The subject of study of this work was learning algorithm of neuro-fuzzy systems with different membership functions. In the prior works there are no published studies of such studies, making it difficult synthesis of neuro-fuzzy control system with new objects in the application of biotechnology, including technological agribusiness entities. A comparative analysis of learning algorithms of neuro-fuzzy system with different membership functions using the method of error back propagation and а hybrid method. For this we used a training sample that contains data of temperature and concentration of dissolved gas in the culture liquid: oxygen (pO2), carbon dioxide (pCO2) of a biotechnological process. It is shown that the hybrid method carries out training of a neural network for the number of periods is 23 times smaller than the algorithm back-propagation errors. The studies found that the two-sided Gaussian membership function provides the smallest learning error of the network δ equal of 3,28•10–3, compared to the other, giving the largest error of training the neural network δ=0,138. Therefore, the task of running the fermentation process effective is the use a hybrid method of education and two-sided Gaussian membership functions. According to the research, we can conclude that for the adaptation of neuro-fuzzy network ANFIS and fuzzy inference system Sugeno zero order to solve biotechnological process control tasks microbiological production efficiency is to use a hybrid method of education and bilateral Gaussian membership functions
The preconsolidation pressure of natural loamy soils was researched with the method of constant rate of loading (CRL) and the continuous measuring of de-formations at the apparatus AKP-6NM. The results of the rate of deformation analysis have been offered. The method of the preconsolidation pressure and the crumple pressure values has been suggested based on the rate of deformation analysis
In the article we have considered the self-balanced
potato sorting machine differing from existing designs
of self-balanced potato sorting machines with an oscillatory
electric drive. That drive uses a linear induction
motor. As the counterbalancing device, the method of
the duplicating mechanism is applied. The duplicating
mechanism is a specular reflection of the main working
body, and also participates in technological process.
Its application in the drive of machine allows not
only to increase efficiency of cleaning, drying and
sorting of potatoes, but also to increase reliability of
sorting installation that corresponds to the newest
tendencies of development of technology. We have
brought the mathematical model of the offered electric
drive of potato sorting machine, which is implemented
in the environment of object and visual modeling of
Matlab |Simulink|. The mathematical model allows
investigating influence of parameters of the linear induction
electric drive on parameters of oscillatory process
of working body in dynamics. The developed
technique of research and the created experimental
potato sorting machine with the linear induction drive
have confirmed theoretical researches. The main kinematic
sizes of machine and technical parameters of
individual nodes are determined. It is shown, that the
efficiency of inertial transportation can be increased
for 20% in comparison with the classical drive from
the motor of rotation. Results of research will allow to
realize energetically and technologically effective potato
sorting machines with the linear induction drive
To determine the mass fraction of linolenic acid in flax
seed oil the authors developed a rapid method based on
the pulse method of nuclear magnetic resonance with
the use of NMR analyzers low resolution. This method
has some significant advantages compared with the
known method on the basis of chromatographic
method for the determination of the mass fraction of
linolenic acid in the oil extracted from the seeds of
flax, namely, does not require additional sample
preparation (extracting oil by solvent), eliminates the
use of toxic chemicals, requires a much lower
investment of time to conduct a single analysis, highly
automated, and eliminates the influence of human
factor on the research results. The method is based on
the revealed correlation between the weighted average
time spin-spin relaxation of protons contained in the
triacylglycerols of the oil of the flax seed, and a mass
fraction of linolenic acid in the seed oil at a
temperature of 23ºc. In article results of research of
influence of temperature on the weighted average time
spin-spin relaxation of protons contained in the flax
seed oil. Installed directly proportional linear
relationship between the temperature of flax seeds and
measured the average time spin-spin relaxation of
protons oil contained in the flax seeds with the correlation coefficient is 0,930, the temperature change
of the sample of seeds at 1 ° C leads to the change of
the coefficient is 0.011 kt. The developed
methodaccuracy in the temperature range from 20 to
26 ºC
There were presented comparative analysis’ results of
criteria rice quality used in Russian and international
standards. Also there were shown summarized
statistical analysis’ results of rice quality that was
processed in 2013-2016 on rice mill plant at
“Krasnodarzernoprodukt” Company. The purpose of
research was to find factors that had influence on
combining rice bulk effectiveness on specialized
elevator. It was found that diapason of rice moisture
content was quite small. Red rice content was ranging
from 1.08 to 5.10 % with confidence interval of 1.98-
2.56 % that is evidence of small amounts of rice bulks
with extreme high or low red rice content. Broken rice
content varied in the wide range from 3.43 to 9.26 %
with confidence interval from 5.84 to 7.26 % that
shows necessity in length grading procedure applying
for broken kernels removing at the elevator unit.
Grain impurities content exceeded basic norm in 3-4
times that is a sign of low efficiency in the process of
separation. During the treatment of rice bulks
combined at the grain cleaning unit of rice plant
broken rice cereals output varied in wide range from 3
to 9 % during single shift that is seems to be connected
with kernels cracking increasing and rice moisture
decreasing. Moreover, rice bran output also varied in a
wide diapason from 10.2 to 17.0 % that could be
explained by big swing in rice moisture, glassy texture
and cracking of rice grains. According to our research
it was investigated that current rice quality control system that is a base instrument of rice bulks forming
for further processing is not quite efficient and as a
result there should be created and implemented new
methods and technical devices for quick and proper
grain control that could assure stable technological
regimes at different stages of rice treatment process
and that would provide the best quality and the biggest
output of rice cereals
The article shows that for the essential oil industry of
the Russian Federation to solve the problems of import
substitution it is urgent to develop effective technology
to extract fatty oil from the crushed and deprived of
ether by water vapor coriander fruits. The existing
technologies for obtaining fatty oils from deprived of
ether coriander were considered. It is shown, that at
small-scale enterprises it is most advisable to remove
fatty oil by screw pressing. A brief analysis of main
factors affecting the completeness of oil extraction in
screw presses was carried out. The conclusion about
the necessity of preparation of deprived of ether
coriander for extraction with the purpose of increasing
the yield of fatty oil was made. It is shown, that one of
the stages of such preparation is the removal of part of
the shell from deprived of ether coriander, which also
increases the nutritional value of the oilcake.
Fractional composition of deprived of ether coriander
after extraction of essential oil by water vapour and
drying was investigated by sieve analysis method. On
the basis of obtained data, prior to separating the shell
by pneumatic separation, it is recommended to
separate deprived of ether coriander into three
fractions by size – coarse (riddling from the sieve with
the diameter of 1.8 mm), medium (riddling from the
sieve with the diameter of 1 mm) and fine (undersize
from a sieve with a diameter of 1 mm). Component
composition of the coarse fraction of deprived of ether
coriander was studied. It was found, that the content of the shell in this fraction is 36%, which indicates the
possibility of its effective separation by pneumatic
separation
The low rate of planting and harvesting mechanization is a common drawback of the equipment for hydroponic green fodder (HGF) cultivation. The system, that has been developed, is represented by a rack construction with a tray positioned on each rack; a tray is the place where the processes of seed dispersal and growth and the takeout of grown HGF occur. A tray with a pipe sways in opposite directions over pipe axis, and, as a result, moving bulk of seeds covers the entire tray surface with a seed layer of uniform thickness. The mathematical model of dynamics in the “granular medium – tray” system has been developed; on the basis of this model, after execution of simulation modeling, practical angles of tray tilt were identified as being within the range of 20 – 30 degrees. Experimental investigations have been carried out on the barley of “Vakula” cultivar with seed moisture contents 10%, 14%, 16% and 18%. During the research, the main factors influencing the process under consideration were identified: the moisture content in barley seeds; initial angle of tray tilt; the number of tray swaying motions. It was determined that the increase of tray filling rate can be achieved by means of setting up the initial angle of its tilt from a reference level, which depends on the moisture content in the barley seeds used in the process of HGF production. This angle is 22 – 24 degrees for the dry barley seed with moisture content 10-14% and 25 – 27degrees for the barley seed with higher moisture content. The upper limit of moisture content in the seeds used in work process should be specified to achieve high quality of a final product. The barley seeds with moisture content 10-15% and the angles of tray tilt 23 – 25 degrees should be used during HGF production
This article presents the results of moisture-absorbing properties of the bedding material depending on the type, size and time of moisture absorption
The goal of the study is to research time characteristics
of pseudo-noise sequence acquisition method for
multicarrier communication systems of MC-DSCDMA,
that allows low hardware requirements and
able to operate in low quality channels. The way of
achieving this lies in using a mathematical model of
algorithm. The achieved results show suitability of
MC-DS-CDMA-signals optimization for pseudo-noise
sequence acquisition efficiency increase. The achieved
results can be applied in MC-DS-CDMA
communication systems development
The article describes the synthesis and verification of
statistical and system-cognitive models of the
influence of environmental factors on the quality of
life of the population of the region. This stage of the
ASC-analysis is performed in the system called
"Eidos". As a result, we have created and validated
(verification stage) all the specified systemic cognitive
models. It is expected that reliability for the models of
knowledge is sufficiently high for a given subject area,
that is why we can state the discovery of a dependence
of life expectancy and causes of death from
environmental conditions. Typically, knowledge
models are approximately 20% higher in accuracy than
statistical models, which operate on the principle of
positive pseudo-prediction. Making decisions based on
the model of Abs (matrix of absolute frequencies) is
not appropriate because of the different number of
instances of classes (generalized categories) and
dependence of the solutions of this amount. In the
model called Prc2 (conditional and unconditional
percentage distribution) the dependence of the model
values of the number of examples in classes has been
removed, but the accuracy of it is usually same low as
in the Abs. In addition, for decision-making based on
this model, one has to compare the values of
conditional and unconditional probabilities manually,
which is laborious and hardly possible for large
dimensional models. The knowledge model called
Inf3, based on a measure similar to the Chi-square, is
the result of the automated comparison of values of
conditional and unconditional probabilities presented
in the model of Prc1, which is similar to Prc2, and
usually has a fairly high accuracy, especially
considering the high complexity of the subject area,
which we simulated. Therefore, in accordance with the
technology of the ASC-analysis data conversion into
information, and afterwards - into knowledge, it is the
model of Inf3 which is planned to be used for the
solution of problems of identification, forecasting, decision-making and exploring the modeled subject
area, through the study of its models