Accurate stator current measurement is mandatory to maximize electric drive precision and safety. During motor operation, current sensors may be affected by various phenomena, leading to amplification errors (G) or offsets (OFF), which gradually accumulate over time. Detecting and correcting these deviations at an early stage is crucial to enhance overall system performance. This paper presents a simple fault-tolerant system capable of detection, classification, and online correction of measurement errors and complete signal losses. The proposed approach relies on the analysis of the amplitudes and mean values of both measured and estimated currents. A dedicated algorithm is used to mitigate the impact of variations in motor parameters on the quality of the estimate. The results of simulation tests performed under various sensor conditions and motor operating points demonstrate the effectiveness of the proposed method, confirming its potential for practical implementation in real-world applications.
The dataset contains the results of simulation studies of a system for detection, classification, and compensation of stepwise and progressive current sensor faults, dedicated to an induction motor drive under vector control. The system is based on analytical relationships, does not require artificial intelligence techniques, nor additional diagnostic signals. Sensitivity to changes in motor parameters is reduced by implementing an advanced diagnostic algorithm.
The simulation studies were carried out in MATLAB/Simulink 2021b. The system considered has two current sensors (on phases A and B), each of which may be subject to the following faults:
OC – complete signal loss; G – gain error; OFF – DC offset.
Simulation conditions:
Faults of type G and OFF were applied gradually, with a growth rate of 0.02 [p.u./s], until reaching an error of 0.1 p.u. Motor parameter changes were realized by changing the rotor resistance: an inertial increase to 120% of its nominal value was assumed.
File Description:
1) "steady_state_w100_m100.xlsx" contains measurements for operation at 100% speed and load.
2) "dynamic_operation.xlsx" contains measurements for operation with variable speed and load (between 100% and 20% of nominal values).
In the "Metadata" sheet, file metadata is included.
In the "Data" sheet, the signals are provided (all expressed in per-unit, p.u.):
deltaGA → corrective factor for Gain-fault in phase A;
deltaGB → corrective factor for Gain-fault in phase B;
deltaOFFA → corrective factor for OFFset-fault in phase A;
deltaOFFB → corrective factor for OFFset-fault in phase B;
deltaPCA → corrective factor for ParameterChange for phase A;
deltaPCB → corrective factor for ParameterChange for phase B;
lambdaOCA → fault identifier for OpenCircuit-fault in phase A;
lambdaOCB → fault identifier for OpenCircuit-fault in phase B;
iAest → estimated current in phase A;
iBest → estimated current in phase B;
iAim → real current in phase A;
iBim → real current in phase B;
iAmea → measured current in phase A;
iBmea → measured current in phase B;
tref → reference load torque;
temIM → estimated electromagnetic torque;
wref → reference speeed;
wmCOR → measured speed with FTC applied;
wm_NoCor → measured speed without fault correction.
(2026-02-11)
