JP7498396B2 - Rubber material mixing process control system and rubber material mixing method - Google Patents

Description

The present invention relates to a rubber material mixing process control system and a rubber material mixing method, and more specifically, to a rubber material mixing process control system that can grasp with high accuracy the actual mixing state of the rubber material during the mixing process using a kneading roll device, and a rubber material mixing method using this mixing process control system.

Various rubber products such as tires and conveyor belts are manufactured by vulcanizing unvulcanized molded products. To mold these molded products, kneaded rubber is used, which is a rubber material made of raw rubber and various compounding agents. For example, rubber material kneaded in a Banbury mixer (closed mixer) is repeatedly kneaded in a kneading device called an undermixer to produce kneaded rubber of specified specifications (see, for example, Patent Document 1).

An open-type kneading roll device is known as this type of kneading device (see, for example, Patent Document 2). The kneading time in the kneading process using the kneading roll device is set according to the specified specifications of the kneaded rubber to be produced. If this kneading time is too short, the dispersion of the various compounding ingredients in the raw rubber may be insufficient, making it impossible to ensure a kneaded rubber of the specified specifications. On the other hand, if the kneading time is too long, productivity decreases. Generally, this kneading time is managed by the operating time of the kneading roll device.

However, the actual mixing time of the rubber material is the time from when the rubber material is wrapped around the kneading roll device in a preset normal state and kneading begins to when kneading ends. Some preparation time is required before kneading of the rubber material can begin, such as operating the kneading roll device and winding the rubber material in the normal state. Since this preparation time can vary for each kneading batch, the actual kneading time of the rubber material cannot be accurately grasped if only the operating time of the kneading roll device is used as an indicator. Therefore, there is room for improvement in order to grasp with greater accuracy the actual kneaded state of the rubber material during the kneading process by the kneading roll device.

JP 2016-43486 A JP 2005-66877 A

The object of the present invention is to provide a rubber material mixing process control system that can grasp with higher accuracy the actual mixing state of the rubber material during the mixing process using a mixing roll device, and a rubber material mixing method using this mixing control system.

In order to achieve the above object, the rubber material kneading process management system of the present invention is characterized in that it comprises a camera device that acquires image data of a predetermined position of a kneading roll device, a memory unit that inputs and stores the image data, operation information of the kneading roll device, and unique information of the rubber material kneaded by the kneading roll device, and a calculation unit communicably connected to the memory unit, and is configured such that for each kneading batch of the rubber material, the calculation unit determines whether the rubber material is wound around the kneading roll device in a preset normal state based on the image data, and the kneading time of the rubber material by the kneading roll device is calculated using the determination result and the operation information, and the calculated kneading time is associated with the operation information and the unique information and stored in the memory unit.

The rubber material mixing method of the present invention is characterized in that the rubber material is mixed with the mixing roll device using the rubber material mixing process management system described above to produce mixed rubber of predetermined specifications.

According to the rubber material kneading process management system of the present invention, the calculation unit determines whether the rubber material is wound around the kneading roll device in a preset normal state based on the image data. Therefore, the calculation unit can more accurately calculate the kneading time of the rubber material by the kneading roll device using this determination result and the operation information. Then, by storing the calculated kneading time in the memory unit in association with the operation information and the unique information, it becomes possible to grasp with higher accuracy the actual kneading state of the rubber material in the kneading process by the kneading roll device for each kneading batch.

The rubber material kneading method of the present invention allows the actual kneading state of the rubber material during the kneading process to be grasped with greater precision, which is advantageous for producing kneaded rubber with preset specifications.

1 is an explanatory diagram illustrating an embodiment of a rubber material kneading process control system of the present invention installed in a kneading roll apparatus, as viewed from the side of the kneading roll apparatus. FIG. FIG. 2 is an explanatory diagram illustrating the kneading roll device of FIG. 1 in a plan view. 2 is an explanatory diagram illustrating a state in which a rubber material is properly wound around the kneading roll device of FIG. 1 and preparation for a kneading step is completed; FIG. FIG. 4 is an explanatory diagram illustrating the kneading roll device of FIG. 3 in a plan view. 4 is an explanatory diagram illustrating a state in which a rubber material is kneaded by the kneading roll device of FIG. 3 . FIG. 6 is an explanatory diagram illustrating the kneading roll device of FIG. 5 in a plan view. FIG. 4 is an explanatory diagram illustrating a side view of the kneading roll device from which a produced kneaded rubber has been removed. FIG. FIG. 2 is an explanatory diagram illustrating an example of image data acquired by a management system. FIG. 2 is an explanatory diagram illustrating an example of a database stored in a storage unit. 10 is an explanatory diagram illustrating an example of image data showing that a rubber material is wound around a kneading roll device in a normal state; FIG. 1 is an explanatory diagram illustrating an example of image data showing that a rubber material is wound around a kneading roll device in an irregular state; FIG. 11 is an explanatory diagram illustrating image data showing a state in which nothing is wound around the kneading roll device.

The rubber material kneading process management system and rubber material kneading method of the present invention will be described below based on the embodiment shown in the figure.

The embodiment of the rubber material kneading process management system 1 (hereinafter referred to as the management system 1) of the present invention illustrated in Figures 1 and 2 includes a camera device 2, a memory unit 3 to which image data D acquired by the camera device 2 is input and stored, and a calculation unit 4 communicably connected to the memory unit 3. In this embodiment, the management system 1 further includes a temperature sensor 5 and a monitor 6.

This management system 1 is installed in an open-type kneading roll device 7. The kneading roll device 7 is disposed, for example, below a Banbury mixer. The kneading roll device 7 has a pair of kneading rolls 8 and a blender roll 9 disposed above the pair of kneading rolls 8. The kneading rolls 8 and the blender roll 9 are arranged in parallel. The kneading rolls 8 rotate in opposite directions. The blender roll 9 is provided with a traverse section 9a that moves the outer circumferential surface of the roll in the roll width direction. The kneading roll device 7 is controlled by a kneading control device 10.

The rubber material R mixed by the Banbury mixer is fed between a pair of mixing rolls 8 and passes through it, after which the rubber material R is passed around a blender roll 9 and fed again between the pair of mixing rolls 8. In this way, the rubber material R is repeatedly mixed by the mixing roll device 7 to produce mixed rubber Rm of the specified specifications.

The camera device 2 acquires image data D of a predetermined position P of the kneading roll device 7. This predetermined position P is set to a position where it can be confirmed from the acquired image data D that the rubber material R is wound around the kneading roll device 7 in a preset normal state. This normal state is a state in which the rubber material R is wrapped around a pair of kneading rolls 8 and a blender roll 9 so that it can be kneaded normally by the kneading roll device 7 without any problems. The predetermined position P is set, for example, to the center of the width of the blender roll 9 (a range of about 50% to 70% of the length of the blender roll 9 at the center of the width). In this embodiment, the area between the pair of traverse portions 9a at the center of the width of the blender roll 9 is set to the predetermined position P.

A digital camera (CCD camera) capable of acquiring still or video image data D is used as the camera device 2. In this embodiment, the camera device 2 includes a front camera device 2A that acquires image data D of a predetermined position P in a front view of the kneading roll device 7, and a side camera device 2B that acquires image data D of a predetermined position P in a side view of the kneading roll device 7. The number of front camera devices 2A and side camera devices 2B is not limited to one, and multiple devices can be provided. The side camera device 2B can be provided as desired, and the device can also be configured to have only the front camera device 2A.

Various computers can be used as the memory unit 3 and the calculation unit 4. Therefore, the memory unit 3 is a computer memory, and the calculation unit 4 is a CPU. The memory unit 3 and the calculation unit 4 are communicatively connected to the temperature sensor 5, the monitor 6, and the kneading control device 10.

The temperature sensor 5 detects the external environmental temperature (temperature of the installation location) of the kneading roll device 2. The detected temperature data detected by the temperature sensor 5 is input and stored in the memory unit 3. The temperature sensor 5 can be installed as desired.

The monitor 6 displays the data stored in the memory unit 3 and the data calculated and computed by the calculation unit 4. The memory unit 3 and the calculation unit 4 can be connected separately to a general monitor for use, so a dedicated monitor 6 that is constantly connected to the memory unit 3 and the calculation unit 4 can be provided as an option.

The kneading control device 10 inputs unique information F of the rubber material R for each kneading batch (data such as batch number, specifications of raw rubber and compounding ingredients, set kneading time, etc.) to the memory unit 3 and stores it in the memory unit 3. In addition, for each kneading batch, the kneading control device 10 inputs operation information M of the kneading roll device 7 (data such as operation time of the kneading roll device 7, rotation speed of the kneading roll 8, movement speed and movement range of the traverse section 9a, etc.) to the memory unit 3 and stores it in the memory unit 3.

The calculation unit 4 judges whether the rubber material R is wound around the kneading roll device 7 in the correct state based on the image data D. To make this judgment, for example, the calculation unit 4 stores the allowable range of the color density at the specified position P when the rubber material R is wound around the kneading roll device 7 in the correct state. In the correct state, the rubber material R is wound around the specified position P of the blender roll 9, so the color of the specified position P is not the color of the outer surface of the blender roll 9 but the color of the rubber material R. Therefore, the allowable range of the color density at the specified position P is determined based on the color density of the rubber material R. Then, if the density at the specified position P of the image data D is within the allowable range, it is judged that the rubber material R is wound in the correct state. If this density is outside the allowable range, it is judged that the rubber material R is not wound in the correct state.

In this way, based on the density (shade) of the predetermined position P in the image data D, it is determined whether the rubber material R is properly wrapped around the kneading roll device 7. When using the color shade of the predetermined position P, the color of the outer surface of the blender roll 9 needs to have a large difference in shade from the color of the rubber material R. Therefore, the color of the outer surface of the blender roll 9 is a light color such as silver or white if the rubber material R is black-based, and a dark color such as black-based if the rubber material R is white-based.

Next, an example of the procedure for the rubber material kneading method of the present invention using this management system 1 will be described.

The rubber material R that has been mixed in a Banbury mixer is fed into the mixing roll device 7. This rubber material R is in a state where the various compounding ingredients are not sufficiently dispersed in the raw rubber.

In the kneading roll device 7, the kneading roll device 7 is operated in preparation for the kneading process, and as shown in Figures 3 and 4, the rubber material R is inserted between a pair of kneading rolls 8, passed through the kneading rolls, and wrapped around the blender roll 9. When the rubber material R is wrapped around the blender roll 9 in this manner, preparation for the kneading process is complete.

Then, the kneading roll device 7 starts operating, starting the kneading process illustrated in Figures 5 and 6. In the kneading process, as illustrated in Figure 5, the rubber material R is fed between a pair of kneading rolls 8 in a ring-shaped state and passes through it, then it is wrapped around a blender roll 9 and fed again between the pair of kneading rolls 8. In this way, the rubber material R is repeatedly kneaded by the kneading roll device 7 for a set time, and kneaded rubber Rm of the specified specifications is produced. During the kneading process, the traverse parts 9a of the blender roll 9 repeatedly move back and forth in the roll width direction while maintaining a constant distance between each other on the outer circumferential surface of the roll, promoting the dispersion of various compounding agents in the rubber material R.

As shown in FIG. 7, the manufactured kneaded rubber Rm is removed from the kneading roll device 7 and used as a component for molding a molded product in a separate process. The molded product is then vulcanized to complete a rubber product. Examples of rubber products include tires, conveyor belts, and fenders.

In the kneading process by the kneading roll device 7, as shown in FIG. 8, for each kneading batch of rubber material R, image data D of a predetermined position P is acquired by the camera device 2 (front camera device 2A). The image data D acquired by the side camera device 2B is an enlarged image of the predetermined position P shown in FIG. 5. Then, based on the image data D, the calculation unit 4 determines whether the rubber material R is wound around the kneading roll device 7 in the correct state. Using this determination result and the operation information M, the calculation unit 4 calculates the actual kneading time T of the rubber material R by the kneading roll device 7.

Specifically, it is determined that the rubber material R is wound in the correct state based on the image data D, and the time during which the kneading roll device 7 is operating is calculated as the actual kneading time T of the rubber material R. The calculated kneading time T is associated with the operation information M and the unique information F and stored in the memory unit 3.

A database as shown in FIG. 9 is constructed in the memory unit 3, and unique information F, operation information M, image data D, detected temperature data, and calculated kneading time T are stored in association with each other for each kneading batch. The operation information M, image data D, and detected temperature data are stored in chronological order. Therefore, by inputting the unique information F (batch number) of a desired kneading batch to the calculation unit 4, the kneading time T of the rubber material R of that batch number stored in the memory unit 3 can be displayed and understood on the monitor 6. Furthermore, other unique information F, operation information M, image data D, and detected temperature data of the rubber material R of that batch number can also be displayed and understood on the monitor 6.

In this way, the management system 1 determines whether the rubber material R is wound around the kneading roll device 7 in the correct state based on the image data D, so the actual kneading time of the rubber material R by the kneading roll device 7 can be calculated more accurately. Then, by storing the calculated kneading time T in the memory unit 3 in association with the operation information M and the unique information F, the actual kneading state of the rubber material R during the kneading process by the kneading roll device 7 can be grasped with higher accuracy for each kneading batch.

The actual mixing state of the rubber material R can be accurately grasped, so that mixing batches that are insufficiently mixed can be identified. Furthermore, by analyzing the various data described above, the cause of the insufficient mixing can be identified, and the mixing conditions for obtaining mixed rubber Rm according to the specified specifications can also be determined. By feeding back the results of this analysis and performing the kneading process, it is advantageous to stably obtain mixed rubber Rm in which the various compounding ingredients are more evenly dispersed.

By having a temperature sensor 5 as in this embodiment, the external environmental temperature during the kneading process for each kneading batch can be grasped. Therefore, the influence of the external environmental temperature on the characteristics of the manufactured kneaded rubber Rm can be analyzed in more detail for each kneading batch.

When the camera device 2 includes a front camera device 2A and a side camera device 2B, it is possible to determine whether the rubber material R is wound around the kneading roll device 7 in the correct state using not only the image data D acquired by the front camera device 2A, but also the image data D acquired by the side camera device 2B. This makes it even more advantageous to accurately grasp the actual kneading time T of the rubber material R in this kneading process.

In the previous embodiment, whether the rubber material R is properly wound around the kneading roll device 7 is determined based on the shade of color at a specific position P in the image data D, but it can also be determined using other techniques, such as AI technology.

For example, a prediction model PM that has been machine-learned using various image data D as training data can be used to determine whether the rubber material R is wound around the kneading roll device 7 in the normal state. In this configuration, the prediction model PM that has been machine-learned using training data including normal state image data D1 in which the rubber material R is wound around the kneading roll device 7 in the normal state as illustrated in FIG. 10, non-normal state image data D2 in which the rubber material R is wound around the kneading roll device 7 in a state other than the normal state as illustrated in FIG. 11, and no-rubber state image data D3 in which nothing is wound around the kneading roll device 7 as illustrated in FIG. 12 is stored in the memory unit 3.

This prediction model PM is a computer program for predicting whether the rubber material R is properly wound around the kneading roll device 7, based on image data D of a predetermined position P input to the memory unit 3. In this embodiment, the predetermined position P from which the image data D is acquired by the camera device 2 is the kneading roll 8, the blender roll 9, and the area between them.

The image data D1, D2, and D3 differ in the range of the kneading roll device 7 covered by the rubber material R (the position of the rubber material R), the size, and the shape (form) of the rubber material R. Therefore, by using the image data D1, D2, and D3 as training data, the artificial intelligence (AI) can learn the differences in the kneading roll device 7 between the case where the rubber material R is wrapped around the device in the normal state and the case where the rubber material R is wrapped around the device in a non-normal state (the case where the rubber material R is wrapped around the device in an abnormal state and the case where nothing is wrapped around the device), and the characteristics of each case, thereby generating the above-mentioned prediction model PM. Examples of machine learning methods using the memory unit 3 include deep learning, neural networks, regression, clustering, and pattern matching.

In the kneading process by the kneading roll device 7, for each kneading batch, the calculation unit 4 judges whether the rubber material R is wound around the kneading roll device 7 in the correct state based on the image data D acquired by the camera device 2 using the prediction model PM. The actual kneading time T of the rubber material R by the kneading roll device 7 is calculated using this judgment result and the operation information M, as in the previous embodiment. Also, the calculated kneading time T is associated with the operation information M and the unique information F and stored in the memory unit 3, as in the previous embodiment. In order to improve the prediction accuracy of the prediction model PM, it is preferable to store as many and diverse image data D1, D2, and D3 as possible in the memory unit 3 and perform machine learning.

Reference Signs List 1 Management system 2 Camera device 2A Front camera device 2B Side camera device 3 Memory unit 4 Calculation unit 5 Temperature sensor 6 Monitor 7 Kneading roll device 8 Kneading roll 9 Blender roll 9a Traverse unit 10 Kneading control device R Rubber material Rm Kneaded rubber

Claims (6)

The system includes a camera device that acquires image data of a predetermined position of a kneading roll device, a memory unit that inputs and stores the image data, operation information of the kneading roll device, and information specific to a rubber material kneaded by the kneading roll device, and a calculation unit that is communicatively connected to the memory unit,
A rubber material kneading process management system characterized in that, for each kneading batch of the rubber material, the calculation unit determines whether the rubber material is wound around the kneading roll device in a preset normal state based on the image data, and the kneading time of the rubber material by the kneading roll device is calculated using this determination result and the operation information, and this calculated kneading time is associated with the operation information and the unique information and stored in the memory unit. The rubber material kneading process management system according to claim 1, wherein it is determined whether the rubber material is wound around the kneading roll device in the correct state based on the color shading at the specified position of the image data. The rubber material kneading process management system according to claim 1, wherein the storage unit stores a prediction model that has been machine-trained using as training data image data of a normal state in which the rubber material is wound around the kneading roll device in the normal state, image data of an abnormal state in which the rubber material is wound around the kneading roll device in a state other than the normal state, and image data of a rubber-free state in which nothing is wound around the kneading roll device, and the calculation unit uses the prediction model to determine whether the rubber material is wound around the kneading roll device in the normal state based on the image data acquired by the camera device. The rubber material kneading process management system according to any one of claims 1 to 3, comprising a front camera device that acquires the image data of the predetermined position when viewed from the front of the kneading roll device, and a side camera device that acquires the image data of the predetermined position when viewed from the side of the kneading roll device, as the camera device. A rubber material kneading process management system according to any one of claims 1 to 4, which has a temperature sensor that detects the external environmental temperature of the kneading roll device, and the detected temperature data detected by the temperature sensor is stored in the memory unit in association with the operation information and the unique information together with the calculated kneading time. A method for kneading rubber material, characterized in that the rubber material is kneaded by the kneading roll device using the rubber material kneading process management system described in any one of claims 1 to 5, thereby producing kneaded rubber of predetermined specifications that have been set in advance.

Images