Fault-tolerant control algorithm of the manned submarine with multi-thruster based on quantum-behaved particle swarm optimisation

Abstract

To cite this article: Daqi Zhu , Qian Liu & Zhen Hu (2011) Fault-tolerant control algorithm of the manned submarine with multi-thruster based on quantum-behaved particle swarm optimisationA thruster reconfiguration control approach of manned submarine with 7000 m operation depth based on quantum-behaved particle swarm optimisation (QPSO) is presented in this article. The manned submarine has eight thrusters. When thruster faults happen, the corresponding weight matrix is updated to restrict the usage of the faulty thruster. But the solution with this method may become unfeasible (exceed the rated valve of the thruster) and cannot be directly applied to the thrusters. In order to complete an appropriate control law reconfiguration, a novel control reallocation method based on QPSO is proposed. Not only is the solution obtained by the approach of QPSO limited in the whole feasible space, but also the control error of the fault-tolerant control is very small. Eight dimensions of the particles are used in this article, and each particle represents the components of the control vector, it searches in the range of the restricted factor value to make sure that all the reconfiguration control solutions are feasible. Compared to the weighted pseudo-inverse method, the error of the obtained control vector with the QPSO is very small. Finally, simulation examples of multi-uncertain faults are given to illustrate the advantages of the proposed method.

Gradient-Based Parameter Identification Algorithms for Observer Canonical State Space Systems Using State Estimates

“…The simulation results indicate that the proposed algorithms can deal with the identification problem of systems with unknown state variables. The methods in this paper can be extended to study identification algorithms for other linear and nonlinear systems with colored noises [16][17][18] and linear-in-parameters systems [37,38] and applied to other fields [43][44][45].…”

Fault detection and accommodation of saturated actuators for underactuated surface vessels in the presence of nonlinear uncertainties

“…The existing works [4][5][6][7][8][9][10][11] for the FDA problem of ocean vehicles only considered fully or over-actuated systems. In these results, additional actuators were required to achieve given missions successfully in the presence of the actuator faults.…”

“…Pseudo-inverse methods were derived for the thruster force allocation [6,7]. A quantum particle swarm optimization method was proposed to reallocate the fault thrusters precisely [8]. Other control allocation results have been successfully applied to unmanned ocean vehicles (see [9] and the reference therein).…”

A Novel Tracking Controller for Autonomous Underwater Vehicles with Thruster Fault Accommodation

“…The GI approach cannot handle the constrained control problem since the solution of this approach is only based on the attainable command set. To handle such cases where attainable control inputs cannot be 594 allocated, T-approximation (Truncation) and S-approximation (Scaling) methods were proposed (Zhu et al, 2011). The solution obtained by the above approximation methods can be contained inside the entire attainable command set, but the magnitude and direction errors caused by approximation still exist and the AUV cannot follow the desired trajectory completely.…”

“…Zhu et al (2011) and Liu et al (2009) use the Quantum-behaved Particle Swarm Optimisation (QPSO) method to deal with the fault tolerant control of the 7000m manned submarine and URIS AUV model respectively while the fitness value is based on direction error and magnitude error. Inspired by the above design and in order to do a precise reallocation of the fault thrusters, the QPSO method is applied to the control reallocation problem and a thruster Fault Accommodation (FA) method is proposed for the Odin AUV.…”