JP2015214119A - Quality evaluation method for rubber kneaded product - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a quality evaluation method for a rubber kneaded product, capable of relatively simply grasping a degree of quality of abnormal products to normal products, and characteristics causally related to the occurrence of abnormal products.SOLUTION: Before a rubber kneaded product R is manufactured by mixing a raw material rubber and various compounding agents by a closed type rubber kneading machine 1, based on data on mixing characteristics related to mixing specifications of normal products of the rubber kneaded product R and data on material characteristics related to material specifications, a Mahalanobis distance Dis calculated, and a threshold C of the Mahalanobis distance which is a boundary between normal products and abnormal products is set. The Mahalanobis distance Dof the rubber kneaded products R for each batch is calculated, and then whether the rubber kneaded product R is a normal product or an abnormal product is determined in comparison between the calculated Mahalanobis distance Dand the threshold C. When it is determined to be an abnormal product, based on a correlation between the mixing characteristics and the material characteristics of the rubber kneaded product R, a causal relation between the mixing characteristics and the material characteristics is grasped.

Description

  The present invention relates to a method for evaluating the quality of a rubber kneaded product, and more specifically, a rubber kneaded product capable of relatively easily grasping the quality level of an abnormal product with respect to a normal product and the characteristics causally related to the occurrence of the abnormal product. It relates to the quality evaluation method.

  When manufacturing rubber products such as tires and rubber hoses, for example, raw rubber and various compounding materials such as carbon black are put into a sealed rubber kneader in a predetermined amount and kneaded. By this kneading, various blended materials are uniformly dispersed in the raw rubber, and a rubber kneaded material having an appropriate viscosity reduced to a certain viscosity is obtained. The closed rubber kneader includes two rotors juxtaposed in parallel in a chamber, and kneads a rubber kneaded material by rotating these rotors. The rubber kneaded product is kneaded by being rotated about the rotor shaft so that a shearing force is applied between the rotor and the inner wall surface of the chamber.

  By the way, the quality (specific gravity, tensile strength, elongation at break, 100% modulus, hardness, etc.) of the rubber kneaded product obtained by the mixing process using the closed rubber kneader varies somewhat from batch to batch. In general quality control, for example, standard values (control values) are individually set for each characteristic, and various characteristics of samples collected for each batch are compared with the standard values of each characteristic. Judging whether the product is normal or abnormal. For example, a method has been proposed in which each kneading information for each kneading batch is stored in a memory unit in association with a bar code, and quality is controlled by determining whether it is suitable for use based on the reading of the bar code (Patent Literature). 1).

  However, in the conventional quality control method, further analysis is necessary to grasp the reason why the quality of the abnormal product is inferior to that of the normal product and why the quality is inferior. Detailed analysis is not easy because such analysis requires advanced skills. Therefore, there is a demand for an evaluation method that can more easily grasp the quality level of an abnormal product with respect to a normal product and the characteristics that are causally related to the occurrence of the abnormal product.

JP-A-11-198153

  An object of the present invention is to provide a method for evaluating the quality of a rubber kneaded product, in which the quality level of an abnormal product with respect to a normal product and the characteristics causally related to the occurrence of the abnormal product can be grasped relatively easily.

  In order to achieve the above object, the rubber kneaded product quality evaluation method according to the present invention is a method for mixing a raw rubber and various compounding agents to produce a rubber kneaded product. Calculate the Mahalanobis distance based on the property data and the material property data related to the material specifications, set the threshold of the Mahalanobis distance that becomes the boundary between the normal product and the abnormal product, and knead the rubber for each batch. Calculate the Mahalanobis distance of the product and determine whether it is normal or abnormal based on the comparison between the calculated Mahalanobis distance and the above threshold, and if it is determined to be abnormal, the mixing characteristics and material characteristics of the rubber kneaded product The causal relationship between the mixing property and the material property is grasped based on the degree of correlation between the material property and the material property.

  In the present invention, when judging whether a rubber-kneaded product is a normal product or an abnormal product, it is not determined using only the data for each property, but between many properties (multivariables). The Mahalanobis distance, which is a comprehensive judgment scale based on correlation, is used. These characteristics include mixed characteristics (causal characteristics) and material characteristics (result characteristics). Therefore, it becomes easy to grasp the quality level of the abnormal product with respect to the normal product and the characteristics that are causally related to the occurrence of the abnormal product.

It is a longitudinal cross-sectional view which illustrates the inside of the closed type rubber kneader which knead | mixes the rubber kneaded material. It is a longitudinal cross-sectional view which illustrates the internal structure of the sealing-type rubber kneader of FIG. It is AA sectional drawing of FIG. It is a graph which shows the condition of the effect of each characteristic in samples 1, 2, and 5.

  The rubber kneaded product quality evaluation method of the present invention will be described below based on the embodiments shown in the drawings.

In the present invention, when manufacturing a rubber kneaded product R by mixing raw rubber and various compounding agents, data of mixing characteristics related to the mixing specifications acquired in advance for normal products, and material characteristics related to the material specifications The Mahalanobis distance D ² is calculated on the basis of the data and the Mahalanobis distance threshold C that is the boundary between the normal product and the abnormal product is set. Next, the Mahalanobis distance D ² of the sample collected from the rubber kneaded product R for each batch to be evaluated is calculated, and based on a comparison between the calculated Mahalanobis distance D ² and a preset threshold C, whether the product is normal or abnormal Judge whether it is a product.

  When it is determined that the product is abnormal, the causal relationship between the mixing characteristic and the material characteristic is grasped based on the degree of correlation between the mixing characteristic and the material characteristic of the rubber kneaded material R. The mixing characteristics are the characteristics of the causative system that causes the quality of the manufactured rubber kneaded material R. As mixing characteristics, kneading time T for one batch, total electric power P required for kneading one batch, temperature of rubber kneaded material R, and the like can be used.

  The kneading time T for one batch is, for example, a kneading step (S1), an oil mixing step (S2) for adding and mixing oil, a carbon mixing step (S3) for adding and mixing carbon-containing material, and mixing. It consists of the time of a plurality of steps of the uniformizing and kneading step (S4) for uniformly kneading the material. Of these, the oil mixing step (S2) is optional. The amount of electric power required for kneading in each step and the temperature of the rubber kneaded product R at the end of each step can also be used as mixing characteristics.

  The material properties are the resulting properties resulting from the quality of the rubber kneaded product R produced. The material characteristics include rubber characteristics when the rubber kneaded product R is not vulcanized and rubber characteristics after vulcanization. As the rubber characteristics when not vulcanized, for example, a scorch time and a minimum viscosity obtained using a Mooney viscometer can be used. As rubber characteristics after vulcanization, for example, specific gravity, tensile strength, elongation at break, 100% modulus and hardness can be used.

  In order to manufacture the rubber kneaded product R, a closed rubber kneader 1 (hereinafter referred to as a kneader 1) exemplified in FIGS. 1 to 3 is used. The kneader 1 has a material input port 3 in the middle of the casing 2 in the vertical direction, and has a chamber 7 for accommodating the rotor 8 and a material discharge port 4 at the lower part of the casing 3.

  The chamber 7 is provided with two rotors 8 juxtaposed in parallel. The two juxtaposed rotors 8 are rotationally driven in opposite directions around the respective rotor shafts 9 juxtaposed in parallel. The type of the rotor 8 is not particularly limited, and various types such as a tangential type and a meshing type can be adopted. Each of the rotors 8 is a two-blade rotor, but the number and shape of the blades are appropriately determined.

  A pressure weight 6 that moves up and down is provided on the top of the rotor 8. When the raw rubber and various compounding agents are charged into the casing 2, the pressure weight 6 is disposed at an upper standby position that does not interfere with the charging of the raw rubber and various compounding agents. After the raw rubber and various compounding agents are put into the casing 2, they move downward from the standby position, and are arranged at a position where the upper portion of the rotor 8 is covered and the chamber 7 is almost sealed. The rubber kneaded product R contains raw rubber and carbon black, and in addition, reinforcing agents other than carbon black, oil, anti-aging agents, processing aids, softeners, plasticizers, vulcanizing agents, vulcanization accelerators, vulcanization delays. Various compounding materials such as an agent are appropriately blended.

  When the rubber kneaded material R is kneaded, the material discharge port 4 provided at a position below the rotor 8 is closed by the discharge port flap 5. When the kneaded rubber kneaded product R is discharged from the material discharge port 4, the discharge port flap 5 moves to a standby position that does not interfere with the discharge of the rubber kneaded product R and opens the material discharge port 4. The structure of the pressure weight 6 and the discharge port flap 5 is not limited to the illustrated structure. A mixer having a so-called kneader structure may be used.

  For example, a drive motor or the like is employed as the rotor drive unit 10 that rotationally drives the rotor shaft 9. The rotor drive unit 10 is provided with a tachometer 10a and a power meter 10b. The tachometer 10 a detects the rotation speed N of the rotor 8 (rotor shaft 9), and the wattmeter 10 b detects the instantaneous power p required for rotational driving of the rotor 8.

  Data detected by the tachometer 10a and the wattmeter 10b is input to the arithmetic unit 11 configured by a computer or the like connected to the rotor drive unit 10. By integrating the instantaneous power p with the kneading time T, the total power P required for one batch of kneading is calculated. Data on the outer diameter D of the rotor 8 and the clearance H between the outer diameter position of the rotor 8 and the inner wall surface 7 a of the chamber 7 are also input to the arithmetic device 11. The data of the outer diameter D and the clearance H can also be mixed characteristic data.

  A sample is taken from the manufactured rubber kneaded material R, and the above-mentioned various characteristic data of the result system are measured.

The calculation method of the Mahalanobis distance D ² is as follows.
Data (number of data: j = 1, 2, 3,... N) of the characteristics to be evaluated (number of characteristics i = 1, 2, 3,... M) are collected as unit space data. Then, an average value m and a standard deviation σ of each characteristic are calculated. Next, normalization of the data x _i and _j is performed by the following equation (1).
X _i , _j = (x _i , _j −m _i ) / σ _i (1)

ここで、Ｓ（Ｘ_i,Ｘ_i）は下記のとおりである。

Next, correlation coefficients r _i and _j between the following characteristics are obtained.

Here, S (X _i , X _i ) is as follows.

Next, the following correlation matrix R is obtained.

Next, the following inverse matrix A of the correlation matrix R is obtained.

Next, the following Mahalanobis distance D ² is calculated from the inverse matrix A and the normalized unit space data.

The Mahalanobis distance D ² is calculated in advance using unit space data consisting of data of only normal products, and based on the Mahalanobis distance D ² and abnormal product data, an appropriate boundary between the normal product and the abnormal product is obtained. A threshold C for the Mahalanobis distance is set.

Then, the Mahalanobis distance D ^{2 of the} sample collected from the rubber kneaded product R for each batch is compared with the threshold value C, and if the Mahalanobis distance D ^{2 of the} sample is equal to or less than the threshold value C, it is determined as a normal product. If so, it is judged as an abnormal product. As described above, according to the present invention, when determining whether the product is normal or abnormal, it is not a method using only the data of each property for each property, but is based on the correlation between a large number of properties (multivariables). The Mahalanobis distance D ² is used as an objective judgment scale. Therefore, the degree of quality can be grasped by how much the Mahalanobis distance D ² of the sample of the kneading batch is deviated from the threshold value C. That is, it is possible to grasp the degree of quality of an abnormal product with respect to a normal product.

If it is determined that the product is abnormal, the correlation coefficient r _i , _j between each characteristic is calculated to check the degree of correlation. That is, when the correlation coefficient r _i , _j (degree of correlation) is high, it can be determined that an abnormal causal relationship exists between these characteristics. Therefore, it is possible to easily grasp the characteristics that are causally related to the occurrence of abnormal products. In order to prevent the occurrence of abnormal products, it is sufficient to take measures for the characteristics having a high degree of correlation. The effectiveness of each variable (factor) is evaluated, and for example, a main factor of abnormality is specified using a two-level orthogonal table.

  Using the present invention, quality evaluation was performed on seven types of samples shown in Table 1, which were kneaded in different batches using the same ingredients and the same composition and the kneader having the same structure as that illustrated in FIG. It was. Table 1 shows the data of the mixing characteristics and material characteristics of each sample. The kneading step includes a kneading step (S1), an oil mixing step (S2) for adding and mixing oil, a carbon mixing step (S3) for adding and mixing carbon-containing materials, and a uniform kneading of the mixed materials. This was performed in 4 steps of the kneading step (S4).

The data on the mixing characteristics are the kneading time T for one batch, the temperature of the kneaded product at the end of each step of S1 to S4, the total electric power P required for kneading the batch, and the electric energy required for each step of S2 to S4. Was used. Material property data include Mooney scorch time (MS) and minimum viscosity (Vm) as rubber properties when unvulcanized, and specific gravity, tensile strength (TB), elongation at break (EB), 100 after rubber properties. % Modulus (M100) and hardness were used. The Mahalanobis distance D ² of the normal product is calculated in advance, and the threshold value C that is the boundary between the normal product and the abnormal product is set to a numerical value 10. Table 2 shows the determination results of whether each sample is a normal product or an abnormal product. Samples 1, 2, and 5 that exceed the threshold C value of 10 are determined as abnormal products.

Each characteristic data is obtained by the following measurement method.
MS, Vm: Measured according to the method defined in JIS K6300: 2013.
Specific gravity: Measured according to the method defined in JIS K6268: 1998.
TB, EB, M100: Measured according to the method defined in JIS K6251: 2010.
Hardness: Measured according to the method defined in JIS K6253-3: 2012.

  Next, Table 3 shows the degree of correlation (degree of effect) of each characteristic of the sample determined to be an abnormal product.

  FIG. 4 is a graph showing the degree of effect of each characteristic in Samples 1, 2, and 5 shown in Table 3. Referring to FIG. 4, it can be seen that in Sample 1, the temperature at the end of S1 is high, the amount of power in S4 is large, and as a result, TB and M100 are high. In sample 2, the kneading time T is short, the amount of power in S2 and S3 is small, and as a result, it can be seen that M100 is low. In sample 5, the total amount of power is small, and as a result, it can be seen that Vm is high.

DESCRIPTION OF SYMBOLS 1 Sealed rubber kneading machine 2 Casing 3 Material inlet 4 Material outlet 5 Discharge flap 6 Pressurization weight 7 Chamber 7a Inner wall surface 8 Rotor 9 Rotor shaft 10 Rotor drive part 10a Power meter 10b Tachometer 11 Arithmetic unit R Rubber kneading object

Claims (3)

  When mixing raw rubber and various compounding agents to produce a rubber kneaded product, based on data on mixing characteristics related to the mixing specifications of normal rubber kneaded materials and data on material characteristics related to material specifications Calculate the Mahalanobis distance, set the threshold of the Mahalanobis distance that becomes the boundary between the normal product and the abnormal product, calculate the Mahalanobis distance of the rubber kneaded material for each batch, and calculate the Mahalanobis distance and the threshold value. Based on the comparison, it is determined whether the product is normal or abnormal, and when it is determined that the product is abnormal, based on the degree of correlation between the mixing characteristics of the rubber kneaded material and the material characteristics, A method for evaluating the quality of a rubber kneaded product, characterized by grasping a causal relationship.   The mixing characteristics include one batch of kneading time, rubber temperature, and total electric power required for kneading, and the material characteristics include unvulcanized rubber characteristics and rubber characteristics after vulcanization of the rubber kneaded product. The quality evaluation method of the rubber kneaded material of Claim 1.   The kneading time is composed of at least a time of a plurality of steps of a kneading step, a carbon mixing step of adding and mixing carbon, and a homogenizing kneading step, and the rubber temperature at the end of the step in these steps is the mixing temperature. The amount of electric power required for kneading in the plurality of steps is included in the mixing characteristics, the rubber characteristics when unvulcanized include scorch time and minimum viscosity, and the rubber after vulcanization. The method for evaluating the quality of a rubber kneaded product according to claim 2, wherein the characteristics include specific gravity, tensile strength, elongation at break, 100% modulus, and hardness.

Images