Tatiana Atanasova

Towards semantic-based process-oriented control in digital home

The purpose of this work is to investigate the specifics in the development of technologies for heterogeneous data and process integration in digital home and to show possible solutions during design of integrated applications. The analysis of the integrated data can be useful for the development of improved algorithms for monitoring and control of digital networked home.

Distributed heterogeneous knowledge data base for control system design: multiagent development and support

It is proposed to construct an automatic control system design environment with engineering knowledge-based management as an integral part. The control design environment consists of a set of control design modules and a number of control systems models that have to be designed or are to be designed. The knowledge objects are distributed in multiple heterogeneous data sources. A multi-agent system approach to the design of resource management components in the heterogeneous environment is discussed. Cooperative agents communicate through networked environment. The database structure is determined dynamically by XML tools. Meta-knowledge is described as an RDF-file that makes it possible for intelligent retrieval of information. Communications are established using CORBA.

ICT-Based Beekeeping Using IoT and Machine Learning

Integrating information and communication systems into different industries enables storage, management and processing of data and then transforming it into useful information and knowledge. The gradual integration of information technology into beekeeping may provide remarkable results in control over care for bee hives and increase profitability. The aim of this paper is to provide ICT-based means to predict the survival of bee families when the winter season is over. The procedures of collecting and processing of the data from beehives is examined. Several machine learning algorithms are applied and tested. The performance of classification models used for the prediction is estimated. The experimental results are presented.

Problems with quality enabling of information functions composition in smart buildings

The aim of this paper is to point out and summarise existing problems and research trends that arise when assessing and ensuring quality of information services. On the focus of this work are smart buildings. Basic information functions that are composed in different services are described. The sets and structures of the composition form various scenarios. Problems in determination the quality of these scenarios, and their implementation to meet the requirements of the customers are considered.

Optimizing the management and control of apparel enterprise by information technologies

The integrated system for optimizing the management and control of apparel enterprise is presented in the paper. The purpose of the integrated system is to design a framework tailored to operational and organizational structure of the apparel company. The system allows the executives of the company to effectively organize, analyze, control and optimize both the production process and the marketing of its production.

Neural network integration in control system structures: a case study towards intelligent control by neural network integration

Neural networks techniques have been found to be useful for controlling highly uncertain, nonlinear and complex systems. The possibility of applying neural networks for modelling of unknown functions in dynamic environments has been demonstrated by several empirical studies. Allowing unknown nonlinearities in the system to be controlled we extend the class of uncertain systems that adaptive control techniques can be applied to. In the paper an integration of neural networks in the control systems is considered. Neural network application schemes are described. Some advantages of neural modelling and control are presented.