Current sensors fault detection and tolerant control strategy for three-phase induction motor drives

Abstract

In the case of failure of one or more components of a drive system, the emergency shutdown of the system is not always the best way to act. Therefore, simultaneous reconfiguration of the drive control strategy is mandatory to enable an uninterrupted operation to cater for the catastrophic failure. In this context, this paper presents a current sensors fault-tolerant control method for induction motor drives, based on vector control and currents estimation. Several important issues are considered in the proposed method, namely, the detection of sensors failure, isolation of the faulty sensors, and reconfiguration of the control system by proper currents estimation. A new adaptation of the Luenberger observer is proposed and used to perform the task of stator currents estimation. Furthermore, a developed logic circuit is used to detect the faulty current sensors and isolate them with simultaneous generation of logic impulses allowing switching to a proper estimation. The proposed fault-tolerant control strategy is firstly tested in MATLAB/Simulink environment in order to illustrate its high-performance. Then, several experimental tests are carried out on a 1.1 kW three-phase induction motor to validate the theoretical results and to confirm the effectiveness of the proposed algorithm.

Appendices

Appendix A

See Tables 2 and 3.

Table 2 Specifications and parameters of the induction motor used in simulation and experimental tests

Table 3 PI regulator values used in simulation and experimental tests

Appendix B

MAX	$\left| {\Delta I} \right|$ at TR	$\left| {\Delta I} \right|$ at CR	$\left| {\Delta I} \right|$
Threshold determination
$T_{\text{L}} = 0\,{\text{N}}\;{\text{m}}$; $\omega_{r} = 105 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.315	0.265	$\left| {\Delta I_{a} } \right| < 0.35$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.319	0.271	$\left| {\Delta I_{b} } \right| < 0.35$
$\left| {\Delta I_{c} } \right|$	0.323	0.269	$\left| {\Delta I_{c} } \right| < 0.35$
$T_{\text{L}} = 2\,{\text{N}}\,{\text{m}}$; $\omega_{r} = 105 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.306	0.249	$\left| {\Delta I_{a} } \right| < 0.35$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.348	0.257	$\left| {\Delta I_{b} } \right| < 0.35$
$\left| {\Delta I_{c} } \right|$	0.361	0.284	$\left| {\Delta I_{c} } \right| < 0.40$

MAX	$\left| {\Delta I} \right|$ en TR	$\left| {\Delta I} \right|$ at CR	$\left| {\Delta I} \right|$
$T_{\text{L}} = 5 \,{\text{N}}\;{\text{m}}$; $\omega_{r} = 105 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.326	0.214	$\left| {\Delta I_{a} } \right| < 0.35$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.312	0.234	$\left| {\Delta I_{b} } \right| < 0.35$
$\left| {\Delta I_{c} } \right|$	0.334	0.229	$\left| {\Delta I_{c} } \right| < 0.35$
$T_{\text{L}} = 0\, {\text{N}}\,{\text{m}}$; $\omega_{r} = 21 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.348	0.265	$\left| {\Delta I_{a} } \right| < 0.35$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.315	0.2612	$\left| {\Delta I_{b} } \right| < 0.35$
$\left| {\Delta I_{c} } \right|$	0.387	0.239	$\left| {\Delta I_{c} } \right| < 0.35$
$T_{\text{L}} = 2\, {\text{N}}\;{\text{m}}$; $\omega_{r} = 21 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.379	0.281	$\left| {\Delta I_{a} } \right| < 0.40$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.366	0.272	$\left| {\Delta I_{b} } \right| < 0.40$
$\left| {\Delta I_{c} } \right|$	0.328	0.233	$\left| {\Delta I_{c} } \right| < 0.35$
$T_{\text{L}} = 5 \,{\text{N}}\;{\text{m}}$;  $\omega_{r} = 21 \,{\text{rad/s}}$
$\left| {\Delta I_{a} } \right|$	0.364	0.251	$\left| {\Delta I_{a} } \right| < 0.40$	${\text{Th}} = 0.45$
$\left| {\Delta I_{b} } \right|$	0.337	0.278	$\left| {\Delta I_{b} } \right| < 0.35$
$\left| {\Delta I_{c} } \right|$	0.346	0.250	$\left| {\Delta I_{c} } \right| < 0.35$

1. TR transient regime, CR continuous regime

Appendix C

Comparison between the suggested method and relevant methods over the last 10 years

	[26]	[28]	[27]	[51]	[30, 31]	[37, 38, 40]	Suggested method
Failure succession in the different sensors	Unsuitable to detect	Unsuitable to detect	Suitable to detect	Suitable to detect	Suitable to detect	Suitable to detect	Suitable to detect
Number of current estimators	3	3	3	0	0	1	1
Number of used current sensors	2	1	1	0	0	0	0
Number of controllers	1	1	1	4	3	1	1
Considered sensor faults	Complete outage	Complete outage and gain fault	Complete outage and gain fault	Complete outage and gain fault	Complete outage and gain fault	Complete outage and gain fault	Complete outage, gain fault, and offset
Sensitive to parametric variation	Sensitive	Sensitive	Sensitive	Sensitive	Sensitive	Sensitive	Insensitive except in high temperature operation