Abstract-This paper concerns the optimization of an accumulator used in an industrial elastic web processing plant (paper, fabric, polymer, metal …). A nonlinear model of an industrial accumulator is first detailed which enables to deduce a linear model. These models are derived from the physical laws describing web tension and velocity dynamics in each web span. The effects of time-varying rheological and mechanical parameters, such as web Young modulus, web length and rolls inertia, on accumulator dynamics and performances are analyzed. The second part presents several optimizations of industrial PI controllers using evolutionary algorithm for a realistic non-linear model, in conjunction with the controllers interpolation strategy. Finally, simulations made in the Matlab/Simulink software environment show performances improvements compared to hand tuned controllers.

Industrial systems usually contain uncertain parameters. These uncertainties come from lack of knowledge of physical phenomena or from evaluation difficulties. The stochastic behaviour of these parameters have to be taken into account during the design process of industrial systems. To tackle with these uncertainties, the Design for Multi-objective Six Sigma method is used. This method permits to fix six sigma constraint and it is applied on the synthesis of web tension controller for an industrial large-scale roll-to-roll system. The key idea is to synthesize the controller in frequency domain by Hinfinity synthesis. Then the six sigma constraint is fixed on output web tension in time domain. The robust controller synthesis is then compared with standard Hinfinity synthesis approach. This paper presents for the first time a controller synthesis including six sigma constraints.