Application Of Polyploid Based Genetic Algorithm In Gravity Solid Tide Signal Analysis

Study on gravity tide signal can be used in many fields, especially for earthquake precursors. The tidal harmonic waves of gravity tide signal reflect the effects caused by the position change of earth, moon and sun, and these waves are independent components. So, in the analysis of the tidal harmonic waves of gravity tide signal, it is should be extract the independent components, in order to mapped to the gravity tide effects.A orthogonal decomposing model which describes the gravity tide effects caused by the position change of earth, moon and sun in three directions is introduced in order to better analysis of the gravity tide signal.The signal processing used to analyze the gravity tide signal can realize the relation described in the model. The gravity tide signal is a multi-harmonic and mixed signal, therefore, using independent components analysis to extract the independent components form the gravity tide signal. Combined with the intelligent optimization algorithm, a new independent components analysis is be proposed called independent components analysis based on the improved genetic algorithm. This kind of improved genetic algorithm with real-coded crossover on random gene and polyploidy strategy (RGPGA) is proposed, based on improve in crossover, searching and save strategy to repress the premature convergence. The RGPGA is applied to solve the benchmark problems. The effectiveness of independent components analysis based on the improved genetic algorithm is also be proved by the experiment.Finally, using the independent component analysis based on the improved genetic algorithm, three independent components are extracted from gravity tide signal, which are respectively reflect gravity tide effects caused by the position change of earth, moon and sun. The tidal harmonic waves of gravity tide signal are also extract from the three independent components. And then the algorithm is applied to specific measurement signal.