Markov model of a digital stochastic generator

A stochastic process generator based on a chain of modulo-two adders with asynchronous feedback loops is investigated. The generator dynamics is described by the Erlang equations. A mathematical model of the generated stochastic process is proposed. The implementation of the generator using modern electronic devices is considered.     

Output control algorithm with the compensation of biased harmonic disturbances

The paper develops the methods of harmonic disturbance compensation using the measurements of plant output variable. An approach for biased harmonic disturbance compensation is proposed. This approach is superior to known ones in some characteristics.     

Accounting of heterogeneity in the estimation of radiation risks

The problem is considered of taking account of heterogeneity in the individual radiosensitivity in evaluating the radiation risk of the origin of a disease in a group of persons subjected to the radiation action. Mathematical models are described for the estimation of the radiation risk of the origin of the disease with and without regard for heterogeneity. The use of the described method is demonstrated with an example of the estimation of the risk of the origin of radiation-induced diseases from the class of “solid cancers” among the persons taking part in the liquidation of consequences of the accident at the Chernobyl’ Atomic Power Station (CAPS) in the years 1986–1991. It is shown that the heterogeneity disregard leads to understated estimates of the radiation risk of the origin of radiation-induced solid cancers. The effect is investigated of the length of the latent period in the origin of radiation-induced “solid cancers” in taking account of heterogeneity.     

Petersen’s lemma on matrix uncertainty and its generalizations

Various generalizations and refinements are proposed for a well-known result on robust matrix sign-definiteness, which is extensively exploited in quadratic stability, design of robustly quadratically stabilizing controllers, robust LQR-problem, etc. The main emphasis is put on formulating the results in terms of linear matrix inequalities.     

Multidimensional analogs of the kendall equation for priority queueing systems: computational aspects

An analogy between celebrated Kendall equation for busy periods in the system M|GI|1 and analytical results for busy periods in the priority systemsM _r |GI _r |1 is drawn. These results can be viewed as generalizations of the functional Kendall equation. The methodology and algorithms of numerical solution of recurrent functional equations which appear in the analysis of such queueing systems are developed. The efficiency of the algorithms is achieved by acceleration of the numerical procedure of solving the classical Kendall equation. An algorithm of calculation of the system workload coefficient calculation is given.     

Adaptive algorithms to estimate parameters of the optimal policies of inventory control under limited stock-out

Consideration was given to the inventory control by the criterion for minimum of the cumulative mean costs over a sufficiently long planning horizon in conditions of uncertain demand characteristics. Adaptive algorithms to estimate the parameters of the optimal two-level control policies oriented to the supply systems with high level of customer servicing were constructed for the stationary mode of operation.     

Nonlinear parity relations: A logic-dynamic approach

Consideration was given to construction of the parity relations for systems described by the nonlinear dynamic models. To solve this problem, a logic-dynamic approach was proposed, and the realizability conditions providing insensitivity to the perturbing actions were given for it.     

RCGNN：一种基于实数编码的遗传神经网络预测方法

将遗传算法与神经网络相结合,提出一种实数编码、自适应选择、算术交叉、高斯变异、爬山操作的改进遗传BP神经网络RCGNN,利用遗传算法对神经网络权值和阈值进行优化。以时间序列预测的实例进行编程计算表明,用遗传算法进行网络训练,其收敛速度快,最终总误差最小,预测准确率高。对算法中参数进行的相应研究表明,增加爬山操作次数能很好地提高网络训练的速度,同时使误差下降快;爬山操作越多,收敛速度越快,最终误差越小,但计算运行时间也会增加。     

基于A^＊算法的八数码问题的优化实现

用人工智能领域中经典的A^＊算法解决了人工智能中常见的八数码问题。本文首先介绍了八数码问题,然后对A^＊算法进行了解释。针对八数码问题给出并证明了路径不存在时的条件,并事先作了判定。定义了灵活的估价函数,分析了估价函数对程序效率的影响,并对Closed表进行了若干优化,提高了搜索效率,取得了较好的效果。     

人民币号码自动识别方法研究

针对人民币印刷中号码的识别问题,本文研究了其中的识别算法,包括号码图像的二值化、号码区域的定位、号码字符的分割及归一化和号码字符的识别,提出了一种号码区域定位及分割方法。在此基础上,建立了相应的识别系统并进行了实验研究。结果表明,系统具有较高的识别精度,能较好地满足实际需求。