Biomass gasification process is a complex non-linear thermo-chemical conversion process, the operation parameters of which directly affect the operation of the gasification results, thus the parameters is particularly important to determine. On the basis of consulting a large number of domestic and foreign-related literature, grasping fully the mechanism of biomass gasification process. In this paper, based on Gaussian radial function, and sigmoid kernel function to establish a wood gasification process least squares support vector machine model, forming the relationship between gasification process control parameters and the components of gasification gas. Fitting a multi-objective optimization function between gasification process control parameters (gasification temperature, catalyst quantity) and the main gasification performance indicators (gas production rate, gas heating value and the gasification efficiency), at the same time, the paper optimized the objective function by using the method of particle swarm optimization and genetic algorithms. The optimization target value of control parameters are obtained when the three main performance indicators of wood gasification process achieve synthetically satisfactory results.
|