The subject of study in this article is a method for industrial equipment vibration diagnostics that uses discrete Fourier transform and Allan variance to increase precision and accuracy of industrial equipment vibration diagnostics processes. We propose IoT-oriented solutions based on smart sensors. The primary objectives include validating the practicality of employing platform-oriented technologies for vibro-diagnostics of industrial equipment, creating software and hardware solutions for the IoT platform, and assessing measurement accuracy and precision through the analysis of measurement results in both time and frequency domains. The IoT system architecture for industrial equipment vibration diagnostics consists of three levels. At the autonomous sensor level, vibration acceleration indicators are obtained and transmitted via a BLE digital wireless data transmission channel to the second level, the hub, which is based on a BeagleBone single-board microcomputer. The computing power of BeagleBone is sufficient to work with artificial intelligence algorithms. At the third level of the server platform, the tasks of diagnosing and predicting the state of the equipment are solved, for which the Dictionary Learning algorithm implemented in the Python programming language is used. The verification of the accuracy and precision of the vibration diagnostics system was carried out on the developed stand. A comparison of the expected and measured results in the frequency and time domains confirms the correct operation of the entire system.

Subject of Study: The article focuses on methods and tools for developing flexible distributed environmental monitoring systems for ambient air in urbanized areas and zones of high technogenic load. Objective: To enhance the reliability, timeliness, and spatial resolution of information regarding pollutant concentrations by developing architectural solutions based on modern Internet of Things (IoT) technologies, incorporating mobile measurement platforms and machine learning methods. Tasks: To analyze the limitations of existing stationary monitoring stations; to define the architecture of a hybrid monitoring system combining ground-based sensor networks and mobile segments based on Unmanned Aerial Vehicles (UAVs) to expand coverage; to perform a comparative analysis of sensor technologies, energy-efficient communication protocols (LoRaWAN, NB-IoT), and data processing platforms; to investigate the effectiveness of applying artificial intelligence algorithms to improve the measurement quality of low-cost sensors. Results: The study investigates the shortcomings of traditional monitoring systems, particularly their high cost and low coverage density. A comprehensive approach to deploying IoT systems is proposed, involving the use of distributed sensor nodes for real-time data collection. Special attention is paid to the use of Unmanned Aerial Vehicles as carriers of measurement equipment, which enables the collection of data on the vertical distribution of pollutants and monitoring in hard-to-reach places or directly in the emission plumes of pollution sources. Key challenges in implementing such systems have been identified: the accuracy of low-cost sensors, the energy efficiency of autonomous devices, and data transmission reliability in complex conditions. Scientific Novelty: The scientific novelty lies in the systematization of architectural approaches to building heterogeneous IoT environmental monitoring systems, which, unlike existing solutions, dynamically combine stationary and mobile measurement tools. The effectiveness of applying machine learning methods for automatic sensor calibration and cross-sensitivity compensation has been proven. Conclusions: Implementing the proposed IoT solutions enables a significant increase in the density of monitoring points and reduces network deployment costs. The use of UAVs ensures monitoring flexibility, while the integration of artificial intelligence facilitates the necessary measurement accuracy for decision-making.

Предметом дослідження в статті є методика калібрування акселерометрів у складі цифрової платформи вібраційної діагностики промислового обладнання. Мета – підвищити інформативність процесів вібродіагностики промислового обладнання шляхом розробки та впровадження IoT-орієнтованих рішень на основі інтелектуальних датчиків та актуаторів за стандартом IEEE 1451.0-2007. Завдання: обґрунтувати доцільність використання платформено-орієнтованих технологій для вібраційної діагностики промислового обладнання та вибрати хмарний сервіс для реалізації платформи; розробити програмно-технічні рішення платформи IoT для вібраційної діагностики промислового обладнання; відкалібрувати систему вібродіагностики та перевірити точність вимірювань. Використані методи - мікросервісний підхід, багаторівнева архітектура та оцінка стану обладнання на основі даних про вібрацію. Ми отримали наступні результати. Розроблена та представлена в статті архітектура системи IoT для вібродіагностики промислового обладнання є трирівневою. Рівень автономних датчиків забезпечує зчитування індикаторів віброприскорення та передає дані на рівень Hub, який реалізований на базі одноплатного мікрокомп’ютера BeagleBone через цифровий бездротовий канал передачі даних BLE. Обчислювальна потужність BeagleBone забезпечує роботу з алгоритмами штучного інтелекту. На третьому рівні серверної платформи вирішуються задачі діагностики та прогнозування стану обладнання, для чого застосований алгоритм Dictionary Learning, реалізований мовою програмування Python. Перевірку методики калібрування акселерометра для вібраційної діагностики промислового обладнання проведено на унікальному стенді. Висновки. Коректна робота всієї системи підтверджується збігом очікуваних і виміряних результатів. На наступному етапі ми плануємо розробку додаткових мікросервісів, які забезпечать можливість використання методів аналізу часових рядів і сучасних технологій штучного інтелекту для комплексної діагностики та прогнозування стану обладнання.

The subject of study in the article is the method of industrial equipment vibration diagnostics using Allan variance. The goal is to increase the precision and accuracy of industrial equipment's vibration diagnostics processes by developing and implementing IoT-oriented solutions based on intelligent sensors and actuators per the IEEE 1451.0-2007 standard. Tasks: justify the feasibility of using platform-oriented technologies for vibration diagnostics of industrial equipment and choose a cloud service for the implementation of the platform; develop software and hardware solutions of the IoT platform for vibration diagnostics of industrial equipment; calibrate the vibration diagnostics system and check the measurement precision and accuracy. The methods used are the microservice approach, multilevel architecture, and assessing equipment state-based Allan variance. We obtained the following results. The architecture of the IoT system for vibration diagnostics of industrial equipment developed and presented in the article is three-level. The level of autonomous sensors provides readings of vibration acceleration indicators and transmits data to the Hub level, which is implemented based on a BeagleBone single-board microcomputer through the BLE digital wireless data transmission channel. BeagleBone computing power provides work with artificial intelligence algorithms. At the third level of the server platform, the tasks of diagnosing and predicting the condition of the equipment are solved, for which the Dictionary Learning algorithm implemented in the Python programming language is applied. Verifying the accelerometer calibration method for vibration diagnostics of industrial equipment was performed using a unique stand. Correct operation of the entire system is confirmed by the coincidence of expected and measured results. In the next step, we plan the development of additional microservices that will provide the possibility of using time series analysis methods and modern artificial intelligence technologies for complex diagnostics and forecasting of the equipment state.

This paper proposes hardware and software solutions and a data processing method for vibration diagnostics of industrial equipment, which uses discrete Fourier transform and Allan dispersion to increase the accuracy and stability of measurement processes and result processing. The object of this study is the use of vibration diagnostic methods to implement the concept of maintenance of industrial equipment based on monitoring its current and future condition. The subject of this research is the hardware and software solutions for vibration diagnostics systems and methods for processing measurement results. The purpose of this work is to develop a new resource-saving IoT-oriented wireless solution for vibration diagnostics, where the contact method and MEMS accelerometers are used to measure vibration parameters and to evaluate the effectiveness of new methods and algorithms for processing experimental data. The task: justify the need to find new hardware and software solutions and methods of processing the obtained results for the implementation of the service concept based on the tracking of vibration indicators of technical equipment; provide basic hardware and software solutions for the implementation of the cloud platform of vibration diagnostics; develop methods of processing results; check the developed methods and algorithms using mathematical modeling methods and in an on-site experiment; compare the effectiveness of own and competitive solutions; draw conclusions and formulate a plan for further research. Conclusions. It has been proven that the combination of known analysis methods in the time and frequency domains with multi-level processing gives better results than analogous methods. The developed hardware and software tools and the method of processing measurement results effectively implement the contact method of vibration measurement, which provides the possibility of tracking the state of technical equipment. The developed equipment for the calibration of vibration acceleration sensors can reduce accelerometer errors. Further areas of research are the search for the optimal distribution of calculations on IoT levels, reducing the computational complexity of algorithms, increasing the time of continuous autonomous operation of the lower-level microcontroller, creating micro services for time-series analysis, and researching the dependence of the technical state of the equipment on the calculated Allan deviation.

The application of artificial intelligence in e-commerce offers new avenues for enhancing personalization, resulting in more relevant and user-centered shopping experiences. This study introduces an AI-driven recommendation system within a children’s goods e-commerce platform, developed to analyze both individual user preferences and collective consumer behavior patterns. By leveraging this data, the platform provides personalized product suggestions and assists users in selecting appropriate clothing sizes, optimal delivery options, and real-time stock availability. These features simplify the decision-making process, enhance convenience, and increase user engagement. The growing demand for adaptive e-commerce platforms highlights the relevance of this research, as personalized recommendations foster customer satisfaction and encourage repeat usage. Moreover, this platform's AI integration provides a solution tailored to the children's product market, where reliability, safety, and precision are crucial. This study contributes significant insights to the field by demonstrating the role of AI in improving the quality of digital consumer interactions, particularly within specialized retail environments. This project exemplifies how AI can dynamically respond to both individual and broader consumer data, presenting a valuable framework for future developments in adaptive e-commerce solutions and marking a step forward in the intelligent, responsive design of personalized shopping experiences.