Search

Designing a color hologram with conventional metasurfaces usually resorts to a supercell strategy or single-sized approach with different incident angles. However, these designs still have their own drawbacks that need to be further solved. Herein, we show a new, to the best of our knowledge, single-sized strategy to design full-color geometric meta-holograms by utilizing the conjugation property of two circularly polarized lights with opposite handedness and diffraction dispersion. The experimentally captured holographic color images are reconstructed with high quality and without cross talk, which agrees well with our theoretical prediction. Moreover, only with an appropriate combination of wavelength and polarization state can color images be observed accurately. Our strategy provides a simple and effective approach for full-color meta-holography and offers significant potential in image display, information storage, etc.

Aiming at the problem of missing part of the characteristic data of the incoming target during the threat assessment of air defense operations in important area at sea, a new threat evaluation method considering the cosine distance between the characteristic vector of the incoming target and the standard target vector is proposed. First, determine the threat assessment index system and attribute weights of the attacking target according to the characteristics of the air defense mission on the key area. Secondly, according to the combat and exercise experience, experts are invited to give a standard sample of high threat. Finally, the vector cosine-based value threat assessment method to calculate the threat value of each target. The simulation results show that the method proposed in this paper can better solve the threat assessment problem when the characteristic data of the attacking target is incomplete, and it has certain reference value for actual combat applications.

Though already being one of the standard modules of SCADA for years, the fault diagnosis software still exists the problem of high misrecognition rate. To mitigate the issue, this paper proposed a new fault diagnosis model and algorithm based on logistic regression deep neural network (LRDNN) to improve the diagnosis accuracy. The basic process of the proposed method is as follows: 1) construct a deep fully connected neural network for each power equipment. The input of the neural network is the vector of the fault signals of equipment, including switching signal, protection signal and etc., and the output is the failure probability; 2) train the LRDNN based on RMSprop method. Since the proposed method can dig out the complicated relationship between the fault signals and the fault probability, it owns high diagnosis accuracy. Moreover, this paper designed a sample expansion method based on random sampling strategy to successfully realize the training of the LRDNN, skillfully overcoming the defect of the lack of sample. The basic process of which is: 1) extend the historical fault records of equipment by including the fault records of equipment at nearby areas, whose type is the same as the one to be diagnosed; 2) generate adequate number of samples by probability statistics and random sampling operation. A comprehensive comparison of the proposed method and the method based on expert system was performed on practical power systems, which shows that the proposed method reduces the misrecognition rate from 31.25% to 0 compared with the one based on expert system.

Precise identity recognition is a pre-condition for robots to enter the human living environment. Most of the existed face identification methods cannot work on the non-frontal face since the severe texture loss. In this paper, we propose a novel system to deal with multi-angle face identification in video sequence based on frontal face generation, which replaces the process of detection, alignment in the typical face identification system. To solve the problem of face texture loss in large pose variation, we creatively combine generative adversarial networks (GAN) with the state-of-the-art facial landmark localization method. The proposed system was tested on video database containing multi-angle faces, and the experimental results indicate that our system can recognize more faces in the frames, and improve the accuracy of identification for multi-angle face by 130%.

Power grid dispatching is a complex system engineering, and its working pattern and operation principle are implicit in the operation behavior of the dispatcher's daily work. This paper proposes a method which applies the big data technology to mine the operation patterns of grid dispatching. Firstly, data preprocessing is performed on a large number of historical operational records through session division, routine distinguishing and page hierarchy division, where a method of dynamically determining session division threshold and routine distinguishing reference value based on statistical indicators is proposed. Then, the clustering model based on session similarity and affinity propagation algorithm is adopted. This algorithm clusters the operation records after data preprocessing to find the daily operation behavior pattern of the dispatcher. The operation log data of a regional dispatching center in China is taken as an example to verify the algorithm. The results show that the method can effectively identify the operational behavior pattern in power grid dispatching, which provides strong support for the application of auxiliary decision based on the operating mode.

Lane detection plays an important role in improving autopilot&rsquo, s safety. In this paper, a novel lane-division-lines detection method is proposed, which exhibits good performances in abnormal illumination and lane occlusion. It includes three major components: First, the captured image is converted to aerial view to make full use of parallel lanes&rsquo, characteristics. Second, a ridge detector is proposed to extract each lane&rsquo, s feature points and remove noise points with an adaptable neural network (ANN). Last, the lane-division-lines are accurately fitted by an improved random sample consensus (RANSAC), termed the (regional) gaussian distribution random sample consensus (G-RANSAC). To test the performances of this novel lane detection method, we proposed a new index named the lane departure index (LDI) describing the departure degree between true lane and predicted lane. Experimental results verified the superior performances of the proposed method over others in different testing scenarios, respectively achieving 99.02%, 96.92%, 96.65% and 91.61% true-positive rates (TPR), and 66.16, 54.85, 55.98 and 52.61 LDIs in four different types of testing scenarios.

Climbing stairs is an important ability for humanoid robots. This paper proposes a target location and gait planning method for humanoid robot climbing stairs. Target location enables the humanoid robot to accurately locate the stair position, and the gait planning algorithm makes the humanoid robot to climb the stairs. In order to enable the humanoid robot to climb multiple stairs, this paper proposes control strategy of motion for humanoid robot. The effectiveness of our proposed method was confirmed by experiments. The experimental results show that this method improves the accuracy of target location, and enables the humanoid robot Nao to climb a staircase with 2.928 cm steps.

This paper presents a controller for lane keeping system (LKS), and the method of the reference trajectory generation is interpolating five preview points obtained by sensors. Its mean purpose is to reduce the demand for sensors by only using five points. Also the distance between these five preview points is determined by the longitudinal velocity. The computation of the controller has been achieved by casting the model predictive control (MPC) for tracking problem into the quadratic programming (QP) problem. The steering angle is minimized to generate the optimal control sequence and the effectiveness and robustness of the proposed approaches are demonstrated through co-simulations of Matlab/Simulink and CarSim under the different velocity. And the simulation results show that the MPC controller is suitable for reducing the lateral displacement and achieving better performance.

The traditional lane detection methods based on the RANSAC algorithm used to cause many false detection and unable to accurately detect the lanes in complex road environment, because of the existence of interferential noise points in the set of sampling points. Aiming at these issues, this paper presents a new lane detection method combined fuzzy control with RANSAC algorithm. The first process of the new lane detection method is pretreatment, the purpose of which is to denoise the image preliminarily through filtering and binarization. And then it selects the region of interest (ROI) that contains lanes in the input image and extract the initial boundary candidate points of the lanes in ROI. So far, there are still a lot of irrelevant noise points in the set of lane boundary candidate points. It would analyze the relationship between the interferential noise points and the boundary points of the lane, and then remove the interferential noise points from the set of lane boundary candidate points by using the fuzzy control. After that, fit the lane model by using RANSAC algorithm in the set of effective lane boundary points. The experiment shows that the method proposed in this paper has high robustness and effectiveness which can accurately detect the lanes in complicated city road.

Active distribution network is a development mode of future distribution network and a new power distribution system adapting to the active control and management development trend of distribution network. Therefore, this network can effectively cope with large-scale and high-permeability grid connection of distributed energy resource. The Comprehensive Demonstration Project of Active Distribution Network in Suzhou is an engineering practice of active distribution network technology application. This paper describes overview and implementation plan of the demonstration project and explains with emphasis the key technologies for coordinated control of the active distribution network from such aspects as situation awareness, flexible DC, and coordinated control of power grid, power source and load. Through construction of the demonstration project, it is able to realize flexible absorption of high proportional and distributed energy resource in the demonstration zone, intelligent allocation of high-quality electric energy, coordinated and optimized control of multiple resources from power grid, power source, load and stored energy. In addition, a new DC power supply mode has been explored and developed for the distribution network, providing a reference and example for in-depth development and engineering practice of the active distribution network.

UHVDC commutation failure and DC block need to be studied in detail, for they will evidently affect the safety and stability of AC power grid. Firstly, the main circuit and control system are studied and analyzed in depth. Then, the simulation model is established in PSCAD software. After that, the curves of commutation failure and DC block in different cases are obtained. Finally, the correctness of this simulation platform is verified by comparison of some recording curves.

Power electronics based load equipment has now more and more used in power grid these years. Their frequency characteristic is worthy of study in detail for it is very different from that of traditional load equipment, and the regular load model may not be suitable such as in the security and stability analysis of modern power system. Firstly, so much these new-type load equipment are classified in several typical categories. Then, response curves are obtained from physical experiment. Thereafter, the static frequency characteristic of new-type loads is discussed and the trend of parameters is predicted. Finally, compared with the recording curves of a real bulk power grid during large frequency disturbs, the change of parameter is verified and the potential impact on power grid is analyzed.

This paper discusses an automatic parking control method based on the combination of the sliding mode variable structure control (SMVSC) and fuzzy logical control. SMVSC is applied to drive the vehicle from a random initial position and pose, to the designated parking position and pose. Then, the vehicle is driven from the designated parking position to the target parking slot using the method of fuzzy logical control, whose rules are limited to the range of the effective initial position. To combine SMVSC with the fuzzy logical control, the experimental results demonstrate that effective parking can be guaranteed, even if the initial position is out of the effective parking area of the fuzzy logical control.

Considering the influence of the limited measurement data on the effectiveness of the image reconstruction algorithms in industrial multiphase flow tomography, an iterative filtering reconstruction algorithm is proposed by redefining inverse problem in this article. In order to improve the image quality in iterative reconstruction algorithm, Wiener filter and soft-threshold algorithm are employed as constraints for the reconstructed image. The developed algorithm is applied to gamma-CT based two-phase flow measurements, both simulated and experimental. The results demonstrate that the proposed method cannot only realize the flow regime identification on horizontal pipe, but also improve the iterative reconstruction process, i.e., the reconstructed image quality is enhanced and the iteration number is reduced.

Objective: Realize the accurate segmentation and tridimensional visualization of cardiac medical image and finish the design of cardiac aided diagnosis. Methods: Combine clinical experts diagnosis experience, CT image prior feature and graph cut algorithm model, adopt GPUs parallel data processing to realize the segmentation of cardiac struction and tridimensional visualization. Result: Finish the accruate, quick and robust segmentaion and visualization of CT image sequence, initially realize the visualized cardiac aided diagnosis system based on GPUs. Conclusion: The study makes full use of the powerful parallel comouting capacity of computer graphics processing unit, solves the problems in medical image processing and segmentaiton, and improves the operating efficiency of the program and the uer experience.

Electrical capacitance tomography (ECT) is gaining acceptance as a laboratory and industry tool to analyze gas/liquid two phase flow. To overcome the shortcomings of the existing analogue ECT system, a new 16-electrode digital electrical capacitance tomography system based on FPGA and DSP is designed. The new scheme can easily realize single frequency and multi-frequency excitation and demodulation. Furthermore, two kinds of methods for multi-frequency demodulation, DFT demodulation and digital quadrature demodulation, are discussed deeply. Static and dynamic experiments under condition of single frequency are carried out to verify the performance of the digital ECT system. Experimental results show that this system is of high precision and stability.