Rotary-centrifugal shredder for forage preparation

. Existing grain shredders have disadvantages, the main of which are the following: high metal and energy consumption, uneven granulometric composition of the crushed product, high percentage of pulverized fraction, rapid wear of working bodies, product heating. To eliminate the disadvantages, the design of a rotary-centrifugal grain shredder is pro-posed. The presented design requires optimization of a number of structur-al and kinematic parameters. To solve this problem, the method of a multi-factor experiment was chosen. The feed of grain material, the rotor speed, the opening of the separating surface, and the number of knives on the inner and outer rings were taken as factors that varied at two levels. As optimization criteria, the plant performance, power consumption during grain grinding, and compliance of the resulting product with zootechnical requirements were considered. As a result of data array processing, adequate and reliable mathematical models were obtained. As a result of model analysis, the influence of factors on optimization criteria was established.


Introduction
Feed intended for animals must be prepared in accordance with zootechnical requirements. When preparing feed mechanically, grinding is mandatory.
Grinding is considered more energy-intensive and labor-intensive operation, which takes up more than 50% of the total labor costs in the preparation of compound feeds The development, research and creation of an optimal and efficient way to grind grain feed encourages many researchers to create new designs for shredders [1][2][3][4][5][6].
The resulting scientific task is to improve the most efficient machines for grain crops grinding, one of which is a centrifugal-rotary shredder.
The purpose of the research is to reduce the specific energy consumption of forage grain grinding and increase the uniformity of the granulometric composition of the finished product by improving the main design parameters of the centrifugal-rotary shredder.

Materials and methods
During the research, the following was used: forage grain, centrifugal-rotary shredder. The experiment was planned using the Box-Benkin matrix. The Excel data analysis package was used to analyze the data array and build mathematical models.

Results and Discussion
Of practical interest for grain grinding are the designs of shredders operating in the border area, combining the process of grinding by chipping and cutting with timely output of the finished product [1,4,7,8].
In the course of the study, barley of conditioned humidity was used as the initial grain material. Adjustment of the grain supply to the device x1 was carried out by changing the speed of the motor shaft of the blade feeder. The x2 rotor speed was regulated by changing the speed of the electric motor by a frequency converter. Opening of the separating surface x3 -by setting the appropriate size between the parallel planes of two adjacent knives on the outer ring 9 ( Fig. 1 and 2). The number of knives on the inner x4 and outer x5 ring -as a result of adding or removing them.  Analysis of zootechnical requirements for concentrated feed for various groups of animals allowed to identify several main criteria for evaluating the quality of the product. This is the grinding size, the percentage of particles in the grinding more than 3 mm and the presence of whole grains in the grinding. However, the results of sieving on a vibrating screen of selected samples showed that the most critical parameter for the resulting product is the content of particles in the grinding more than 3 mm. At the same time, the presence of whole grains and the grinding modulus were not considered, since no whole grains were observed in the sieving, and the modulus value corresponded to coarse grinding. It should be noted that the grinding modulus largely depended on the particle content of more than 3 mm, so, for example, their content varied in the range from 0.4 to 60.8% The data was analyzed in Excel. The results of data analysis in Excel are written as regression equations (1), (2), (3), (4) and are presented in table 2. Based on the results, it is concluded that the coefficients of the regression equations are significant at the 5% level; mathematical models (1), (2), (3), (4) due to the variation of the selected factors, they are statistically significant and cannot describe the processes occurring only at the 5% level. During the analysis, attention was paid to the normalized R-square, Fisher test results, tstatistics, and p-value (significance) for each coefficient of the equation [9][10]. The normalized R-square, unlike the R-square, can decrease when new explanatory variables are introduced into the model that do not significantly affect the dependent variable, while the R-square can increase when new explanatory variables are added, although this does not necessarily mean that the quality of the regression model improves. We give the regression equations and below the equation a table with the main indicators for each of the results y1, y2, y3, y4.
(1) After analyzing the regression models of power consumption (1), productivity (2), as well as the dust fraction content (3) and the content of particles in the finished product more than 3 mm (4), then using the Solution search tool, we obtained the following conclusions.
Optimal values of factors under the condition of minimizing energy consumption, increasing throughput, reducing the content of the dust fraction and the screening residue:

Conclusions
1. The most significant influence on the energy consumption y1 is exerted by factor x1, followed by factors x3 and x5. An increase in the grain feed x1 increases the power consumption y1, an increase in the opening value x3 and a decrease in the number of knives on the outer ring x5 leads to a decrease in the power consumption y1.
2. The most significant influence on the throughput y2 is exerted by factor x2, followed by factors x1 and x4. The increase in speed x2 and the number of blades on the inner ring x4 leads to a decrease in the throughput y2, the increase in grain feed x1 leads to increase of the throughput y2. 3. Factors x1 and x4, followed by factor x3, have the most significant influence on the percentage of the y3 dust fraction. The increase in the grain feed x1 and the opening value x3 increase the percentage of the dust fraction y3, the increase in the number of knives on the inner ring x4 leads to a decrease in the percentage of the dust fraction y3. 4. The most significant influence on the percentage of particles with a diameter of more than 3 mm y4 is exerted by factor x4, followed by factors x1 and x3. Increasing the number of knives on the inner ring x4, grain feed x1 and increase of the opening of the separating surface x3 increases the content of particles with a diameter of more than 3 mm. 5. The factor that significantly affects all results is the grain feed factor x1. There are no factors that could be excluded from consideration.