Search

Maintaining a fuel cell (FC) system in correct operating conditions when subjected to fast load changes requires good system control. The best results are typically achieved using model-based control strategies. This paper elaborates a control-oriented proton exchange membrane FC system model in two distinct steps. First, the FC auxiliary systems (i.e., air and hydrogen supply along with thermal management) are described in a state-space approach linking user input to FC operating conditions. A steady-state FC stack model then describes the effect of those operating conditions on system performance. The model then allows the implementation of advanced control strategies such as multivariable control with decoupling, actuator sensitivity analysis or maximum efficiency tracking algorithms. Index Terms--Fuel cell (FC), modeling, state-space control.

In symmetric multilevel inverters, there is a tradeoff between the output quality and the reliability and efficiency of the converter. New asymmetric and hybrid solutions, using different voltages and devices in various parts of the inverter, promise significant improvements for medium-voltage industrial drives. This paper investigates such a hybrid asymmetric nine-level inverter. It consists of a three-phase three-level integrated gate-commutated thyristor inverter (main inverter), with a two-level insulated-gate bipolar transistor H-bridge (subinverter) in series with each phase. To keep the power part simple and the efficiency high, the subinverters have no feeding from the net and can only supply reactive power. This is a very interesting solution in terms of power quality, efficiency, reliability, and cost. But the nonsupplied intermediate-circuit capacitors form an unstable system. This paper proposes a control method to stabilize their voltages. Power balancing is guaranteed by varying the common-mode voltage, using an online nonlinear model-predictive controller. The controller predicts the system evolution as a function of the control inputs. A cost function of system and control quantities is iteratively minimized in real time, to find the optimal control to apply to the system. Simulations and measurements demonstrate stable behavior in steady state and during transients. Precharging of the nonsupplied capacitors is also an issue to consider. This paper proposes a startup method that charges them in parallel with the supplied ones, without any additional equipment. Measurements show its successful application in the proposed drive system. Index Terms--Converters, motor drives, multilevel systems, optimal control, power electronics, predictive control.

In recent years, power variations and energy criteria have been the main motivations for developing regenerative drive converters for elevators [1]. A better performing solution for power smoothing can be easily found by using a supercapacitor-based storage device, connected to the intermediary circuit of a variable-speed-drive system. In this paper, power and energy considerations are being accounted for in the design of the storage tank and regarding the maximum power demand from the feeding network. For the power-conversion circuit, which is necessary to compensate the voltage variations of the supercapacitors during charge and discharge, a high-efficiency converter topology is proposed, which allows the bidirectionnal energy flow under soft-commutation conditions. Additionally it offers a good flexibility for the optimal sizing of the supercapacitor voltage level. The typical behavior of the special converter is given, together with an analysis of the advantages related to the specific application. Index Terms--Energy storage, load leveling, medium-frequency transformers, supercapacitors.