Linear-in-Parameter Models Based on Parsimonious Genetic Programming Algorithm and Its Application to Aero-Engine Start Modeling

Ying-hong LI, Xun-kai WEI
2006 Chinese Journal of Aeronautics  
A novel Parsimonious Genetic Programming (PGP) algorithm together with a novel aero-engine optimum data-driven dynamic start process model based on PGP is proposed. In application of this method, first, the traditional Genetic Programming(GP) is used to generate the nonlinear input-output models that are represented in a binary tree structure; then, the Orthogonal Least Squares algorithm (OLS) is used to estimate the contribution of the branches of the tree (refer to basic function term that
more » ... not be decomposed anymore according to special rule) to the accuracy of the model, which contributes to eliminate complex redundant subtrees and enhance GP's convergence speed; and finally, a simple, reliable and exact linear-in-parameter nonlinear model via GP evolution is obtained. The real aero-engine start process test data simulation and the comparisons with Support Vector Machines (SVM) validate that the proposed method can generate more applicable, interpretable models and achieve comparable, even superior results to SVM. Key words aerospace propulsion system linear-in-parameter nonlinear model Parsimonious Genetic Programming (PGP); aero-engine dynamic start model 1 During use and operation of aero-engine, the start process is an important phase, which is a key precondition before going into normal working status. The start process is the foundation of start process control design and simulation. But due to its extreme complexity and lacking of low speed part characteristics, the traditional thermal dynamic model is hard for practical use. Data-driven modeling technique only requires observed input-output data regardless of complex physical or chemical process to obtain a commendable model representing the given complex process. This virtue makes it popular and effective for nonlinear dynamic process modeling. Recently, there are some research achievements of nonlinear modeling available for complex nonlinear plant ·296· LI Ying-hong, WEI Xun-kai CJA
doi:10.1016/s1000-9361(11)60331-2 fatcat:b6c33hcqyfa2tktzb3cbtynkcm