Search

We considered a sensor formation problem for target localization based on the time difference of arrival (TDOA) data. It is assumed that the target is located far off from the sensors while the sensors are relatively close to each other. The estimators of the target location expressed in terms of the spherical coordinates are found to be uncorrelated when the sensors are arranged in a concentric ring formation. The proposed optimal formation of sensors is in the concentric ring formation and is shown to change in accordance with sensors' angular positions with respect to the line-of-sight vector from the reference sensor to the target. The proposed optimal sensor formations are compared in the context of estimation accuracy both in theory and by numerical experiments. They are shown optimal in general through refined numerical experiments. [ABSTRACT FROM AUTHOR]

In this paper, we propose a neural tree classifier, called the convex objective function neural tree (COF-NT), which has a specialized perceptron at each node. The specialized perceptron is a single layer feed-forward perceptron which calculates the errors before the neuron’s non-linear activation function instead of after them. Thus, the network parameters are independent of non-linear activation functions, and subsequently, the objective function is a convex objective function. The solution can be easily obtained by solving a system of linear equations which will require less computational power than conventional iterative methods. During the training, the proposed neural tree classifier divides the training set into smaller subsets by adding new levels to the tree. Each child perceptron takes forward the task of training done by its parent perceptron on the superset of this subset. Thus, the training is done by a number of single layer perceptrons (each perceptron carrying forward the work done by its ancestors) that reach the global minima in a finite number of steps. The proposed algorithm has been tested on available benchmark datasets and the results are promising in terms of classification accuracy and training time. [ABSTRACT FROM AUTHOR]

In the preset study, a new identifier system has been proposed for optimal model estimation of Solid Oxide Fuel Cell (SOFC) stacks. To afford an appropriate identification system, the voltage-current profile of the system has been considered. The main idea is to minimize the Mean Squared Error (MSE) value between the actual output voltage of the stack and the output value achieved by the proposed model. Here, the MSE minimization has been established based on a new improved metaheuristic technique, called Fractional-order Social Network Search algorithm. The main advantage of the proposed technique is that it provides better trade-off between the global optimum and local optimum values. After system designing, it has been implemented to a studied case and its results are put in comparison with several latest techniques with considering two scenarios. One scenario with constant pressure and variable temperature and one other scenario with constant pressure and variable temperature. Final results indicate that the presented identification system provides satisfying results against the other compared methods in optimum system identification of the SOFCs. • New model identification of Solid Oxide Fuel Cell (SOFC) stacks based on a new metaheuristic. • Fractional-order Social Network Search algorithm is used as the new metaheuristic technique. • Voltage-current profile of the system is considered for identification. • The idea is to minimize the MSE between the model output data and empirical data. [ABSTRACT FROM AUTHOR]