×

You are using an outdated browser Internet Explorer. It does not support some functions of the site.

Recommend that you install one of the following browsers: Firefox, Opera or Chrome.

Contacts:

+7 961 270-60-01
ivdon3@bk.ru

  • Implementation of a real physical object control controller using methods of the neuroevolutionary algorithm NEAT

    In this experiment, a solver (NEAT) and a simulator (an inverted pendulum cart object) are implemented, where the solver will influence the object in order to keep it in a stable state, i.e. don't let the pendulum fall. The main objective of the experiment is to study the possibility of implementing a simulator of a real physical object and use it to determine the target function of the neuroevolutionary algorithm NEAT. Solving this problem will make it possible to implement controllers based on the NEAT algorithm, capable of controlling real physical objects.

    Keywords: machine learning, non-revolutionary algorithms, genetic algorithms, neural networks

  • Analysis of the effectiveness of bridge reinforcement structures

    Economic development directly depends on the creation of a network of roads of the highest category. Major road construction requires large capital investments and less funds are allocated for the operation of existing roads and infrastructure. Instead of replacing bridge structures, it is necessary to use their reconstruction. The article discusses the problem of reconstruction of bridge structures, namely the restoration of carrying capacity, which has decreased during many years of operation. The relevance of the study, its goals and objectives are noted. A classification of reinforcement structures according to various criteria is proposed. The types of bridge reinforcements often used in practice, their advantages and disadvantages, are analyzed. The presented material is illustrated with photographs of objects. Successful reconstruction directly depends on the qualifications of designers and contractors, since with insufficient reinforcement, the span continues to work and is overloaded and gradually collapses. A modern method of reinforcement based on the use of carbon composite is presented. Significant disadvantages of this method for strengthening bridges and its modification using a tensioning device to secure and tension the carbon lamellas have been noted. The use of a tensioning device allows the superstructure to be partially unloaded from permanent and temporary loads. The main conclusions are presented.

    Keywords: bridge, strengthening, reconstruction, truss, carbon composite, lamella, load-carrying capacity, load-bearing capacity, tension

  • Application and comparison of evolutionary algorithms in the framework of the problem of reinforcement learning for unstable systems

    The aim of this work is the implementation and comparison of genetic algorithms in the framework of the problem of reinforcement learning for the control of unstable systems. The unstable system will be the CartPole Open AI GYM object, which simulates the balancing of a rod hinged on a cart that moves left and right. The goal is to keep the pole in a vertical position for as long as possible. The control of this object is implemented using two learning methods: the neuroevolutionary algorithm (NEAT) and the multilayer perceptron using genetic algorithms (DEAP).

    Keywords: machine learning, non-revolutionary algorithms, genetic algorithms, reinforcement learning, neural networks

  • System for predicting electricity consumption in food production based on streaming data

    The purpose of this work is to implement a system for predicting electricity consumption in food production and to select the most appropriate method for training the forecasting model. In this work, a system was implemented for predicting electricity consumption based on streaming data, which receives them in "real time". The system is implemented on the principle of microservice architecture, where the following were implemented: a service for collecting data from meters, a service for data aggregation and forecasting services. Two forecasting services were implemented: using the classical learning approach based on the ARIMA model and the online learning approach using the HATR online model, the results of which were compared using tests for predicting anomalous values and forecasting under conditions of a change in the data concept, or drift concepts.

    Keywords: machine learning, online learning, online model, concept drift, data drift

  • Influence of the method of grinding the charge on the electrophysical parameters of the piezoceramic material ZTBS-3

    Piezoceramics based on the lead zirconate-titanate (PZT) system is the basis of most transducers operating on the piezoelectric effect and are used in various electronic devices operating in the sections of electrical and hydroacoustics, ultrasonic technology, etc. In turn, piezoceramic materials created in the last century have substantially exhausted their potential for use in modern devices, and it is rather difficult to develop new compositions and is not economically efficient. In connection with the above, an urgent task is to find modern ways to increase the parameters of piezoceramic materials.

    Keywords: piezoceramics, piezophase, piezomaterial, ceramic technology, microstructure

  • Abstracts

    `

    Keywords: