Based on the analysis on shortcomings of controlling strategies of direct power on the traditional 3-phase Boost-type PWM Rectifier through space vectors, a new controlling strategy is proposed in this paper. With this new strategy, the space vector is divided into 12 basic space vectors to form a new switching table. The instantaneous power controlling technology is realized through the control of instantaneous active power and reactive power by choosing the best switching state of PWM Rectifier. This new controlling strategy can put the input side unit power into operation, improve the efficiency of power and reduce harmonics of the input current. The effectiveness and feasibility of this controlling strategy have been justified through the simulation and experiment.

Face detection and feature extraction are widely used in face recognition, expression recognition, video-phone conference, human-computer interface, and intelligent monitoring system and so on. In order to solve the problem of low accuracy in the use of the traditional face detection technology in the detection of human face under complex background, in this paper, we propose a new view for face detection method, which is Snake model base on shape restricted. Consider character of the human facial shape is similar to an ellipse, the traditional Snake model is modified by adding the shape energy item in its energy function, which strong the relationship among the Snake points so as to improve the accuracy. Experiments show that it is an effective approach for extracting facial contour under complex background.

As the requirement for the power quality is increasing nowadays, the distributed source energy has been widely used. The association of different Distributed Generation System (DGS) could cause as many problems as it may solve. Micro-grid associates the distributed generation and the loads as a subsystem. The generation and loads could separate from the distribution system to isolate the micro-grid's load from the disturbance without harming the distributed network's integrity. This paper discusses the associated operation of the distributed generation system and the influence it introduced into the distributed network. The design the distributed generation system would influence the voltage distribution of the system. Simulation and analysis has been done to verify it and the methods to avoid the harm it introduced to the network are proposed.

To implement distribution network reconfiguration better,the divide and conquer method is introduced in,and then a method to implement the network reconfiguration by parallel Tabu search algorithm is proposed.Simplifying distribution network as the ring network that only contains branch-groups,necessary conditions for distribution network reconfiguration are given.During the reconfiguration,firstly the switched off branch-groups are chosen and depth-first algorithm is adopted to form father and son linked list for judging network structure;then the thinking of divide and conquer that can decrease the complexity is introduced in and to decompose the solution neighborhood of tabu search into sub-neighborhood,and switchable off branch-groups are taken as the solution space for the grouping and multi-processors are applied to perform parallel tabu search.Simulation results of IEEE 69-bus system show that the proposed method is effective.

This paper develops a new error concealment algorithm for the whole-frame losses in wireless video transmission based on H.264/AVC.At first, we propose a forward motion vector extrapolation algorithm based on variable-size block motion vector estimation to obtain the motion vector field of the loss frame without yielding the hole problems. Then, we partition the missing frame into several object regions by clustering the motion vectors, furthermore, we apply median filter to reconstruct the motion vectors for hole areas based on the reliabilities of neighboring motion vectors. Finally, all macroblocks in the lost frame are estimated by using the extrapolated motion vectors. Experimental results show that the proposed algorithm provides better image quality than conventional methods both objectively and subjectively

The distributed resource energy has been widely used to cope with the higher demand for power quality. The association of different Distributed Generation System (DGS) could cause as many problems as it may solve. The generation and loads could be considered as a micro-grid and separate from the distribution system to isolate the micro-grid's load from the disturbance without harming the distributed network's integrity. This paper discusses the operation principle of the distributed generation system and the influence it introduced into the distributed network. The location and penetration of the distributed generation system would influence the power loss of the grid. Simulation and analysis has been done to verify the results. The methods are proposed to properly design the location and penetration of distributed generation system and improve the power quality.

To make more effectively use of the mass data on the SIS(Supervisory Information System) platform of thermal power plant,its economical indexes are sorted into controllable and uncontrollable according to their characteristics.An optimized model of dynamically controllable indexes is built up to evaluate the selected indexes:generated power,plant electricity rate and other indicators,and the dynamic optimization strategy based on the neural network model is built up on SIS platform to evaluate the shift groups.With the SIS of a power plant as an example,by applying the optimized controllable indexes as the supervisory parameters of shift group,the optimized model of dynamically controllable indexes based on SIS data can comprehensively reflect the unit operational status and the dynamic optimization strategy based on neural network model can result in better management effectiveness than that of the traditional indicator evaluation system.

Wireless video communication is an important application supported by 3G mobile communication systems. In many applications,the pre-encoded video needs to be transcoded to lower spatial resolution to fit for various display screen sizes on mobile terminal devices.In this paper, we propose a fast video transcoder,working entirely in the frequent-domain, to convert high quality H.264/AVC bitstream into low bit rate stream with low spatial resolution for wireless video access. Simulation results show that the computational complexity of the proposed transcoder is less than 40% of that required by cascaded pixel-domain transcoder, however, the video quality achieved by the proposed method and by the pixel-domain methods is hardly distinguishable.

Spatial resolution reduction of compressed video is highly desirable in video transcoding. Many studies have addressed this problem, but they either cause drift error or have high computation complexity. A new spatial downsampling algorithm for DCT-Domain video transcoder is proposed in this paper, in which LxM sub-blocks of NxN are combined into a LxN rows and MxN columns blocks in DCT-Domain firstly, by extracting the left-top NxN low-frequency coefficients from this block, the new NxN downsampled block is the obtained. Experimental results confirm that the proposed algorithm performs better than other algorithms in terms of PNSR of reconstructed video sequences, furthermore, the computation complexity reduced greatly. sheet, as illustrated by the portions given in this document.

Wireless video communication is an important application supported by 3G mobile communication systems. Motivated by the wide adoption of H.264 and the demand of universal multimedia data access over wireless network, this paper presents a fast video transcoder, working entirely in the DCT-domain, to convert high quality H.264 bitstream into low bit rate stream with low spatial resolution for wireless video access, furthermore, a motion vector estimation algorithm based on mapping and merging process is proposed to utilize the original motion information to predict motion vectors for the downscaled video. Simulation results show that, compared with the conventional transcoder, the proposed method significantly reduces the transcoding complexity while maintaining comparable rate distortion performance.

GPS-based synchronous phasor measurement technology is a powerful tool for the security and reliable operation of the inter-connected electric power system. This paper presents an ACO-based approach to optimize the phasor measurement unit (PMU) placement problem. The pheromone trail persistence coefficient adaptive adjustment mechanism and stochastic perturbing progress are introduced into the ant colony system (ACS), in case the algorithm entering the stagnation behavior and getting stuck at local minima. The improved algorithm outperforms the ACS in obtaining global optimal solution and convergence speed, when applied to optimizing the PMU placement problem. A graph-theoretic procedure based on depth first search is adopted to analyze system observability. Simulation results in optimizing a provincial 46-bus system PMU placement problem show that the improved ACS algorithm is effective.

This paper develops a new error concealment algorithm for the whole-frame losses in wireless video transmission based on H.264/AVC. At first, we propose a forward motion vector extrapolation algorithm based on variable-size block motion vector estimation to obtain the motion vector field of the loss frame without yielding the hole problems. Then, we partition the missing frame into several object regions by clustering the motion vectors, furthermore, we apply median filter to reconstruct the motion vectors for hole areas based on the reliabilities of neighboring motion vectors. Finally, all macroblocks in the lost frame are estimated by using the extrapolated motion vectors. Experimental results show that the proposed algorithm provides better image quality than conventional methods both objectively and subjectively.

In this paper, we develop an improved error concealment scheme for entire frame loss of H.264/AVC by using variable-size block motion vector extrapolation. At first, the proposed algorithm extrapolates MVs from available neighboring frames onto the lost frame and determines the MV of each 8×8 block, for those blocks with dissimilar candidate MVs, they are divided into four sub-blocks and an re-estimation method is implemented on every sub-blocks. Then, a more complete motion vector field is obtained by u sing median filtering. At last, the blocks without extrapolation MVs are concealed by using linear interpolation algorithm. Experimental results show that this method can alleviate the shortcomings of the fixed-size method effectively and provide better performance than other concealment methods.

GPS-based synchronous phasor measurement technology is a powerful tool for the security and reliable operation of the inter-connected electric power system. This paper presents an ACO-based approach to optimize the phasor measurement unit (PMU) placement problem. The pheromone trail persistence coefficient adaptive adjustment mechanism and stochastic perturbing progress are introduced into the ant colony system(ACS), in case the algorithm entering the stagnation behavior and getting stuck at local minima. The improved algorithm outperforms the ACS in obtaining global optimal solution and convergence speed, when applied to optimizing the PMU placement problem. A graph-theoretic procedure based on depth first search is adopted to analyze system observability. Simulation results in optimizing a provincial 46-bus system PMU placement problem show that the improved ACS algorithm is effective.

Based on the N330-17.75-540/540 steam turbine unit, the thermodynamic methods were firstly used to test the influence of the economics in different loads with different input pressures (fixed-pressure operation and sliding-pressure operation). And then the variation of the slightly increasing output was explored, which was caused by the exhaust pressure changes of the steam turbine. The best operation mode was determined after considering the unit security and the economic benefits. Finally, we obtained the optimally fixed-sliding pressure curve in the variable condition, as well as the unit relationship between a slightly increasing output and the exhausted pressure changes, those were beneficial for the economical operation and on-line monitoring.

Multimedia applications have become the most promising applications over wireless networks.However, wireless networks have lower bandwidth than wired networks, this calls for fast bitrate reduction techniques through video transcoding that can deliver the best video quality to the receiver while adhering to the bitrate constraints of the wireless network.This paper presents a hybrid homogenous H.264 video transcoding method in order to reduce the bitrate for wireless video communication. In the paper, we propose to transcode intra frame with cascaded pixel-domain architecture and inter frame with open-loop architecture, furthermore, we propose a drift error compensation algorithm for P frames in order to maintain the transcoding quality. Simulation results show that the hybrid transcoding method has lower computation complexity and higher transcoded quality; it can be effectively applied to transcode H.264 video stream in wireless networks.

In this paper, we develop a whole frame loss recovery method for H.264 by using block-based motion vector extrapolation in the decoder. Firstly,we propose a variable-size block motion vetor extrapolation algorithm to construct the motion vectors for the lost frame from available neighboring frame. Then, a more complete motion vector field is obtained by performing median filtering. At last, the blocks without motion vectors are concealed by using linear interpolation algorithm. Experimental results show that this method can alleviate the shortcomings of the fixed-size method effectively and provide better performance than other concealment methods.

Wireless video transmission is expected to experience entire frame losses since data packets always carry a whole frame in low bitrate coding scenario. In this paper, we propose an entire frame loss recovery algorithm for H.264/AVC by using variable-size block motion vector extrapolation. At first, the proposed algorithm extrapolates MVs from available previous frames onto the lost frame and determines the MV of each 8×8 block, for those blocks with dissimilar candidate MVs, they are divided into four sub-blocks and one re-estimation method is implemented on every sub-blocks. Then, a more complete motion vector field is obtained by using median filtering. At last, the blocks without extrapolation MVs are concealed by using linear interpolation algorithm. Experimental results show that this method can alleviate the shortcomings of the fixed-size method and it is very effective in dealing with the entire frame loss occurring in wireless video transmission.

This paper presents a new method of rule discovery in time series of power system based on rough set. These techniques used here are sample window and rough set based rule reduction. It can produce meaningful series time rules In the paper, an instance of rule discovery of power system alarm is analyzed.

This paper proposes a novel object-based rate control algorithm to transmit real-time video over variable ad-hoc wireless channels by employing an adaptive rate control algorithm. The algorithm characterizes the variable data rate of the ad-hoc wireless system by by the packet-level Gilbert twostate Markov Model, estimates the available channel bandwidth between the sender and receiver, and adjusts the output rate of Video encoder using R-D framework and frame-skip control mechanism to fit the estimated time-varying channel bandwidth. Simulation results show that this proposed algorithm can effectively utilize the network capacity while avoiding packet loss as the number of hops between source and destination changes, improves the realtime video quality.