Applied Mechanics and Materials Vols. 58-60

Abstract: Shikaku is a logic puzzle published by Nikoli at 2005. Shikaku has a very simple rule. This puzzle is played on a rectangular grid. Some of the squares in the grid are numbered. The main objective is to create partitions inside the grid. Each partition must have exactly one number, and the number represents the area of the partition. Then the partition’s shape must be a rectangular or a square. The aim of this research is discussing how can computer software be able to solve the Shikaku problem by implementing heuristic technique and genetics algorithms. Initially the Shikaku problem is inputted into the system. Firstly, the software will solve the problem by applying heuristics methods with some logic rules. All logic rules are created and implemented into the software so that the software can minimize the partitions possibilities to the problem. If this heuristics method still can not solve the problem then genetic algorithms will be executed to find the solution. This paper elaborates from how the problem be modelled and also be implemented until software testing to ensure that the solver worked as expected. The implementation consists of a virtual puzzle board with three different size, genetic algorithms parameters, and ability to create, save, load, and solve puzzle. Software testing is conducted to find how fast the system can solve the problem.

Abstract: It is important to own the capability of debugging application and system software and multitasking system in the design of microprocessor at present. In this paper, we proposed a debugging structure based on JTAG port and combined with boundary-scan technology and interrupt system, which can allow users to debug software procedure by the debugger which is connected to JTAG port. It is easy to form a standardized structure and applicable to all general-purpose CPU and only increases the difficulty of the circuit logic design.

Abstract: Some projective points (2D-points) of the same 3D-point captured by different cameras are called corresponding points. The task of finding the corresponding points is named as point correspondence. It is a essential problem in computer vision, and it paves the way for 3D reconstruction. Epipolar constraint is a widely-used geometric constraint for point correspondence, some methods based on it have been proposed for this task. However, the threshold is set by empirical approach in these methods; this approach is influenced by the parameters of cameras. On the other hand, epipolar constraint is not a sufficient condition for this task; this will result in the mismatch for two 2D-points. According to the rule of Propagation Uncertainty and practical situation of the computer vision system, this paper introduces the enhanced epipolar constraint comprising normalized coplanar constraint, nonparallel constraint and interval constraint. The normalized coplanar constraint provides a simple and rational method to set the threshold of epipolar constraint; the nonparallel constraint and interval constraint can reduce the occasion of mismatch.

Abstract: A new idea, using chirplet as the staff to define fractal dimension, is proposed in this paper, based on self- similitude of knowing essence of things from collectivity to part, from macroscopy to microcosm, in fractal theory and chirplet transformation. Chirplet fractal dimension is defined as the sum of high-frequency values of decomposed signals. The edge of infrared image is detected through chirplet fractal dimension, experimental results show that this new algorithm is simple and effective to detect whole contour and detail information, and is better than other traditional operators.

Abstract: A new method for image edge detection based on envelope curve of histogram is proposed, aimed at the characteristics of high voltage transmission line. In order to inhibit noise influence on infrared image, envelop curve of histogram, got from Savitzky-Golay(S-G) filter, is used to smooth the image, and the edge of infrared image can be detected based on the extreme points of the envelop curve. Experimental results show that this new algorithm is simple and effective to detect whole contour and detail information, and is better than other traditional operators.

Abstract: Firstly, this paper analysed the existed problem on traditional controlling way of city water supply system, then elaborated the controlling principle of frequency conversion. Finally, a designed practical example by frequency conversion technical is given on the water supply system. It is proved that frequency conversion can maintain constant pressure water, and it is a very ideal efficient measure for energy saving. Therefore design and transformation on water supply system has a broad market application prospect.

Abstract: This paper designs a GPRS-based system to effectively monitor and control large scale Photovoltaic Power Plants. Single-Chip microcomputer with high performance is utilized to collect data from different types of photovoltaic devices; the collected data are then sent to the server via GPRS; the server receives and intelligently analyzes the data. The operators can monitor and control the running status of photovoltaic devices by executing a simple client tool. The implemented system has been applied in several Photovoltaic Power Plants.

Abstract: Referring to the military sonar hunting submarine principle，the author converts the multi-constraint issue into the multidimensional space search issue and calculates the differences between the current solution and expectation in each dimension to determine the direction and the scope for the next search. Then the probability of better solutions and the efficiency are both improved. This article mainly make experiments in the auto-generation of test paper. The auto-generation of test paper is proved to meet the needs in practice.

Abstract: The intrinsically redundant series manipulator’s kinematics were studied by the exponential product formula of screw theory, the direct kinematics problem and Inverse kinematics problems were analyzed, and the intrinsically redundant series manipulator’s kinematics solution that based on exponential product formulas were proposed; the intrinsically redundant series manipulator’s kinematics is decomposed into several simple sub-problems, then analyzed sub-problem, and set an example to validate the correctness of the proposed method. Finally, comparing the exponential product formula and the D-H parameters, draw that they are essentially the same in solving the manipulator’s kinematics, so as to the algorithm of the manipulator’s kinematics based on exponential product formulas are correct, and the manipulator’s kinematics process based on exponential product formula is more simple and easier to real-time control of industrial.

Abstract: In order to settle such a problem that the multi-sensors data fusion results are not good due to data confliction in the coal-rock interface recognition, the paper first carries out the fusion with D-S evidence theory. The fusion results are not correct when there are high-conflicting in the evidence, so a distance function is introduced and weight fusion correction algorithm is put forward. Through test simulation, fusion results respectively with D-S evidence theory, weight correction algorithm and fuzzy neural network are analyzed. The results show: the good results are achieved in the multi-sensor data conflict of coal-rock recognition through weight fusion correction algorithm, and influence of signal conflict is avoided effectively.