УПРАВЛЕНИЕ БОЛЬШИМИ СИСТЕМАМИ
на главную написать письмо карта сайта

Выпуск 62


  • Rudko I. Applying order statistic to detection problems in the frequency domain
  • We study detection systems operating in the frequency domain and implementing two statistical hypothesis testing with energy criterion. We propose an algorithm of signal detection in the presence of interference based on the statistical properties of the truncated order statistics (TOS). The performance of the algorithm is measured by simulation where the signal is modeled as colored noise and the interference modeled as white noise. The detection system that implements the algorithm-based two-threshold TOS criterion outperforms the “classic” one-threshold algorithm.

  • Chirkova J. Price of anarchy for maximizing the minimum machine load
  • The maximizing the minimum machine delay game with uniformly related machines is considered. Players choose machines with different speeds to run their jobs trying to minimize job’s delay, i.e. chosen machine’s completion time. The social payoff is the minimal delay over all machines. For the general case of N machines we find the lower bound for Price of Anarchy (PoA), and for the case of 3 machines we find its exact value. We prove that the PoA either remains the same or increases when an additional third machine is included into the system with two machines. Also we propose a method of computation the PoA value and illustrate it for 3 machines.

  • Eremenko Y., Glushchenko A. Neural network structure selection method to solve linear controllers parameters adjustment problem
  • The problem of a neural network structure selection is considered. It is used as a part of a neural tuner to adjust P-, PI- or PID-controllers parameters online to control nonlinear plants. A method to find a number of network layers, neurons in each of them, choose activation functions and calculate delay time for network delayed inputs is proposed. Such method does not need a plant model. An algorithm to synthesize and initialize the neural network for the neural tuner with the help of a-priori known data about the plant is developed. Having made the experiments with plant models and two electroheating furnaces, we conclude that neural tuner helps to achieve both time and energy consumption decrease to complete setpoint schedule in comparison with conventional linear controller. This fact shows that proposed method is valid.

  • Alekseev A., Spirina V., Korgin N. Commercial real estate management technology taking into account consumer preferences
  • The management technology of commercial real estate object differing in accounting of consumer preferences and also two problem definitions of management of commercial real estate object is given. The technology includes three stages at the integrated level: identification of consumer preferences, verification of models and management. Using model it is possible to perform scenario modeling and forecasting including game imitating modeling. During business games drafts of administrative decisions which efficiency can be developed it is checked during scenario modeling "what \'85 if \'85". Search of administrative decisions can be carried out by method of the analysis of sensitivity of the chosen efficiency indicator to change of controlled parameters.

  • Chebotarev P., Malyshev V., Tsodikova Y., Loginov A., Lezina Z., Afonkin V. The optimal majority threshold as a function of the variation coefficient of the environment
  • We consider the model of social dynamics determined by collective decisions in a stochastic environment (ViSE model). In the paper we investigate the case of a homogeneous society consisting of classically rational economic agents (or homines economici, or egoists). Alternatives generated by the environment are random vectors of utility increments for each agent. Increments can be positive or negative. An agent votes for those alternatives which increase her utility. An alternative which has gained more votes than the given majority threshold is accepted and all agents receive corresponding increments to their utilities. Expressions for the optimal majority threshold and the maximum expected utility increment as a function of the parameters of the environment are obtained. An estimate of the rate of change of the optimal threshold at zero is given and this value is an absolute constant.

  • Amelin K., Andrievsky B., Tomashevich S., L. Fradkov A. Data exchange with adaptive coding between quadrotors in the formation
  • In this paper we study the control dynamics of a group of quadrotors interacting over a network with limited bandwidth and with feedbacks between agents. We investigate how the quantization process affects the accuracy of the data transfer between the quadrotors moving in formation. The parameters of altitude tracking for quadrotor ”QuadSwarm” was experimentally identified based on two datasets: first where data records were stored onboard and second where data were transmitted via digital communication link. The adaptive coding procedure with second-order dynamical encoder is presented and numerically studied. The results obtained are compared with the theoretical predictions. The experiments showed the efficacy of the adaptive coding procedure in the case when the transmitted signals are nonstationary.

  • Antonenko A., Loshkarev I., Pankov V., Ugolnitsky G. Game theoretic models of motivation of innovations and their identification (on the example of the electricity power industry).
  • The article deals with game theoretic models of coordination of interests in the power industry. We consider the problem of multi-tariff electricity meters integration. Four game-theoretic models catching different aspects of the problem are investigated. The players represent local authorities, electricity consumers and power generating company. We developed software simulation toolbox implementing these models. The software can be used to specify parameters of the games and obtain numerical solutions. The results of software simulation are presented and interpreted. Recommendations and conclusions about the feasibility of new-generation electricity meters integration can be given based on the simulation results.

ИПУ РАН © 2007. Все права защищены