process described below.

[0024] Here, frequency analysis was performed to acquire signal intensities over a wide range, such as 0 to 1 MHz. However, since the frequency range in which the effect is significant is determined to some extent depending on the characteristics of the object to be predicted, the preprocessing unit 22 may be configured to acquire only signal intensities within the required frequency range. As a result of studies by the present inventors, it has become clear that when the characteristics of the object to be predicted are the viscosity and shrinkage rate of a resin, the correlation between signal intensity and the characteristics becomes large at frequencies between 40 kHz and 60 kHz. Therefore, when the characteristics of the object to be predicted are the viscosity and shrinkage rate of a resin, it is advisable to acquire signal intensities within a frequency range that includes at least the range between 40 kHz and 60 kHz.

[0025] (Relationship derivation processing unit 23) The relationship derivation processor 23 performs a relationship derivation process to derive the relationship between the intensity spectrum (preprocessed data 32) obtained by the preprocessor 22 and the properties of the resin measured separately. In the relationship derivation process, prior to the real-time monitoring described above, a test production is performed under different extrusion molding conditions. The output of the AE sensor 111 obtained in the test production is then preprocessed to obtain an intensity spectrum, and the properties of the molded body obtained in the test production are determined. In this embodiment, the temperature from the downstream side of the extruder main body 61 to the die 62a is set to the same temperature, and the test production is performed by varying this temperature as a parameter (varying from 140°C to 175°C in 5°C increments). The screw rotation speed is set to 7 rpm, and the linear speed of the conductor 65 is set to a constant 11 m / min.

[0026] The viscosity of the resin immediately before discharge, which is the characteristic to be predicted, is difficult to measure directly. Therefore, in this embodiment, the resin used was measured at three temperatures, 160°C, 180°C, and 200°C, using a capillary rheometer, and the viscosity was converted into a function using the approximate formula (1) below.

number

[0027] The shrinkage rate of the molded body, which is the property to be predicted, was calculated by taking a sample of a specified size from the molded coating material, heat treating it at 80°C for 24 hours, measuring the dimensions, and then using the following formula (2).

number

[0028] The relationship derivation processing unit 23 calculates the relationship between the signal intensity in each frequency band of the intensity spectrum, i.e., the spectral total value in a predetermined frequency range of the intensity spectrum (here, a frequency range in 10 kHz increments), and the characteristic. For example, the relationship between the signal intensity from 0 kHz to less than 10 kHz and the viscosity of the resin, the relationship between the signal intensity from 10 kHz to less than 20 kHz and the viscosity of the resin, etc. are calculated. The relationship derivation processing unit 23 then calculates the correlation coefficient R for each relationship and stores the relationship with the largest correlation coefficient R in the storage unit 3 as relationship data 33. That is, the relationship data 33 is data representing a relational expression that shows the relationship between the total value of the signal intensity in a predetermined frequency range and the characteristic to be predicted. Examples of the relationship data 33 obtained by the relationship derivation processing are shown in FIGS. 4(a) and 4(b). FIG. 4(a) shows the relationship data 33 used to calculate the viscosity of the resin, and FIG. 4(b) shows the relationship data 33 used to calculate the shrinkage rate of a molded body. Both relationship data 33 have a correlation coefficient R of 0.96 or higher, confirming a high correlation.

[0029] (Prediction processing unit 24) The prediction processing unit 24 performs a prediction process (prediction step) to predict the resin characteristics using the relationship data 33 previously determined by the relationship derivation processing unit 23, i.e., the relationship between the spectrum total value in a predetermined frequency range of the intensity spectrum (in the case of FIGS. 4(a) and 4(b), the signal intensity in the frequency band of 40 kHz or more and less than 50 kHz) and the resin characteristics. In the prediction process, the intensity spectrum (signal intensity in the corresponding frequency band) determined by the preprocessing unit 22 is applied to the relationship data 33 to determine a predicted value of the resin characteristics. The determined predicted value is stored in the memory unit 3 as prediction data 34.

[0030] In this embodiment, by predicting properties such as the viscosity of the resin using the intensity spectrum obtained every 300 seconds by the pre-processing unit 22, it becomes possible to monitor changes in the resin properties in real time, and it becomes possible to detect abnormalities in the extrusion molding based on these changes in the resin properties. Therefore, although not shown, an abnormality detection unit may be further provided that detects abnormalities in the extrusion molding based on the predicted values ​​of the resin properties predicted by the prediction processing unit 24.

[0031] Furthermore, the prediction process does not need to be performed in real time simultaneously with the extrusion molding, but rather the prediction of properties such as shrinkage rate may be performed at an appropriate timing after the extrusion molding based on the measurement data 31 acquired during the extrusion molding.

[0032] (Prediction result presentation processing unit 25) The prediction result presentation processing unit 25 performs a prediction result presentation process to present prediction data 34, which is the prediction result obtained by the prediction processing unit 24. In the prediction result presentation process, for example, the prediction data 34 is displayed on the display device 4. Note that in the prediction result presentation process, data other than the prediction data 34, for example, appropriate information such as the measurement data 31 that is the prediction target, may also be presented.

[0033] (Characteristics prediction method) In the characteristic prediction method according to this embodiment, prior to prediction, it is necessary to first create relational data 33 by conducting test manufacturing and collecting past manufacturing data. The flow of this process is shown in FIG.

[0034] 5, first, in step S11, a resin material is fed into the extruder 60 and extrusion molding is performed to carry out test production, and measurement is also carried out by the AE sensor 111. The measurement results by the AE sensor 111 are stored in the memory unit 3 as measurement data 31.

[0035] Then, in step S12, the viscosity of the resin material used in extrusion molding is measured using a capillary rheometer. In step S13, coefficients C1 to C3 are obtained by fitting to the approximation of formula (1). Then, in step S14, the viscosity of the resin is obtained from the obtained approximation of formula (1) and the temperature during test production in step S11, and the process proceeds to step S19.

[0036] On the other hand, after step S11, a sample of the coating material, which is a molded body, is obtained in step S15, and is subjected to heat treatment at 80° C. for 24 hours in step S16. Thereafter, in step S17, the shrinkage rate is calculated based on the dimensional change before and after the heat treatment using equation (2), and the process proceeds to step S19.

[0037] On the other hand, after step S11, preprocessing is performed on the measurement data 31 in step S18. In the preprocessing, as shown in Fig. 6, in step S21, the output of the AE sensor 111 for a predetermined period (for example, one second) is extracted from the measurement data 31, in step S22 the extracted output of the AE sensor 111 is frequency analyzed, and in step S23 an intensity spectrum is obtained from the result of the frequency analysis. Thereafter, in step S24 the obtained intensity spectrum is stored in the storage unit 3 as preprocessed data 32. Thereafter, the process returns and proceeds to step S19 in Fig. 5.

[0038] In step S19, it is determined whether to end data collection. This determination can be made based on, for example, input from the input device 5. Then, in step S20, a relationship derivation process is performed. In the relationship derivation process, the relationship between the signal intensity of each frequency band of the intensity spectrum (preprocessed data 32) and the viscosity and shrinkage rate of the resin is obtained, and the correlation coefficient R is also obtained. The relationship between the signal intensity and the viscosity and shrinkage rate of the resin in the frequency band where the correlation coefficient R is largest is stored in the storage unit 3 as relationship data 33. Then, the process ends.

[0039] (Main routine) In the characteristic prediction method according to this embodiment, a control flow shown in Fig. 7 is executed. As shown in Fig. 7, first, in step S1, a data acquisition process is performed. In the data acquisition process, a data acquisition processing unit 21 acquires data from the vibration sensor 11 (AE sensor 111) and stores the data in the storage unit 3 as measurement data 31.

[0040] Thereafter, in step S2, preprocessing (see FIG. 6) is performed to obtain the intensity spectrum as preprocessed data 32. The preprocessing in step S2 may be performed in real time during extrusion molding, or may be performed after the extrusion molding is completed.

[0041] Thereafter, in step S3, a prediction process is performed. In the prediction process, the signal intensity of a predetermined frequency band of the intensity spectrum, which is the preprocessed data 32 to be predicted, is substituted into a relational expression, which is the relational data 33, to obtain predicted values ​​of the viscosity and shrinkage rate characteristics of the resin. The obtained predicted values ​​are stored in the storage unit 3 as prediction data 34.

[0042] Thereafter, in step S4, a prediction result presentation process is performed. In the prediction result presentation process, the prediction result presentation processing unit 25 displays the prediction data 34 obtained in step S3 on the display 4. Thereafter, the process ends.

[0043] (Actions and Effects of the Embodiments) As described above, the characteristic prediction device 1 according to this embodiment includes a vibration sensor 11 attached to the crosshead 62 of the extruder 60 and detecting vibrations, a pre-processing unit 22 that performs frequency analysis on the output of the vibration sensor 11 to obtain an intensity spectrum, and a prediction processing unit 24 that predicts the characteristics based on previously obtained relationship data 33 (the relationship between the characteristics and the spectral total value in a predetermined frequency range of the intensity spectrum) and the intensity spectrum obtained by the pre-processing unit 22 (pre-processed data 32).

[0044] By attaching the vibration sensor 11 (AE sensor 111) to the crosshead 62, the device is less susceptible to the effects of crushing sounds and the like that occur in the early stages of kneading, making it possible to accurately predict the properties of the resin being extruded. Furthermore, according to this embodiment, it is possible to predict the properties of the resin in real time based on the measured output of the vibration sensor 11, and if the viscosity of the resin becomes abnormal, for example, adjustments can be made quickly to improve yield and reduce manufacturing costs.

[0045] (Variation) In this embodiment, the case where the resin to be extruded is mainly PVC has been described, but the present invention is not limited to this and can be applied to various resins, including resins that do not contain fillers. Also, in this embodiment, the case where the covering material for the electric wire 66 is formed by extrusion molding has been described, but the present invention is not limited to this and can be, for example, a resin linear body or tube omitting the conductor wire 65.

[0046] (Summary of the embodiment) Next, the technical ideas grasped from the above-described embodiments will be described by using the reference numerals and the like in the embodiments. However, the reference numerals and the like in the following description do not limit the components in the claims to the members and the like specifically shown in the embodiments.

[0047] [1] A device for predicting the properties of a resin extruded by an extruder (60), comprising: a vibration sensor (11) attached to a crosshead (62) of the extruder (60) for detecting vibrations; a pre-processing unit (22) for performing frequency analysis on the output of the vibration sensor (11) to obtain an intensity spectrum; and a prediction processing unit (24) for predicting the properties based on the relationship between the properties and a spectral sum value of a predetermined frequency range of the intensity spectrum obtained in advance, and the intensity spectrum obtained by the pre-processing unit (22).

[0048] [2] The characteristic prediction device (1) according to [1], wherein the characteristic is the viscosity of the resin immediately before discharge or the shrinkage rate of the resin after the extrusion molding.

[0049] [3] The characteristic prediction device (1) according to [2], wherein the predetermined frequency range includes at least a range of 40 kHz to 60 kHz.

[0050] [4] The characteristic prediction device (1) described in [1], wherein the vibration sensor (11) is an acoustic emission sensor (111) and is attached to the crosshead (62) via a sensor installation jig (41) that separates the acoustic emission sensor (111) from the crosshead (62) and transmits vibrations from the crosshead (62) to the acoustic emission sensor (111).

[0051] [5] The characteristic prediction device (1) according to [1], further comprising an abnormality detection unit that detects abnormalities during the extrusion molding based on the predicted values ​​of the characteristics predicted by the prediction processing unit (24).

[0052] [6] A method for predicting the properties of a resin extruded by an extruder (60), the method comprising: a pre-processing step of using a vibration sensor (11) attached to a crosshead (62) of the extruder (60) to detect vibrations, and performing frequency analysis on the output of the vibration sensor (11) to obtain an intensity spectrum; and a prediction processing step of predicting the properties based on the relationship between the properties and a spectral sum value of a predetermined frequency range of the intensity spectrum obtained in advance, and the intensity spectrum obtained in the pre-processing step.

[0053] (Addendum) Although the embodiments of the present invention have been described above, the invention according to the claims is not limited to the above-described embodiments. It should be noted that not all of the combinations of features described in the embodiments are necessarily essential to the means for solving the problems of the invention. Furthermore, the present invention can be appropriately modified and implemented within the scope of its spirit. [Explanation of symbols]

[0054] 1...Characteristics prediction device 11...Vibration sensor 111...AE sensor 12...Arithmetic device 2...Control unit 21...Data acquisition processing unit 22...Pre-processing section 23...Relationship derivation processing unit 24...Prediction processing unit 25...Prediction result presentation processing unit 3...Storage section 31...Measurement data 32...Preprocessed data 33...Relationship data 34…Prediction data 41...Sensor installation jig 42...Rod section 43...Base 60...Extruder 62...Crosshead

Claims

1. An apparatus for predicting the properties of a resin extruded by an extruder, comprising: a vibration sensor attached to a crosshead of the extruder to detect vibrations; a preprocessing unit that performs frequency analysis on the output of the vibration sensor to obtain an intensity spectrum; a prediction processing unit that predicts the characteristic based on a relationship between the characteristic and a spectrum sum value of the intensity spectrum in a predetermined frequency range that has been obtained in advance, and the intensity spectrum obtained by the preprocessing unit, Characteristics prediction device.

2. The characteristic is the viscosity of the resin immediately before being discharged or the shrinkage rate of the resin after the extrusion molding. The characteristic prediction device according to claim 1 .

3. The predetermined frequency range includes at least a range of 40 kHz to 60 kHz. The characteristic prediction device according to claim 2 .

4. the vibration sensor is an acoustic emission sensor and is attached to the crosshead via a sensor installation jig that separates the acoustic emission sensor from the crosshead and transmits vibrations from the crosshead to the acoustic emission sensor; The characteristic prediction device according to claim 1 .

5. and an abnormality detection unit that detects an abnormality during the extrusion molding based on the predicted value of the characteristic predicted by the prediction processing unit. The characteristic prediction device according to claim 1 .

6. A method for predicting properties of a resin extruded by an extruder, comprising: A vibration sensor attached to the crosshead of the extruder is used to detect vibrations. a pre-processing step of performing frequency analysis on the output of the vibration sensor to obtain an intensity spectrum; a prediction processing step of predicting the characteristic based on a relationship between the characteristic and a spectrum sum value in a predetermined frequency range of the intensity spectrum, which has been obtained in advance, and the intensity spectrum obtained in the preprocessing step, Property prediction methods.

Images