Short Term Wind Speed Prediction Using Multiple Kernel Pseudo Inverse Neural Network

An accurate short-term wind speed prediction algorithm based on the efficient kernel ridge pseudo inverse neural network （KRPINN） variants is proposed in this paper. The use of nonlinear kernel functions in pseudo inverse neural networks eliminates the trial and error approach of choosing the number of hidden layer neurons and their activation functions. The robustness of the proposed method has been validated in comparison with other models such as pseudo inverse radial basis function （PIRBF） and Legendre tanh activation function based neural network, i.e., PILNNT, whose input weights to the hidden layer weights are optimized using an adaptive firefly algorithm, i.e., FFA. However, since the individual kernel functions based KRPINN may not be able to produce accurate forecasts under chaotically varying wind speed conditions, a linear combination of individual kernel functions is used to build the multi kernel ridge pseudo inverse neural network （MK-RPINN） for providing improved forecasting accuracy, generalization, and stability of the wind speed prediction model. Several case studies have been presented to validate the accuracy of the short-term wind speed prediction models using the real world wind speed data from a wind farm in the Wyoming State of USA over time horizons varying from 10 minutes to 5 hours.

DECOUPLING CONTROL OF TWO MOTORS SYSTEM BASED ON NEURAL NETWORK INVERSE SYSTEM

In accordance with the characteristics of two motors system, the unitedmathematic model of two-motors inverter system with v/f variable frequency speed-regulating isgiven. Two-motor inverter system can be decoupled by the neural network invert system, and changedinto a sub-system of speed and a sub-system of tension. Multiple controllers are designed, and goodresults are obtained. Tie system has good static and dynamic performances and high anti-disturbanceof load.

PARAMETERS INVERSION OF FLUID-SATURATED POROUS MEDIA BASED ON NEURAL NETWORKS

The multi- layers feedforward neural network is used for inversion ofmaterial constants of fluid-saturated porous media. The direct analysis of fluid-saturated porousmedia is carried out with the boundary element method. The dynamic displacement responses obtainedfrom direct analysis for prescribed material parameters constitute the sample sets training neuralnetwork. By virtue of the effective L-M training algorithm and the Tikhonov regularization method aswell as the GCV method for an appropriate selection of regu-larization parameter, the inversemapping from dynamic displacement responses to material constants is performed. Numerical examplesdemonstrate the validity of the neural network method.

Adaptive inverse control of air supply flow for proton exchange membrane fuel cell systems

To prevent the oxygen starvation and improve the system output performance, an adaptive inverse control （AIC） strategy is developed to regulate the air supply flow of a proton exchange membrane fuel cell （PEMFC） system in this paper. The PEMFC stack and the air supply system including a compressor and a supply manifold are modeled for the purpose of performance analysis and controller design. A recurrent fuzzy neural network （RFNN） is utilized to identify the inverse model of the controlled system and generates a suitable control input during the abrupt step change of external disturbances. Compared with the PI controller, numerical simulations are performed to validate the effectiveness and advantages of the proposed AIC strategy.

Improving seismic interpretation: a high-contrast approximation to the reflection coefficient of a plane longitudinal wave

Linearized approximations of reflection and transmission coefficients set a foundation for amplitude versus offset（AVO） analysis and inversion in exploration geophysics.However,the weak properties contrast hypothesis of those linearized approximate equations leads to big errors when the two media across the interface vary dramatically.To extend the application of AVO analysis and inversion to high contrast between the properties of the two layers,we derive a novel nonlinearized high-contrast approximation of the PP-wave reflection coefficient,which establishes the direct relationship between PPwave reflection coefficient and P-wave velocities,S-wave velocities and densities across the interface.（A PP wave is a reflected compressional wave from an incident compressional wave（P-wave）.） This novel approximation is derived from the exact reflection coefficient equation with Taylor expansion for the incident angle.Model tests demonstrate that,compared with the reflection coefficients of the linearized approximations,the reflection coefficients of the novel nonlinearized approximate equation agree with those of the exact PP equation better for a high contrast interface with a moderate incident angle.Furthermore,we introduce a nonlinear direct inversion method utilizing the novel reflection coefficient equation as forward solver,to implement the direct inversion for the six parameters including P-wave velocities,S-wave velocities,and densities in the upper and lower layers across the interface.This nonlinear inversion algorithm is able to estimate the inverse of the nonlinear function in terms of model parameters directly rather than in a conventional optimization way.Three examples verified the feasibility and suitability of this novel approximation for a high contrast interface,and we still could estimate the six parameters across the interface reasonably when the parameters in both media across the interface vary about 50%.