This paper is a report of an assignment of a course Computational Intelligence. The main goal of the assignment is to apply computational intelligence techniques in a practical setting - building a controller for a race car in The Open Racing Car Simulator (TORCS) using artificial neural network and evolving that network with evolutionary algorithm techniques.
Keywords: computational intelligence, car controller, artificial neural network
This work is going to analyze the Forced Van Der Pol's Equation which is used to analyze the electric circuit. In 1927, Balthasar Van Der Pol observed the stable oscillation and heard some irregular noise in vacuum tube circuit. He then proposed the Forced Van Der Pol's Equation to analyze the circuit and suggested the concept of limit cycle and Chaos to explain his observation. In this work, We would like to analyze behavior of the model by observing the phase space, time series, bifurcation diagram and power spectrum. Those points in the figures are calculated by Runge-Kutta Method with the aid of MATLAB. For a better understanding, the RLC circuit, which is a electric circuit consisting the a resistor, an inductor and a capacitor, will be used as an example for explaining the properties. Apart from electric circuit, the Forced Van Der Pol's Equation can be applied to dynamic systems in different aspect, such as the artificial heart, economic market and so on. Therefore, we suggest this work to all students since the Forced Van Der Pol's Equation can be applied to many majors, like Mathematics, Physics, Economics, Sociology, Biology, Engineering and so on.
Principal Components Analysis (PCA) and Canonical Correlation Analysis (CCA) are among the methods used in Multivariate Data Analysis. PCA is concerned with explaining the variance-covariance structure of a set of variables through a few linear combinations of these variables. Its general objectives are data reduction and interpretation. CCA seeks to identify and quantify the associations between two sets of variables i.e Pulp fibres and Paper variables.PCA shows that the first PC already exceeds 90% of the total variability. According to the proportion of variability explained by each canonical variable , the results suggest that the first two canonical correlations seem to be sufficient to explain the structure between
Pulp and Paper characteristics with 98.86%. Despite the fact that the first the two canonical
variables keep 98% of common variability, 78% was kept in the pulp fiber set and about
94% of the paper set as a whole. In the proportion of opposite canonical variable,there were
approximately 64% for the paper set of variables and 78% for the pulp fiber set of variables
kept for the two respectively.