Obtaining source functions in Fourier transform (FT) on the base of some knowing frequency distributions are presented. The proposed approach allows to study the properties of the obtained results. This technique is used for calculation of some improper integrals as well.
This article considers non-parametric models based on feature space modeling in the context of machine learning. The main machine learning models, their advantages and disadvantages are analyzed. The term “non-parametric feature space modeling model” has been considered in detail and compared with other machine learning models. The advantages of these models are justified in comparison with other approaches. The paper contains an analysis that confirms the advantages of using non-parametric feature space modeling models in machine learning tasks.
This paper presents results of two different isentropic analysis types: conventional isentropic analysis which considers the whole steam turbine cylinder and segmental isentropic analysis which considers all cylinder parts (segments). In conventional isentropic analysis is obtained that isentropic efficiency of the analyzed turbine is 73.39%, what is in a range of expected isentropic efficiencies for such steam turbines (in the mechanical power range around 60 MW). Segmental isentropic analysis shows that the last two segments (fifth and sixth segment) of the analyzed turbine did not show proper operation (especially the fifth turbine segment which isentropic efficiency is unacceptably low and equal to 26.73% only). Such isentropic efficiency results, related to the fifth and sixth turbine segment, indicate highly problematic operation, or the most likely malfunction of at least some turbine stages in these segments. For the analyzed steam turbine can be recommended that it should be stopped as soon as possible and that turbine stages mounted in the last two segments should be checked, repaired or replaced.
One of the most critical consumable expenses in the production of wet wipes is the anvil and knife set used in the cutting unit. Blades that lose their cutting ability over time and become unable to perform their task pose a significant challenge. As wet wipes are essential products for maintaining hygiene and directly affect human health, even the slightest disruptions in production conditions are unacceptable. To ensure the quality and consistency of wet wipe production, it is vital to analyze the factors affecting the strenght of cutting blades and implement the necessary precautions. One of the key factors influencing the wear of cutting tools is the mechanical strength of the tool’s carrying unit. This paper explores the importance of cutting tools in wet wipe production, the role of mechanical strength in maintaining their effectiveness and application of mathematical modeling to design.
In this paper we will study an important part of the speech signal processing, the reconstruction of the signal. We will use the Newton interpolation method with a deterministic model and warped frequencies in order, maybe, to get a better reconstruction process of the signal. To determine the efficiency of the method we will use two criteria like: accuracy of reconstruction, noise stability. Then we will test this algorithm using the criteria that we proposed above versus the classic Yen’s algorithm. At the end of our experiment we will see that in noise stability our algorithm performs better but not in the reconstruction accuracy. In overall Yen’s algorithm performs slightly better.
The new platforms coming out on the social media space are faced from the beginning with a fierce contest. The competition for user attention is hard and the main players (Facebook, TikTok, Instagram, Reddit) have a huge advantage. To overcome these problems many of the newly lunched platform are making use of gamification strategy to keep the user engaged. The gamification strategy usually provides the user with a form of digital reputation but on some platform, it can be even a form of digital currency which can be exchanged for benefits or real products. To validate the user actions there is not always an automatic process, and the platforms are relaying on the community to cross-validate the authenticity and the value of the respective actions. The validation of the user actions can be automated but not always. When the validation cannot be automated, and the task requires human intervention the process is in his most part outsourced to the community and only a small number of the validation will be done by platform operators. The validation done by the platform operators will give credits to the users whose actions were validated. After certain amount of credit is given to a user, he became trusted, and his action can validate the actions of other users. The trusted user will act almost as a platform operator by validating other users’ actions. In this paper we simulate and analyse how the trust can be propagated into different social networks configuration. We start by defining a mathematical framework for modelling the users and the actions. After we introduce two different algorithms for propagating the trust in the community and finally for each configuration, we simulate the propagation of the trust and discuss the results.
Agent-based modeling proved to be a powerful tool for studying the complex multifactorial processes that take place in human population. In this paper we describe how this approach can be used to study the spread of airborne infections in a big city and the ways to control them. Agent-based modeling includes three main stages: a creating a synthetic population, simulation of disease spread in synthetic population during a fixed time period, and an analysis of the results. We created the population of 10 million agents and united them in a complex network according to their individual characteristics such as age, sex, marital status and occupation. In addition, each agent has a property that characterizes its state of health: susceptible, infected, and recovered. A susceptible agent can become infected if has an infected agent among its contacts and if an event of disease transmission occurs. Disease transmission is simulated as a random event with probability p which calculated for every pair susceptible-infected and depends on their individual characteristics and on the length of the contact. In so doing, the heterogeneity in a number of contacts and in resistance can be modelled. This approach was applied to model a dynamic of COVID-19 in Moscow during the period between October 2020 and December 2021.
A dose assessment of external neutron irradiation at high-energy accelerating complexes, where neutron energy can be in a wide range of 10-8 to 103 MeV, is important for staff radiation safety. In this case, Bonner multi-sphere spectrometers can be used to register the neutron flow. The work proposes a method for unfolding the neutron spectrum by Bonner spectrometer readings, based on the use of the detector sensitivity functions as the basic functions of decomposition of the spectrum under consideration. To find the decomposition coefficients, the system of integral Fredholm equations of the 1st kind is used. From a mathematical point of view, this problem is the illposed one and is solved numerically using the A.N. Tikhonov regularization method. According to the received energy spectrum of neutrons, an assessment of the dose of irradiation is made at the detection locations. The results of unfolding the neutron spectrum at the JINR Phasotron and the dose to its personnel are presented.
The present study aims to model and analyze the images of induced in vitro blood clots from healthy donors and patients with type 2 diabetes mellitus (T2DM). By using the BioFlux microfluidic system, blood clots are induced at varying shear rates, and for different fibrinogen concentrations: native and highly modified. The fiber diameter of blood clots is analyzed by the scanning electron microscope (SEM). The obtained images of blood clots are imported as input data into the Image J software environment, after which obtained results for the area, number, and fibrin fiber diameter of blood clots, are further processed in a program developed in IntelliJ IDEA. It is found that patients with T2DM at the native concentration of fibrinogen at all studied shear rates form more blood clots having a larger total area, in comparison with the control group. At the higher modified fibrinogen concentration with increasing shear rate, the group with T2DM forms a smaller number of blood clots with a larger area, compared to healthy donors. From the SEM images, it is found, that denser fibrin networks are formed with increased fibrinogen concentration, which contains numerous thick fibrin fibеrs in healthy and diabetic individuals.