Forward Kinematics of the Variable Vector Propeller Based on HGANN

The Variable Vector Propeller(VVP) Controllable Pitch Machine(CPM)´s unique structure presents an particular problem in its forward kinematics(FK) solution. It involves the solution of a series of simultaneous non-linear equation and, usually, non-unique, multiple sets of solutions are obtained from one set of data. This article proposed modified Hybrid encoding Genetic Algorithm Neural Network (HGANN) for solving the FK problem of the VVPCMP. In this article proposed the algorithm concurrently has had the genetic algorithm overall situation optimization ability and the neural network approaches ability formidable regarding the non-linear mapping. Simultaneously has used the binary system and the real number hybrid encoding scheme cooperate with the 3 chromosomic structures, modifies the GANN algorithm, optimizes the network architecture and the weight vector, solves the short genome team actual overlapping, variation opportunity excessively small problem in computation process, enable the descendant population to have a better multiplicity. In addition, the combination of genetic algorithm with progeny generated by Solis& Wets operation enriched the heredity search space, sped up the convergence rate. Simulation and experimental results indicate that the HGANN algorithm proposed in this article effectively sped up the genetic algorithm convergence rate and enhanced VVPCMP´s position posture precision.