Low intrusive efficiency estimative of DC motors

This work presents a low intrusive system for identification of the parameters of a DC motor model to estimate an electrical machine efficiency during operation. A metallic base supports an electrical generator coupled to the motor. The generator output is connected to replaceable electrical resistances in order to control the mechanical load on the shaft. Also, a control system is implemented in a dedicated microprocessor in order to apply different excitation signals to the motor while the current and angular velocity signals are acquired by an ADC. The system communicates with a computer through an USB channel. The dynamic equations that describe the DC motor are discretized and rearranged in order to apply the iterative least squares algorithm, running in the computer. Finally, the electrical and mechanical parameters of the model: winding inductance L, resistance R, mechanical moment of inertia J, friction coefficient b and general torque constant K are estimated. Thus the system estimates the motor losses from parameters R and b and approximates the motor efficiency. The results are validated through the indirect measurement of the mechanical power: a load cell measures the reaction force produced on the generator and a tachogenerator measures the angular velocity. The estimated output power as well the estimated efficiency follow the measured mechanical power on the shaft and the results have shown that the system can be used as an alternative non intrusive method.