On the analytic derivation of Poincaré maps — The forced Brusselator problem