Applied Mechanics and Materials Vols. 128-129

Abstract: This paper discusses a face recognition method based on the fuzzy neural network (FNN). The fuzzy neural network has more advantages than artificial neural network alone. The paper firstly introduces the structure of the FNN. Than proposed the fuzzy rules and the study algorithm. Thirdly it researches on the process of face recognition. The experimental results prove that this method can achieve good location performance and good effect of extraction.

Abstract: Interferometric Synthetic Aperture Radar is one of the most important technologies for topographic mapping. The DEM quality of airborne InSAR system depends on both system hardware performance and data processing methods. To derive large scale topographic and thematic maps up to scale 1:50000 and 1:10000, the whole data processing methods were presented. The methods included SAR imaging, interferometric processing and cartographic processing. Special methods were induced to resolve the problems encountered in project applications. Results using X-band airborne InSAR system data showed validity of the algorithm.

Abstract: The high altitude airship can not have desired performance to control the altitude rapidly and accurately when the elevator or ancillary air bursa charge or deflation is used only, because the elevator has little efficiency when the velocity is low and auxiliary air bursas charge or deflation control is very slow. It is present a method to design flight control system for a high altitude airship using auxiliary air bursas charge or deflation and elevator combination control. This combination control scheme is that the ancillary air bursa and elevator are also used to control the airship attitude to get large raise velocity and the ancillary air bursa control is used to adjust the airship altitude for suspension. In this paper, a high altitude airship model with compound control of elevator and ancillary air bursa charge and deflation is given firstly. Then the combination controller is designed by using fuzzy self-tuning control. Finally, it has been proved by simulation that the flight control system has desirable performance and the compound control scheme is feasible.

Abstract: A optimization method of human-body ADL(activity of daily living) real-time monitoring system design method based on wireless sensor network is proposed, which is able to detect elderly people body posture and biological signal at rehabilitation centers and nursing homes and doctor or their family can know patients’ body state through mobile phone or PC. The number of nodes in each base station increase can cause network congestion. A optimization data transmission algorithm based on Bayesian network is presented, and sensors under the Bayesian network distribution model and algorithm are built. Finally, experiments results indicate that system’s real-time performance is improved and parameters of Bayesian network training method is effective.

Abstract: An IMC-PID based on V-norm decoupling for multivariable stable process with time delay was proposed in this paper. Previous multivariable IMC is usually designed after the decoupling of the process, however the industrial process modeling is difficult to be precise, so the controller design is often not ideal after decoupling. The method in this paper designs both the decoupling and controller at the same time, the IMC is not only a controller, but also a decoupling compensation. A filter parameter optimization method based on a new performance function was adopted. The controller can be realized by time delay compensation when there is an advance part. The IMC-PID designed by this method has strong robustness when the model is mismatch and excellent performance both in decoupling and control. The simulation results show that the method is very effective and has a wide range of application in actual industry area.

Abstract: Ensemble Empirical Mode Decomposition (EEMD) is a new noise-assisted data analysis (NADA) method. The effect of EEMD depends on two key parameters which are the amplitude of white noise and the ensemble times. However, the shortcoming of EEMD is that it lacks adaptability and reliability because these two key important parameters are obtained by experience and human intervention. An Improved Ensemble Empirical Mode Decomposition method is proposed in this paper, by adding white noise and ascertaining ensemble number adaptively. The criterion of adding white noise in Improved EEMD is established, by which a composite simulation signal could be adaptively and accurately decomposed into IMFs without mode mixing. The proposed method is applied to a gear fault detection of hot strip finishing mills. The result shows that Improved EEMD method successfully extracts the gear fault feature with high precise diagnosis results.

Abstract: Wavelet image de-noising has been well acknowledged as an important method of de-noising in Image Processing. Lifting scheme is not only a fast algorithm of existing wavelet transforms, but also a tool to produce new wavelet transforms. In this paper, the principle of several wavelet de-noising algorithms are described, and we compares with these algorithm, gives three kinds of improved algorithm. The simulation experiment shows that it is practicable and effective.

Abstract: This paper proposes an intelligent Built-in Test (BIT) technology based on wavelet packet analysis and gray neural network. The aim is to improve the fault diagnosis and prediction capability of intelligent BIT. Firstly, the energy of each frequency-band was computed to form the eigenvectors by using the wavelet packet decomposition, then the energy eigenvectors were used as samples to the forecasting model, which were based on wavelet packet analysis and gray neural network. Finally, the proposed method was applied to the BIT system of the airborne mechatronics, and the results have shown that the proposed method could improve the performance of the intelligent BIT system.

Abstract: Intelligent control methods of missile guidance and control system (GACS) are studied in this paper. Secondly, the component and principle of GACS is introduced. Based on the fuzzy neural network, this paper constructs a basic structure of the intelligent control method of missile. Meanwhile, a new intelligent control method of rolling channel of missile based on Fuzzy Cerebella Model Articulation Controller (FCMAC) is designed. Under complicated environmental conditions, the missile can be accurately controlled with this method. Finally, the application value is illustrated. It’s very meaningful to improve the combat capability.

Abstract: According to the urgent requirement of water quality monitoring, water bloom prediction and emergency treatment decision-making, a multi-functional information system based on GIS (Geographic Information System) is designed and developed, which integrates water quality monitoring and analysis, water bloom prediction and emergency treatment decision-making. VS2010 platform and SQL Server 2005 database are used in the system, and the remote data communication is realized by GPRS. The system has functions such as lakes and reservoirs’ water quality real-time monitoring, teletransmission and location, water quality data analysis, water bloom prediction and emergency treatment decision-making. The medium-term prediction of water bloom is realized by wavelet neural network, and the optimal choice of water bloom emergency treatment decision-making projects is realized by multi-attribute decision-making theory. This design offers effective decision-making system for environmental protection departments, which also has significant meaning to the protection and renovation of water environment, and its application and promotion have considerable social and economic benefits.