Adaptive image segmentation using genetic and hybrid search methods

This paper describes an adaptive approach for the important image processing problem of image segmentation that relies on learning from experience to adapt and improve the segmentation performance. The adaptive image segmentation system incorporates a feedback loop consisting of a machine learning subsystem, an image segmentation algorithm, and an evaluation component which determines segmentation quality. The machine learning component is based on genetic adaptation and uses (separately) a pure genetic algorithm (GA) and a hybrid of GA and hill climbing (HC). When the learning subsystem is based on pure genetics, the corresponding evaluation component is based on a vector of evaluation criteria. For the hybrid case, the system employs a scalar evaluation measure which is a weighted combination of the different criteria. Experimental results for pure genetic and hybrid search methods are presented using a representative database of outdoor TV imagery. The multiobjective optimization demonstrates the ability of the adaptive image segmentation system to provide high quality segmentation results in a minimal number of generations     

On the electrodynamics of counter stream in pulsar’s inner annular cap region

The electric properties of pulsar’s inner annular gap are explored in this paper. Under two main assumptions, (1) the pulsar is alive, (2) the total charge of pulsar should not vary with time, the condition for the acceleration of negative particle in the annular region is derived. The acceleration condition is j₋ ? 0.5j₊, i.e., the current carried by negative particles is greater than or equal to 0.5 times of the current carried by positive particles. This condition holds even when the backward flow of positive particles exists in the annular region. It is noted that the outflow of negative particles offers good opportunities to understand the current closure problem of pulsar as well as wide radiation beam of pulsar observed at high energy band.     

Comparison of half-bridge off-line, ZCS-QRC and ZVS-MRC

The performance of the half-bridge (HB) zero-current-switched (ZCS) quasi-resonant converter (QRC) and zero-voltage-switched (ZVS) multiresonant converter (MRC) is compared with respect to their efficiency, input voltage range, semiconductor stresses, power density, and reliability. The efficiency of the HB ZVS-MRC at a given nominal input is shown to be highly dependent on the range of the input voltage, and it suffers when the converter has to be designed to cover a wide range. However, this is not the case for the HB ZCS-QRC. Experimental versions of the HB ZCS-QRC and HB ZVS-MRC were designed for the input voltage range from 150 to 350 V and a maximum output power of 100 W, under the same constraints, to facilitate their comparison     

PI controller relay auto-tuning using delay and phase margin in PMSM drives

This paper presents an auto-tuning method for a proportion plus integral（PI） controller for permanent magnet synchronous motor（PMSM） drives, which is supposed to be embedded in electro-mechanical actuator（EMA） control module in aircraft. The method, based on a relay feedback with variable delay time, explores different critical points of the system frequency response.The Nyquist points of the plant can then be derived from the delay time and filter time constant.The coefficients of the PI controller can then be obtained by calculation while shifting the Nyquist point to a specific position to obtain the required phase margin. The major advantage of the autotuning method is that it can provide a series of tuning results for different system bandwidths and damping ratios, corresponding to the specification for delay time and phase margin. Simulation and experimental results for the PMSM controller verify the performance of both the current loop and the speed loop auto-tuning.     

Stability Analysis of a Buck Regulator Employing Input Filter Compensation

The interaction between the input filter and the regulator often causes serious degradation of performance. The reduction in loop gain due to input filter interaction can result in system instability. An exact stability analysis of the buck regulator system is presented. The input filter parameter values are varied and system instability is predicted for the case without feedforward. The eigenvalues of the system can be brought back into the unit circle and the system thus stabilized with the addition of the feedforward loop. Measurements made for the cases with and without feedforward confirm the analytical prediction.