Application of evolutionary optimisation to space plasma turbulence modelling

An evolutionary optimisation technique is presented to solve a problem for turbulent space plasma system modelling, using multi-satellite magnetic field measurements of the plasma turbulence.The application of evolutionary algorithms for system identification allows model structure selection and fitting of parameters for the chosen model using measured inputs and outputs of the system, which can then be used to determine physical characteristics of the system. Genetic algorithms are one such technique that has been implemented. Experimental studies have been performed using multi-point satellite observations providing input and output measurements of the turbulent plasma system. Linear and nonlinear models of the turbulent plasma system are identified and results using genetic algorithms are compared to results obtained from the least squares estimation method.