Real-time task scheduling for wireless networked control systems provides guarantees for the quality of service. This paper introduces a new model for joint network and computing resource scheduling ...(JNCRS) in real-time wireless networked control systems. This new end-to-end real-time task model considers a strict execution order of segments including the sensing, the computing and the actuating segment based on the control loop of WNCSs. The general JNCRS problem is proved to be a NP-hard problem. After dividing the JNCRS problem into four subproblems, we propose a polynomial-time optimal algorithm to solve the first subproblem where each segment has unit execution time, by checking the intervals with 100% network resource utilization and modify the deadlines of tasks. To solve the second subproblem where the computing segment is larger than one unit execution time, we define the new timing parameters of each network segment by taking into account the scheduling of the computing segments. We propose a polynomial-time optimal algorithm to check the intervals with the network resource utilization larger than or equal to 100% and modify the timing parameters of tasks based on these intervals.
In order to optimize the mapping between service composition and operational activities within the process of service oriented architecture design, a method based on the DODAF 2.0 is proposed to ...match operational activities with services. Firstly, related architectural description methods, including Object-oriented method and the formal method, are studied based on the DODAF2.0. Secondly, considering the probability and potential resource conflicts, basic process of Petri Net simulation model was created based on related architectural models, and the rationality of mapping from services to operational activities is verified by the execution of simulation models.
A low-loss, constant-frequency, zero voltage switching (ZVS) modulation strategy for bidirectional buck dc-dc converter, applied to hybrid energy storage system, is proposed in this paper. The ...synchronous rectification and soft-switching techniques are adopted to reduce the power conversion loss. As the modulation strategy is a software-only solution, there are no additional expenses for the active or passive components. The mathematical model and operation principle of the converter are presented respectively. Simulation and experimental results verified the theoretical analysis, and the highest efficiency of the proposed converter is about 95.5%.
Most of the innovation designs of mechanical and electrical products are willing to take an improved design of the present products, which not only inherit the advantages of the current products, but ...also improve the efficiency and reliability of the design. Optimum method of evolutionary design scheme by feature lines and colour based on CBR is studied. This paper provides the division and quantification of product form element, the method and process of extracting form feature lines of case products. The double models of BP neural network of corresponding relation of feature lines, colour parameters and form perceptual image, and of form perceptual image, product function, product capability and performance and product sales is founded, which is further taken as fitness function of GA. The application process of this method is illustrated by an example of innovation design of family car form.
A speed sensorless induction motor drive fed by a 3-level neutral point clamped (NPC) inverter is presented, which combines the principle of sliding mode controller (SMC), sliding mode observer ...(SMO), direct field oriented control (DFOC) and space vector modulation (SVM). PI is widely used in field oriented control for speed, flux and current regulation whereas the performance of drive is usually affected by the invariant parameters of PI. In this paper, SMC is introduced in stead of conventional PI to produce fast and accurate response and reject the external disturbance and motor parameters variation. Moreover, SMO is introduced to observe rotor speed and flux for the implementation of DFOC. The induction motor is driven by a 3-level NPC inverter, which provides more smooth and accurate response than conventional 2-level inverter, and the neutral point potential is well controlled using hysteresis control. By introducing sliding mode theory in both controller and observer, the system demonstrates excellent dynamic and steady-state performance, robustness to motor parameter variations and wide speed range operation, which is confirmed by simulation and experimental results.