Papers by Keyword: Wavelet Neural Network (WNN)

Abstract: The micro-vibration mechanism and fault characteristics of micro-gears was described and the faults were classified with no fault, gear crack, gear face wear, tooth face attrition, tooth face crack. The wavelet neural network was proposed and optimized with differential evolution algorithm. The test was taken with the diagnosis information acquired with vibration experiment and designed as training samples which was normalized for wavelet neural networks, the simulation was taken under MATLAB and the simulation result shows the new algorithm with convergence quality and higher diagnosis precision.

Abstract: Health assessment is one of the key technology in the aircraft operating system. Aiming at the characteristic of aircraft structure, the aircraft fault prediction method based on data mining is presented in this paper. The concept of health assessment is introduced first, the wavelet neural network provide the mathematical model reflecting aircraft health state. The experiment results show that the health prediction applying wavelet neural network works well with high fidelity and real time. Focusing at a typical heavy-duty gas turbine, the critical information collected by the sensor is applied as the network input, then the wavelet neural network is constructed, the quick training and learning speed is proved. The results indicate proposed approach is promising for reliable diagnostics of aircraft.

Abstract: This paper first briefly discussed the basic principle of wavelet neural network, and pointed out the shortcomings and deficiencies existing in wavelet neural network , thus put forward the improved artificial fish-swarm neural network model, and solved the disadvantages of wavelet neural network in the training process. Finally the improved artificial fish-swarm neural network model is applied to the prediction of coal demand in our country, and the predicted results can prove that the model is scientific and feasible.

Abstract: The blending of liquors is a key process in the production of liquors. According to time-frequency localization characteristics of the wavelet transform and advantages of the neural network such as ability to develop, fault-tolerance, self-adaptability, self-learning, and robustness, a mathematic model based on wavelet neural networks is proposed in liquor blending processes with the help of computer-aided design technologies, which makes liquor blending technologies more scientific.

Abstract: Aiming at the feature of failure occurs frequently, check links, difficult to position for channel equipment of measurement and control system of the ship-borne, according as to equipment index that reflects equipment performance-signal power level, analyzes channel equipment reliability by different mathematical methods, and based on the previous equipment test results of data analysis, presents the method of wavelet neural network, to analyze equipment reliability and forecast malfunction .

Abstract: In order to better describe the dynamic characteristics of aircraft through aerodynamic modeling, a Wavelet Neural Network (WNN) aerodynamic modeling method based on Kernel Principal Components Analysis (KPCA) is proposed. Firstly, the training samples are used to execute KPCA for extracting basic features of samples, and then using the extracted basic features, WNN aerodynamic model was established. The simulation result shows that, the modeling ability of the method proposed is better than that of another 3 methods. It can easily determine of model parameters. This enables it to be effective and feasible to establish the aerodynamic modeling for aircraft.

Abstract: Aiming at the problem that it is difficult to predict the highway traveling passenger volume (HTPV), a new prediction model of HTPV based on wavelet neural network (WNN) is proposed. A case study is given to verify the proposed model. The simulation results show that the WNN model has higher convergence speed and prediction precision than the traditional BP neural network model (TBPNNM), and has more practical values.

Abstract: The wavelet neural network (WNN) is presented in the paper which based on the features of wavelet analysis and artificial neural network, and applied to predict the application of compressed air for an enterprise. The result shows that the precision of this model is quite effective. The model , therefore can be use to industry.

Abstract: An improved large envelope nonlinear flight control method using active disturbances rejection control (ADRC) method and wavelet neural network is approved in this paper. Wavelet neural network is used to realize the inversion of the 6-DOF nonlinear airplane model. The wavelet neural network is optimized using simulated annealing particle swarm optimization algorithm to improve the approach precision. In order to improve the robustness and control performance in all disturbances, ADRC is used to realize the high precision flight control. The simulation results show that the large envelope flight controller has excellent control performance.

Abstract: The production forecast is a basic premise in the design of the oil exploration program. Accurate production predictions can provide guidance on the direction of oilfields exploration program adjustments and be able to determine the scale of mining of the oilfields. Because of the varying nature of geological reservoirs, oil field production forecast error is large. Because wavelet neural network is better features of convergence and dealing with complex geological conditions, so it can provide a more accurate prediction than the conventional prediction to heterogeneous reservoir. After the actual reference data are simulated, calculation error is very small, and to prove its production forecast can be used as the reference of the real reservoir exploitation.