Detection of Changes in the Properties of Time-Varying Random Processes

An approach to analyzing and predicting systems described by time-varying series is developed on the basis of identification of intervals in which time-varying processes do not change their parameters and construction of models for cointegration relations in these intervals and for tracking the changes both in the processes themselves and in the long-range cointegration relations formed by them. CUSUM-type detection algorithms are constructed for detecting the changes in time-varying processes and identifying stationary intervals. If the process parameters are known exactly, these algorithms are optimal by the quickest detection criterion. For the case studied in the paper, the process parameters are not known after the changes in properties, only certain assumptions concerning the type of parameter changes can be made. Therefore, property changes are detected with the help of modified CUSUM-algorithms, which are not optimal, but guarantee a reasonable detection quality.     

Algorithms for the detection,localization, and recognition of electro-optical images of the group of isolated ground locations for inertial-sighting systems of navigation and guidance of aircraft

Algorithms are proposed for the detection, localization (i.e., position determination in the observed scene), and recognition of optoelectronic images of a group of plants with a contrasting image background intended for use in inertial sighting systems for the navigation and guidance of aircraft [1, 2]. The structure and parameters of the developed algorithms are presented. The parameters of the quality of the processes of detection, localization, and recognition of halftone electro-optical images of a group of ground objects are given.     

Parameter estimation and reliable fault detection of electric motors

Accurate model identification and fault detection are necessary for reliable motor control. Motor-characterizing parameters experience substantial changes due to aging, motor operating conditions, and faults. Consequently, motor parameters must be estimated accurately and reliably during operation. Based on enhanced model structures of electric motors that accommodate both normal and faulty modes, this paper introduces bias-corrected least-squares （LS） estimation algorithms that incorporate functions for correcting estimation bias, forgetting factors for capturing sudden faults, and recursive structures for efficient real-time implementation. Permanent magnet motors are used as a benchmark type for concrete algorithm development and evaluation. Algorithms are presented, their properties are established, and their accuracy and robustness are evaluated by simulation case studies under both normal operations and inter-turn winding faults. Implementation issues from different motor control schemes are also discussed.     

基于演化算法的客户关系管理系统研究

演化算法是一种模拟生物进化过程的自适应算法.客户关系管理系统是顺应企业改革发展需要的产物,在企业与客户之间建立起友好的桥梁,增进相互的了解和合作.本文通过对演化算法基本概念的描述,初步阐述了如何将其用于客户关系管理系统中,使系统更具有智能性、自适应性和可扩展性.     

Reachability-based feedback control of crystal size distribution in batch crystallization processes

In the paper, we investigate the controllability of crystallization processes by reachability analysis. Crystallization processes are governed by hyperbolic partial differential equations. Given a desired crystal size distribution, we study its reachability by the temperature control from the initial condition without seeding. When the desired crystal size distribution is reachable, we construct an admissible control steering the state to the desired distribution. Our construction is developed based on the discretized model. To ensure that the desired distribution be reached facing model uncertainty, we propose an output feedback control law to correct errors resulted from disturbed parameters of the model.     

Matrix analytical model of an ATM output buffer with self-similar traffic

Renewal processes with asymptotically hyperbolic interarrival time distributions are shown to exhibit self-similar behavior. An output buffer of an ATM switch is modeled as a discrete time queue with a single server, deterministic service times and self-similar renewal process input. A matrix geometric solution is found for the stationary distribution of states. For the case of hyperbolically distributed interarrival times, the mean and standard deviation of queue length are plotted for various values of the queue utilization and the self-similarity parameter of the arrival process. The self-similarity is found to have a significant impact on the performance of the queue.     

Parameter identifiability for distributed parameter systems of hyperbolic type

In this paper we study questions regarding parameter identifiability for distributed parameter systems of hyperbolic type. The unknown parameters are input distribution functions. We consider the systems with continuous-time input-output data and the systems with discrete-time inputs-output data. For these systems we present the necessary and sufficient conditions for identifiability, in the case of a finite number of sensors. We investigate the relations between the continuous-time input-output systems and the discrete-time input-output systems from the viewpoint of identifiability. Moreover we give the sets of input distribution functions which are N-step identifiable, for the discrete-time input-output systems.     

Biodiversity assessment based on artificial intelligence and neural network algorithms

Together with its associated economic activities, Biodiversity has the impact of increasing the global environment to an unprecedented extent. Around the world, countries are focusing on resource consumption and the ecosystem's ability to deliver them. The order is effectively preserved, and the decision-makers are in need of biodiversity indicators and knowledge which needs to be common in such a way that they can be utilized effectively. High-throughput environmental sensing technology is an increasingly important global monitoring of the impact of human activities on ecosystems. More recently, with passive acoustic sensors, the boom has provided a wide range of efficient, non-invasive, and taxonomic tools for studying wildlife populations and communities, responding to environmental changes. A proposed best practice criteria and detailed guidelines are to be proposed for achieving scores used in biodiversity assessment studies based on species distribution models. Artificial Intelligence and Neural Network Algorithms are highly efficient for the overall detection of the low model, which improves the usual model, takes much time. To establish a clear trend, biological assessments are used to a lesser extent than in the data and the model evaluation. Artificial Intelligence and Neural Network Algorithms agree with relevant model standards that promote transparency and reproducibility and argues that the implementation of biodiversity assessments will lead to high-quality models and inferences used in the final assessment. The expansion of Artificial Intelligence and Neural Network Algorithms standards and guidelines encourages a wider community to participate in ongoing improvement.     

Estimation of Parameters and Arrival Instants of a Signal in a Color Noise Background

Algorithms for detecting and estimating arrival instants and parameters of a signal under noise are designed. The algorithms for detecting signal arrival instants are modifications of cumulative sum algorithms. Signal arrival instant can be detected, as is shown, by the instant of change in the maximal eigenvalue of the covariance matrix of observations. The estimation and detection algorithms are applicable to estimate the parameters of the direction vector and arrival instants of P- and S-waves at automatic seismic stations.     

Exact algorithms for computing p-values of statistics-linear combination of 3-nomial variables

Algorithms for computing exact distribution values of statistics-linear combination of some multinomial variables called 3-nomial variables are derived. In the general case, there are no explicit formulae of the cumulative distribution function of these statistics that allow the direct computation of p-values. The exact distribution in the independent case and in some dependent ones is studied. For each of the two cases of dependence, an algorithm based on recursive relationships is proposed, giving the number of favorable cases and combinatorics to minimize computation time. For each of these NFC algorithms, concurrent PGF algorithms and some realized comparisons are presented.     

Real-time analysis of sperm motion using automatic video image digitization

A methodology is described for automatic, real-time analysis of sperm movement characteristics. Video micrographic images of spermatozoa are digitized, compressed, and then analyzed using a Motion Analysis Corporation Expert Vision system. The system includes a VP-100 video processor, a video camera, and a Sun-2/120 minicomputer. Algorithms for detection of and discrimination among sperm are described, and desirable operating characteristics of the system are considered. Salient parameters which characterize the kinematics of sperm swimming trajectories are discussed.     

基于群体智能的算法研究

计算机技术不断发展,从而带动着算法技术不断更新,尤其是在模仿社会性动物的行为领域,产生了很多的智能算法。本文主要介绍当前几种热门研究的算法,阐述了其工作原理和特点,同时对其发展进行了展望。     

Empirical Studies of Optimization Techniques in the Event-Driven Simulation of Mechanically Alloyed Materials

Results of empirical studies of four algorithms for collision detection optimization (CDO) in a framework of event-driven simulation of mechanically alloyed materials are presented. Performance analysis is conducted in a framework of a shaker ball mill model. The results of the study provide insights on how the physical characteristics of the modeled system, such as the number of particles, the distribution of their radii and the density of packing, influence the simulation. Algorithms presented are suitable for parallelization on EREW PRAM architecture with p processors.     

Computation of partial derivatives when crossing a jump of hydromechanical parameters

Algorithms for determining the gradients and second partial derivatives of hydromechanical parameters when crossing the zone of their jump are considered. It is demonstrated that in this case, the integration steps with respect to the spatial coordinates must be variable and nonmonotonic and must pass the minimum. Two examples of determining the values of variable integration-steps for one of the spatial coordinates in relation to the values of this coordinate are presented.