Applied Mechanics and Materials Vols. 668-669

Abstract: In complex battlefield environment all kinds of the uncertainties for target tracking make the analysis and modeling of tracking system irresolvable. Aiming at different levels of system uncertainty, the correspondent modeling methods are introduced. First, the stochastic process model is investigated which includes target process equation and sensor measurement equation. Second, joint state-class probabilistic density is given for realizing target tracking and classification simultaneously. Finally, the joint target detection, tracking and identification based on finite-set statistics theory is represented for multi-sensor multi-target tracking.

Abstract: An optical rain sensor which can be used in automobile is described in this paper. The sensor is mainly composed of light source, light collimation system and light focusing system. When it rains, the rain sensor detects the rainfall on the automobile’s windshield surface through the light from the transmitter, and the light intensity received by the sensor receiver can be used to control the wiper to work automatically. The structure of the sensor is designed and the corresponding optical simulation is finished. With the optical rain sensor, the wiper device can work automatically and adjust the swing speed with the rainfall. The performance of automatic control of automobile and the driving safety are also improved.

Abstract: In order to clear the timeliness issues of metal magnetic memory signal value after unloading, we conduct some static tension experiments on some specimens in materials of KMN, which is commonly used in centrifugal compressors Furthermore, we analyze the specimen, which is prefabricated artificial defects, on the surface normal magnetic field strength decay law with EMS-2003, the intelligent magnetic memory tester. The results show that: the surface magnetic field signal value is reduced by time after unloading, in other words it has certain timeliness.

Abstract: In order to obtain spectra from EFPI interference with the noise in the signal spectrum which can get demodulation precision cavity length, HHT method is proposed based gas detection algorithm processing EFPI fiber optic signals. The principle of HHT algorithm is to deal with the noise spectral signal EFPI through empirical mode decomposition (EMD), a group based on the signal obtained their intrinsic mode function (IMF), through to the IMF for Hilbert transform the Hilbert spectrum for time-frequency analysis. Through simulation and numerical analysis, the algorithm to solve the original non-stationary signal interference spectrum , nonlinear shortcomings, through a series of changes to remove noise, which can reduce the difficulty of data processing and increase the accuracy of the gas signal analysis .

Abstract: In order to develop north seeking system with strong anti-interference ability, the influence caused by the swaying base on the performance of north seeker was investigated. The influence of sinusoidal signal of gyros and tilt sensor's output was analyzed based on four-position north seeking system which was consisted of a single axis gyro and a tilt sensor. Fiber optic gyros and tilt sensor are interference by the same vibration signal, according to the relationship between the output signal of gyro and the angle sensor, a PI controller was designed to compensate the gyro's output error. This real time compensation can effectively improve the north seeking accuracy and speed without complicated calculation or external reference signal. The north seeking accuracy is better than 0.03^o. Simulation and actual north seeking result proved that this method significantly improves the north seeking performance in vibrant environment.

Abstract: This study aims at providing a back propagation-artificial neural network (BP-ANN) model on forecasting the water quality change trend of Qiantang River basin. To achieve this goal, a three-layer (one input layer, one hidden layer, and one output layer) BP-ANN with the LM regularization training algorithm was used. Water quality variables such as pH value, dissolved oxygen, permanganate index and ammonia-nitrogen was selected as the input data to obtain the output of the neural network. The ANN structure with 17 hidden neurons obtained the best selection. The comparison between the original measured and forecast values of the ANN model shows that the relative errors, with a few exceptions, were lower than 9%. The results indicated that the BP neural network can be satisfactorily applied to forecast precise water quality parameters and is suitable for pre-alarm of water quality trend.

Abstract: Fan occupies the important position in many industry, it give rise to that fault diagnosis become the new hot research topic, also is the urgent demand of many manufacturing enterprises. This paper based on the theory of wavelet packet transform, selecting wavelet packet transform and energy spectrum to wavelet de-noising and fault feature extraction the fan vibration signal. And use the MATLAB get the fan vibration signal characteristic vector, lay the foundation for the fan fault diagnosis.

Abstract: This paper discusses the body posture detection problem using low cost Micro-Electro-Mechanical System (MEMS) inertial sensors, for which a complementary sensor fusion solution is proposed. Considering the impact from the noise and bias drifts, through Kalman filter to complete the multi-sensor information fusion, achieved an accurate attitude determination. The experimental results show that, after using Kalman filtering algorithm to fuse acceleration sensor and signal gyroscope, it can effectively eliminate the accumulative error and significantly better dynamic characteristics of attitude angle measurement, Improving the reliability and accuracy of body posture estimation.

Abstract: Among all the kinds of optical fiber sensors, FBG and LPFG become the main current owing to its features and advantages. LPFG sensor is especially suitable in high temperature sensing and stimulates more related research interests. The key point of LPFG sensing is the signal demodulation and processing. In this paper, signal demodulation of the LPFG strain sensor was studied based on Arrayed Waveguide Grating. The fabrication method and high temperature performance of this sensor were presented,and a novel strain signal demodulation method was presented based on Arrayed Waveguide Grating (AWG) device by equi-intensity cantilever beam. The experimental results demonstrated that the signal demodulation technology based on AWG has many advantages, such as simple structure, innovative and reliable performance.

Abstract: Gallium nitride (GaN) is the third generation of semiconductor material; it has a large band gap, high thermal conductivity, low dielectric constant, high drift speed, etc. Radiation detectors based on GaN material have small volume, high radiation resistance, and fast response, can be used to replace the existing Large Hadron Collider vertex detector and track detector. Energy deposition characteristic of GaN detectors to radiation beam is an important factor for detection efficiency, and there are many factors that affect the energy deposition characteristics of the detector, like the detection mechanism, the impact of material properties, the type of incident ray, radiation energy, and many other factors. This paper studies the physical properties of GaN detector by calculation based on Monte Carlo simulation. Energy deposition characteristics are discussed respectively for incident γ-ray with different energy, in the front-end and back-end add PTFE material. The results of our study present the theoretical properties of GaN radiation detectors.