Plants in chemical and biochemical industries are becoming larger and more complex. The growing safety and environmental demands are forcing industry to look for new and more powerful fault detection techniques. This paper presents a method for detecting faults that can appear in some parts of a chemical plant. This method is based on the principle of the generalized likelihood ratio (GLR) and is intended to reveal any drift from the normal behaviour of the process. It solves partially the diagnosis of the fault and allows localizing its physical origin. The work done in this paper is an application of the GLR test to a nonlinear system such as stirred reactors in the presence of exothermic chemical reactions. The abnormal behaviour of a chemical reactor due to two different faults in its control parameters is examined. The aim of this experimental study is to detect the presence of such faults and to pinpoint the moment when each occurs. The evolution of the detection criterion as well as the determination of the delay in detection are observed. The chosen reaction is a very exothermic oxido-reduction one; the oxidation of sodium thiosulfate by hydrogen peroxide.