Open-circuit or short-circuit faults, as well as faults in discrete parameters are the most used models in the simulation method before testing. As the response of an analog circuit to an input signal is continuous, failures in any specific circuit element may not characterize all possible component failures. There are three important features in diagnosing analog circuit faults: faulty component identification, faulty element value determination, and circuit tolerance restrictions. To solve this problem, a fault diagnosis method is proposed in this work using a Particle Swarm Optimization (PSO), where the nonlinear equations of the circuit under test are used to calculate the circuit parameters. The objective is to identify which circuit component has the potential to present the failure by comparing the responses obtained in the real circuit and the response obtained by the optimization process. The Tow-Thomas Biquad Filter was used to evaluate the proposed implementation. The proposed methodology was able to identify the defective components in 5 out 8 cases with 100% accuracy. However, for the remaining 3 cases, a lower accuracy rate of 75% was possible.