Pavel Sukhoi State Technical University of Gomel

The dynamics of a biochemical system is obviously subjected to universal laws of thermodynamics, but analytical forms of the criteria and constraints having been deduced up to date and findable in the literature are to some extent arguable. The paper examines the thermodynamic constraints on chemical reaction pathways and is aimed at fi nding the ways to boost equilibrium-limited reactions under mild conditions by means of blurring the equilibrium or circumventing it. In particular, herein we deal with the thermodynamic analysis of reactions that proceed with exhibiting either the multiplicity of equilibrium states (partial equilibria) or the total absence of an equilibrium. The multiplicity of equilibrium states refers to oscillatory reactions. Thermochemical regularities of such reactions are revealed, and it is shown that they are not determined by positive or negative feedback. The total absence of an equilibrium is inherent in conjugate reactions, i.e. in couples of chemical reactions, one of which (namely, the induced reaction) is impossible in the absence of the other one (namely, the inducing reaction) because of being thermodynamically prohibited. To tackle the challenge of providing thermodynamic control of such reactions we focus on exergetic restrictions and hence deduce general conditions for inducing a heterogeneous homophase reaction.

Effective management of a water supply system amid urban infrastructure growth requires considering numerous factors that determine its technological and energy operational modes. Energy consumption in water intake structures can be optimized through adaptive activation of pumping stations, factoring in equipment timing and composition. A key aspect of implementing such measures is accurately forecasting water consumption dynamics. Most studies analyze the consequences of water consumption factors but overlook their potential for rapid acquisition in building predictive models or their systematization for use in water supply management systems. In contrast, this study aims to identify statistically significant factors that can be rapidly obtained and applied in machine learning models to forecast water consumption. It uses data on the water supply regime of Gomel, a major industrial center in Belarus with over 500,000 residents, from 2017 to 2023. Factors considered include air temperature, precipitation, and temporal parameters, such as hours of the day and months of the year, examining their influence on water consumption dynamics and energy costs within the system. The methodological framework employs correlation and regression analyses to assess relationships among factors, using smoothing (moving average), filtering, and truncation by temperature thresholds, ANOVA, and Tukey’s test to process data and evaluate differences among groups. Results indicate that temperatures above 25 °C increase water consumption by 15.8

Compositions and a methodology for manufacturing composite materials for remote photoluminescent converters of LED radiation based on quartz glass and ultrafine powders of yttrium–aluminum garnet activated by cerium ions (Y3Al5O12:Ce3+) are developed. The conditions for the synthesis of garnet by the method of thermochemical reactions (combustion) of nitrate salts in citric acid are optimized, and agglomerated powders with primary particle sizes of 0.8–12.3 μm are obtained for the reaction with a deficiency of acid and in the range from 49 to 207 nm for synthesis under conditions of excess citric acid. The physicochemical, structural, and spectral-luminescent characteristics of powdered samples and quartz glass–low-melting glass–phosphor composites formed at low sintering temperatures (650–700°C) are studied. It is shown that when the composite plate is illuminated with a blue LED (λmax = 455 nm), a spot of white light is visually recorded, formed from a combination of scattered blue radiation and yellow radiation excited in YAG:Ce3+ particles (λmax = 560 nm).

This paper explores the application of machine learning methods for recognizing automobile light signals to enhance smart traffic light systems. For vehicle detection in video footage, the Keras library was employed along with the RetinaNet neural network architecture [1]. The YOLOv8 architecture was used for identifying the status of vehicle headlights and taillights. Data collection, annotation, and model training were conducted using the Roboflow platform. The research resulted in trained model weights capable of recognizing the state of front and rear lights on various vehicle types under different weather conditions. The paper proposes an adaptation of the YOLOv8-based neural network model for recognizing traffic light signals, which can be utilized for both static recognition in photographs and in real-time or video applications.

The influence of technological parameters of laser processing on tribotechnical characteristics of 45 steel has been investigated, allowing to develop recommendations for obtaining maximum wear resistance. The influence of laser processing modes (pump current I, pulse duration τ) on the values of energy in the pulse E transferred to the surface to be processed has been evaluated. Changing the current strength in the range of 60–120 A and pulse time from 5 to 25 ms allowed us to obtain the change of energy in the pulse in the range from 0.3 to 4.1 J. It was found that tribotechnical properties of the surface treated with the same energy, but different modes of exposure differ. The wear resistance of the sample was estimated by the width of the friction track. It was found that the influence of pulse duration on tribotechnical characteristics is extreme: at a pumping current of 60 A the minimum friction coefficient and maximum wear resistance were obtained at pulse duration of 20 ms. Increase of pulse duration more than 20 ms leads to lamellar fragments peeling off on the surface and deterioration of tribotechnical parameters.