Applied Mechanics and Materials Vols. 602-605

Abstract: Use the photoelectric device-linear CCD as a measurement tool, and use the SCM for the data acquisition and data processing, a diameter measurement system which is a low costing, high-precision, non direct contact, can be created. The system can provide diameter real-time monitor to pipe production process.

Abstract: The traditional temperature and humidity monitoring system of grain depot used to realize in wired network which needs to deploy a large number of cables. The maintainability and scalability is poor. We present a solution of monitoring temperature and humidity of grain depot based on wireless sensor network (WSN). The temperature and humidity data acquisition and control node is composed of digital temperature and humidity sensor and RF wireless chip and the main control chip. Based on the Zigbee technology, these nodes construct a WSN in Ad Hoc in order to collect, monitor and control the temperature and humidity in the monitoring area. The convergence node in the WSN will packed acquisition data from other nodes and then transmit them to a remote monitoring host through the GPRS wireless mobile network. By means of the statistical analysis of the real-time data collected, the monitoring host can alarm and feedback control the temperature and humidity of grain depot. The experimental results show that the system has strong scalability, low power consumption and high efficiency.

Abstract: The safety and reliability of large-scale visual vehicle monitoring system receives the user's attention. In this paper, the status of the development of vehicle monitoring system is analyzed, the necessity of developing scalable component-based vehicle monitoring system is discussed and a practical extensible system model based on component development is proposed. Through this article, it can demonstrate that the design meets to the performance requirements.

Abstract: The accurate method for Trojan group detection under cloud computing environment is studied in this paper. For the problem of Trojan group detection under the cloud computing environment, this paper proposed a Trojan group detection method based on the BP neural network. BP neural network model is constructed and the problem of Trojan group detection is acted on in this model. Experimental results show that the use of this algorithm for Trojan group detection can get the accurate detection results. Thus, it can open the network protection mode to ensure the user's network security and prevent the information loss caused by that the network is intruded.

Abstract: Zinc acetate serves as the catalyst in the synthesis reaction of VAc, the activity of zinc acetate will lose with the changes of the using time, temperature, space velocity, molar ratio. For the realization of soft measurement of the catalyst activity, the mathematical regression method and support vector machine (SVM) method are combined to model. By adopting the method of mathematical regression, the main change trend of catalyst activity can be illuminated with the change of response time; The support vector machine (SVM) method is used to amend the main trend to truly achieve the accurate and reliable soft measurement of catalyst activity^[1]。The simulation is proceed by the MATLAB to compare to the field datas, the simulation results show that the model established by the mixed modeling approach is high-precision and reliable. The requirements of the soft measurement for catalyst activity in VAc synthesis can be satisfied.

Abstract: In this paper, the modified EDF test procedures for testing the normal and exponential distributions with unknown parameters are suggested. The Monte Carlo algorithms are given to approximate the critical values of the EDF test statistics for a wide range of sample sizes. The power simulations show that the Anderson-Darling (AD) test is on the whole better than other EDF tests considered, in particular when the alternative departs from the true distribution in the tails.

Abstract: This research described in the paper proposed a device combining wireless sensor technique, GPS and wireless charge technique. This device can be installed nearby the vehicle tailpipe and transmit measurement data of exhaust gas to the backend system in real-time. It facilitates knowledge of exhaust gas the vehicle generated on the route it passed. With this information, a reference system is developed to estimate gas pollution that will be generated by the vehicle on each route. An artificial intelligent method, case-based reasoning is applied in the system design. Meanwhile, a new mathematical model can be built with additional consideration for gas pollution in route planning and optimization of vehicles. Some optimization approaches, including genetic algorithm, and some other combination algorithm can be applied to get solution of this model with new additional constraints.

Abstract: Soil and water loss monitoring is an important part of work for soil and water conservation, the monitoring and assessment of soil and water loss in large area is related to many security field. The remote GPRS network control system is used for large area soil and water loss monitoring in this paper, monitoring system is designed and implemented for soil and water loss based on dynamic priority mountains of GPRS remote network. The hardware and software of the system is realized, in the design of software, the particle swarm optimization algorithms is used to analyze some information distortion, the calculation of the loss is optimized and it is more accurate. Soil and water loss in Nanjing area with a wide range of incentives in the mountain is taken for test, the system can give the mountain soil erosion data monitoring in real-time, the monitoring area is more than 10 square kilometers with signals not break, and the accuracy is better.

Abstract: With the development of network, an increasing amount of broadcast television information transforms from simulation into digit, which therefor make the security of media information an imminent issue to be concerned. In this paper, a new kind of intrusion detection model was designed for the media information security system. In the system, both the false alarm rate and missing report rate decreased by using support vector machine classification technique in this new model. As a result of the experimental results, our algorithm processed a high classification accuracy and efficiency.

Abstract: As the development of information and automation, expense and aftereffect are severer caused by software defect. This paper represents synthetic retrospect of advanced research process, including data and example of system defect, data-mining system, software test management and standard procedure at domestic and overseas. Employing hybrid automation to obtain location, classification and description of system defect in tests, automation system is established for creditability measurement, standard evaluation with dominance of exploiting effectively and extensible, without heavy workload.