This paper presents an intelligent fuzzy method for maximum power point tracking (MPPT) of a single grid-connected photovoltaic system. The fuzzy inference system can easily synthesize the algorithm to perform MPPT. The MPPT fuzzy method is improved by weighting the control action by the short-circuit current. Hence, it is proposed an estimator of the short-circuit current based on a Takagi-Sugeno fuzzy model of the photovoltaic generator. The short-circuit current can be estimated under any weather condition in real time, without disconnecting the photovoltaic generator and measuring it. Simulation results show attractive features such as, accuracy, fast response, good dynamic performance and efficiency in the conversion process.