No. 6, 2013
Mathematical Problems of Control
N.N. Nepejvoda Algebraic Approach to Control (pp. 2-14)
Abstract. Algebraic structures generated by modeling programs are theoretically studied. The main new notion here is GAPS (General Algebraic Program System). Examples are given how to represent control systems as GAPS. Open problems: algebras of almost invertible processes; discretization and approximation of abstract infinite algebras by finite GAPS; classification of finite GAPS and methods of their composition and decomposition.
Keywords: algebraic programming, magmas, control systems, supercomputers, Landauer limit, Chaitin limit.
Information Technologies in Control
Abstract. Challenges of Big Data technologies to engineers, specialists in applied mathematics and artificial intelligence, as well as to scientists in various branches of applications, are analyzed and structured.
Keywords: Big Data, Data Analysis, Big Control.
Abstract. The paper presents a model of servicing requests for broadband multimedia learning resources and model of data access in hybrid cloud storage system. The models are used to improve the effectiveness of computing resources in distance learning system by distributing the flow of requests between nodes with use of developed module for the «OpenStack» system controller.
Keywords: cloud computing, multimedia learning resources, load balancing, hybrid cloud system «OpenStack».
Control in Social and Economic Systems
V.L. Makarov, A.R. Bakhtizin, E.D. Sushko Computer Simulation of Interaction between Municipalities, Regions, Government Authorities (pp. 31-40)
Abstract. The paper gives an overview of current multi-agent systems developed for analysis of typical problems of territorial entities. The multi-agent model of municipalities within a region wh ere the agents are people and organizations in which they work is presented. The special consideration is given to providing a realistic simulation of the behavior of such agents as participants of production, as well as of the interdependence of behavior and state of agents of various types and environment changes. The example of using the model to test different options of vertical interbudgetary relations is given.
Keywords: agent-based modeling, testing of regional policy, human behavior in socio-economic environment, labor potential.
Abstract. The paper proposes the measures of stability and risk for a separate investor and the structured share market. The research of the market model stability with the use of proposed measures is carried out. It is shown that investors, adhering to various forecasts of the share market development as a result of their various information awareness, and choosing different strategies (portfolios), promote the given system stability.
Keywords: collective risk, risk coefficient, average covariance, stability.
Abstract. The paper gives the analysis of effectiveness of using results of time series neurostructural prediction (exchange rates) in hedging of currency risks using derivative financial instruments. Description of the developed software is given. The technique of computation of effectiveness of use of neurostructural predictions is considered. Results of computational experiments are provided.
Keywords: neurostructural modeling, time series prediction, foreign exchange hedging.
Control of Complex Technological Processes and Productions
Abstract. As a part of managing behavior of an active consumer of electric power in prospective smart grids it is necessary to create a mathematical model that meets his or her economic interests. Existing models either do not take into account all relevant aspects or turn out to be too complicated for the purposes of multi-agent modeling. Mathematical model of an active consumer and its use for investigating the problem of consumption and local generation regimes optimization is proposed. The conditions when the consumer’s problem has a pretty simple and efficient solution are derived. The proposed approach is illustrated by optimizing the operating modes of equipment for a single household.
Keywords: active consumer in power industry, Smart Grid, Demand-Side management, Demand Response.
Yu.A. Zak The Construction Schedule for the Performance of Tasks of Parallel Systems of Machines and Production Lines (pp. 62-70)
Abstract. The development of Flow-Shop-Problem is considered. The mathematical model of scheduling for parallel working systems of machines under a given set of constraints on deadlines for separate tasks is considered. The properties of admissible and optimal schedules are examined and the methods of optimal scheduling by sequential optimization algorithms are proposed. The presented example shows the efficiency of the developed algorithms.
Keywords: parallel system of machines, Flow-Shop-Problem, lim its on assignments, valid and optimal scheduling algorithms, sequential optimization.
V.N. Meshcheryakov, O.V. Meshcheryakova, P.V. Saraev Mathematical Modeling and Process Control in Industrial Automation Systems Using Complex-Valued Neural Networks (pp. 71-75)
Abstract. The paper considers a multi-valued threshold logic, which is the base of multi-valued neurons, discusses the advantages of complex-valued neural networks. The complex-valued activation function is defined for the multi-valued neuron and the backpropagation learning algorithms for the multi-valued neuron and complex-valued neural networks are considered. Further, a method of applying complex-valued neural networks for modeling processes is proposed.
Keywords: mathematical modeling, complex-valued neural networks, neurocontrol.