Search

In this paper, a new method to design a wireless power supply system for trackless trolley car is proposed. Electric field resonant Wireless power transfer (WPT) technology is a method for WPT, which releases a high-frequency electric field in the air. Any conductor, with the same resonant frequency as the frequency of electric field, would receive power with the highest efficiency. For one emitting device, electric field released by the device is centrally symmetric in the air. Based on that, a wireless power supply system is designed for movable devices such as trackless trolley car. Working principle of the system is studied in this paper including analysis of lumped parameter circuit and electromagnetic field model. Two Tesla coils are made for system design and experimented. As a result, this system can transfer power as 18W with efficiency of 15% within a radius of 1.2m. There are some more extra experiment such as quasi-single-wire power transfer and detection of electric field in the air.

Ultrasonic wireless power transfer (UWPT) is a reasonable method to power loads in enclosed metal container. For the existing open-loop UWPT system, the output voltage varies with different loads. In this paper, a closed-loop UWPT system with constant output voltage for different load resistances is proposed. Two pairs of piezoelectric transducers with different resonant working frequencies are adopted, for which one pair is used for power transfer and another is used for voltage signal feedback. PI control algorithm is used to dynamically adjust the PWM duty of the boost circuit, which is able to regulate the input voltage of the high-frequency inverter. As a result of which, the output voltage of the system can be kept constant when the load resistance varies. The experiment validates the proposed system.

A side-scan sonar image often contains a lot of noises and its resolution is very low, the edge of side-scan sonar image is blurred, so these characteristics make it difficult to segment objects. The current common segmentation algorithms generally require setting parameters manually, and these parameters are closely related with the images they collected. Consequently, it is difficult to achieve universality and autonomy. According to the properties of segmentation algorithm and characteristics of side-scan sonar image, this paper proposes an algorithm based on unsupervised feature learning. We use sparse auto-encoder to learn the local binary patterns and Haar-like feature of the side-scan sonar images, then we build an exclusive feature for them. In the end, we test this method by using clustering algorithm based on features we built previously to segment several side-scan sonar images. The results show that the algorithm we proposed has better performance in both adaptability and effectiveness.

Dynamic Data Driven Application System (DDDAS) is a new paradigm for research of interdisciplinary complex problems which entails the ability to incorporate additional data into an executing simulation application and, in reverse, the ability of applications to dynamically steer the measurement process. It needs the support of online simulation to assist decision makers to complete their tasks in time. To facilitate the building of such environments, levels of adaptability and challenging technologies of collaborative DDDAS environment are analyzed and a DDDAS M&S reference framework based on parallel and distributed simulation (PADS) is proposed. Realization mechanism relating to this framework, including multi-paradigm modeling based on Discrete Event System Specification (DEVS), dynamic reconfiguration based on flexible points binding, component substitution and dynamic compilation are also discussed.

Dynamic reconfigurable simulation based on Discrete Event System Specification (DEVS) requires efficient verification of simulation models. Traditional verification method of DEVS model is based on I/O test in which a DEVS model is regarded as a black box or a grey box. This method is low efficient and insufficient because input samples are often limited. This paper proposes a formal method which can translate Parallel DEVS model into a restrict kind of Timed Automata (TA) with equivalent behaviors. By this translation, a formal verification problem of Parallel DEVS model can be changed into the formal verification of according timed automata.