In this work, a fuzzy logic-based system for diagnosis and control of an anaerobic wastewater treatment plant was developed. Three variables (methane production, hydrogen concentration in the gas phase, and intermediate alkalinity/total alkalinity ratio) were selected, according to previous works, as the more adequate for organic overload identification. The diagnosis and control system was developed using expert's knowledge related with normal or overload process states for establishing a set of rules considering the selected variables. The control system computes the feed flow rate for adjusting the organic load applied to the reactor. The control law is based on a response surface obtained according to fuzzy inference Mamdani methodology. The diagnosis and control system was validated at three different levels: (a) simulation with ADM1 (wastewater containing ethanol); (b) open-loop methodology using experimental data from a hybrid upflow sludge bed-filter (USBF) anaerobic pilot plant treating a wastewater containing ethanol; and (c) closed-loop control of a USBF pilot plant treating wastewater containing carbohydrates. The proposed system diagnoses correctly the different episodes, and it demonstrated a high reliability, supplying an adequate control action to achieve the desired set point as well as to manage a sudden change in influent COD. Moreover, the performance of the developed system can depend slightly of the noise in the measured variables used for determining the process state.