This chapter addresses filter algorithms for the diagnosis and prognosis of failure in analog circuits, using Unscented Kalman Filters (UKFs) and Unscented Particle Filters (UPFs). Most of the industrial process uses sensors and actuators to the increase the production without conceding the quality of the product. The sensors and actuators (normally an electric motor) are connected with a power conditioning circuit which consists of several active and passive devices. The sensor is a highly sensitive element and has primary contact with the industrial process. The sensed signals are considered to decide controllers/processors to operate the motors through the power converters. These elements are in the manufacturing/control process round the clock. The failures in anyone the component may be on sensor networks or power converter leads to process failure. To avoid failures in the industrial process and also to ensure the safe operation of associated components, filtering algorithms are proposed. Particularly, small-signal sensor circuits and large-signal power circuits are monitored through filtering algorithms to forecast the failure of individual components in a circuit and also to forecast the remaining useful life (RUL) of that component. This will help the users to replace the circuit to regulate the manufacturing process.