Pseudo Inverse Neural Network Research Articles

Wind speed and power prediction using MK-PINN

A multi-kernel pseudo inverse neural network (MKPINN) is proposed in this paper for efficient wind speed and power forecasting. The proposed model has been compared with Gaussian, wavelet, polynomial and sigmoid kernel. To get best output and learning methodology and stability pseudo inverse neural network is added, which substitutes the hidden layer with kernel function. This helps to achieve more accurate and faster response. Various case studies have been carried out from ten minutes to five hours interval in order to prove it accuracy.

An effective battery management scheme for wind energy systems using multi Kernel Ridge regression algorithm

Abstract Battery Energy Storage (BES) systems are adequate alternative for any Wind Power Generation System (WPGS) for achieving greater operational flexibility by compensating the generation volatility. An efficient Local Energy Management (LEM) for Wind/ Battery Energy Storage (BES) system is described in this chapter to reduce the prediction error and to improve associated battery life. Error in prediction impacts on DG control reference & system stability for local energy management. BES life degrades by prediction error in terms of battery power loss and temperature. A Doubly-Fed Induction Generator (DFIG) based WPGS is considered as primary Distributed Generation (DG), where DC link voltage stability challenges due to wind speed volatility, as well as prediction error is addressed with Lithium-ion (Li-ion) BES stacks. A new online Multi Kernel Ridge Pseudo Inverse Neural Network (MK-RPINN) algorithm is proposed to obtain an impactful decrease in prediction error. A new secondary controller is proposed here for addressing temperature effect of battery. A Model Reference (MR) based battery temperature plan, linked to Rule based switching characteristics of battery stacks with temperature tolerance, is being incorporated in the proposed secondary controller. The effectiveness of this model is represented in various observations through TMS320 C6713 and MATLAB platform.

Short term wind power forecasting using hybrid variational mode decomposition and multi-kernel regularized pseudo inverse neural network

Abstract In this paper a new hybrid method combining variational mode decomposition (VMD) and single or Multi-kernel regularized pseudo inverse neural network (MKRPINN) is presented for effective and efficient wind power forecasting. The original non-linear and non-stationary time series data is decomposed using VMD approach to prevent the mutual effects among the different modes. The proposed VMD-KRPINN (VMD based kernel regularized pseudo inverse neural network) and VMD-MKRPINN methods are then used to predict wind power generation of a wind farm in the state of Wyoming, USA for different time intervals of 10 min, 30 min, 1 h and 3 h ahead. Comparison with empirical mode decomposition (EMD) based kernel regularized pseudo inverse neural networks is also presented in the paper to validate the superiority of the VMD based wind power prediction models. Also to improve the performance of the proposed EMD-MKPRINN and VMD-MKRPINN models, their parameters are optimized using vaporization and precipitation based water cycle algorithm (VAPWCA). Further a fast reduced version of the VMD-KRPINN is presented in the paper to reduce the execution time substantially using randomly selected support vectors from the data set while resulting in a reasonably accurate forecast.

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.

A low complexity nonlinear pseudo inverse neural network for short-term wind power prediction

This paper proposes three different efficient wind power prediction algorithms, i.e., Chebyshev neural network (CNN), Legendre neural network (LNN), trigonometric functional link artificial neural net (trigonometric FLANN), passed through functional expansion block (FEB) and trained by ridge pseudo inverse neural net (R-PINN), for 10 minutes, 30 minutes and 60 minutes ahead prediction. Learning speed rate and computational scalability are very essential in the prediction purpose. Conventional algorithms, such as hybrid models, statistical approaches, and intelligent systems based algorithms are less accurate because they require updation of hidden layer in each iteration. On the contrary, the proposed R-PINN-based prediction algorithm computes the output weight vector in a chunk, where the hidden layer is not being updated. Hence the essential features such as the learning speed and computational scalability has been significantly improved. This allows a faster response. In a comparison study all algorithms are verified using R-PINN technique and it is found that LNN model gives best output results among all algorithms. This is distinguished clearly in the MATLAB%Editor environment and has been illustrated in the simulation and result section by graphical representations.

A low complexity nonlinear pseudo inverse neural network for short-term wind power prediction

A low complexity nonlinear pseudo inverse neural network for short-term wind power prediction

Thouless-anderson-palmer equations for neural networks

Previous derivation of the Thouless-Anderson-Palmer (TAP) equations for the Hopfield model by the cavity method yielded results that were inconsistent with those of the perturbation theory as well as the results derived by the replica theory of the model. Here we present a derivation of the TAP equation for the Hopfield model by the cavity method and show that it agrees with the form derived by perturbation theory. We also use the cavity method to derive TAP equations for the pseudoinverse neural network model. These equations are consistent with the results of the replica theory of these models.

Categorization in the pseudo-inverse neural network

We investigate analytically the emergence of the categorization ability in the pseudo-inverse attractor neural network. More pointedly, we consider the problem of learning an extensive number of concepts by storing a finite number of examples s of each concept. We find that there is a critical value beyond which the categorization error, as measured by the average fraction of unstable sites in the concepts, decreases monotonically with s.