Hybrid improved genetic algorithm for the estimation of biodegradation kinetics parameter

A hybrid estimation method is presented to estimate the kinetics parameter of biodegradation, which includes that a properly Improved Genetic Algorithm (IGA) automatically generates initial parameter estimations for the local optimization method - Gauss-Newton method. The physiological and biochemical factors adjusting of the microorganisms to adapt various circumstances result in the large variation of the kinetics parameters involved in the biodegradation models, which brings the difficulty for the local optimization method that depends on suitable initial estimation. The IGA with excellent searching accuracy can just be employed to resolve this problem. Synthetic data and published data based on the integrated Contois Model are used to verify proposed hybridization method. The independent variable residual sum of the squared errors (IRSSE) gives the criterion to minimize. The results demonstrate the effectiveness of the algorithm. As the IGA does not rely on the mathematical form of objective biodegradation kinetics models, the hybrid method is convenient intelligent and promising.