Temperature sensor of circuit breaker

as an element to real-time monitor temperature of key area of electrical equipment

is an important part of intelligent circuit breaker. Since the temperature measured by the temperature sensor is related to the magnitude of primary current

power CT is used in this paper for measurement of primary current in order to ensure the accuracy of real-time monitoring. However

both operating temperature and magnetic saturation effect of coil have significant influence of the secondary current. As the temperature has significant influence on the secondary current

the secondary current waveform is firstly collected (with 10 kHz sampling frequency) under different temperatures and different primary currents. The morphological filtering method is used for denoising raw signals

the peak of fitted sinusoidal waveform is obtained by way of extracting signal fragment to perform sinusoidal function fitting based on trust-region reflection non-linear least squares method. The normal temperature data-peak - current relationship curve collected in the temperature chamber is taken as the calibration curve. Primary current can be calculated by matching the fitted waveform’s peak and calibration curve. The maximum deviation between the calculated value and real current value is 3.86% under such environment

which

to the maximum extent possible

correct the influence of the temperature and magnetic saturation to the secondary current.