Alireza Payman

Energy Management in a Fuel Cell/Supercapacitor Multisource/Multiload Electrical Hybrid System

In this paper, a flatness-based nonlinear control method is proposed to control a multisource/multiload electrical hybrid system (EHS). The EHS is composed of a fuel cell and a supercapacitor-bank (SCB) as the main and auxiliary sources. They supply two independent loads, connected to a dc-bus through unidirectional buck converters. The proposed method is able to control the fuel cell output power and its dynamics. It also allows limiting the current of an SCB during charging and discharging processes. The proposed control strategy has the advantage of not requiring any commutation between different control algorithms when the operating mode of the system changes (from the normal mode to the overload mode, for example). As the fuel cell output characteristic (FCOC) varies with the physical and environmental parameters, an observer is also proposed and designed to estimate either the fuel cell voltage-power (V-P) output characteristic or voltage-current (V -I) output. The use of the proposed observer allows achieving an efficient control of the system and avoiding overcharging or discharging of an SCB. Experimental results demonstrate the operation of the proposed control strategy and observer in all situations. Experimental test on a system is done with two separated loads of 5 and 12 V, via a dSPACE platform.

An Adapted Control Strategy to Minimize DC-Bus Capacitors of a Parallel Fuel Cell/Ultracapacitor Hybrid System

In this paper, a flatness-based control method is used to control the dc/dc converters of an electrical hybrid system. This system is composed of an ultracapacitor, which is connected in parallel to a fuel cell through a bidirectional converter. This association supplies a load through another dc/dc converter. To control these converters, the mathematical model of the studied system is first presented, and then, it is proven that the system is flat. Considering the electrostatic energy stored in the dc-bus capacitors as the system output, the state variables and control variables are extracted as functions of the system output and its derivative. The system is controlled by planning the desired reference trajectories on the flat output components, and forcing them to follow their own references. The fuel-cell-dynamics control is also studied to observe the criterion of (di/dt)max. Based on the used control strategy, a method is developed to calculate the minimum values of the dc-bus capacitors in the proposed parallel hybrid system. The simulation and implementation results are presented to validate operation of the proposed method in the hybrid system.

Flatness based control of a non-ideal DC/DC boost converter

This paper presents a flatness based method to control a non-ideal DC/DC boost converter. An online trajectory planning algorithm is used to control the total energy stored in the system. The trajectory is planed thanks to polynomial form which is valid on a finite horizon. This method is applied at first to an ideal DC/DC boost converter. Then, an online estimator is used to track the evolution of the converter losses and the load variations and therefore, a more realistic model is developed for a non-ideal converter. The applied control strategy is finally adapted to the completed converter model. Therefore, a high dynamic nonlinear control is deduced for a non-ideal DC/DC boost converter. The experimental results prove validity of the control method and the system losses estimator.

Implementation of a Flatness Based Control for a Fuel Cell-Ultracapacitor Hybrid System

A flatness based control of a hybrid system composed of a fuel- cell and an ultracapacitor is investigated and implemented in this paper. For this purpose, at first, the system is provided to be flat and an expression which takes into account the power losses in converters is extracted to control variables. Three control methods, sliding mode, state feedback and classical PI controllers are considered to ensure the control of the flat outputs to their references. The dSPACE-based implementation results are presented to validate operation of each control method in the purposed hybrid system.

Performance Investigation and Comparison of Two Different Electrical Hybrid System Structures

Two different topologies of an electrical hybrid system composed of a fuel cell as main source and an ultracapacitor as auxiliary source are investigated in this paper. To control these hybrid systems, a flatness-based control method which eliminates algorithm commutation in different operating modes is employed and its operation is studied in each structure. After simulating model of the structures, system efficiency in steady states is calculated while all ohmic, switching and conduction losses are considered. The simulation results show well performance of the proposed control method in both topologies and allow comparing operation of the systems in different situations.

Fuel cell characteristic observation to control an electrical multi-source/multi-load hybrid system

This paper deals with flatness based control of a multi-source/multi-load electrical hybrid system (EHS) composed of a fuel cell and a supercapacitor (SC) as the main and auxiliary source, respectively. The proposed method does not contain any commutation algorithm whereas the different operating modes (normal mode and overload mode) exist in the system. This DC-power source supplies two independent 5 V and 12 V loads which are connected to a DC link through the buck converters. An observer system based on monitoring the supercapacitor voltage is also designed and proposed to estimate the output V-I characteristic of fuel cell which varies with some physical and environmental parameters like temperature, humidity, etc. Experimental results prove validity of the proposed approach.

Control of a low power electrical hybrid system supplied by a fuel cell and a supercapacitor pack

In this paper a combination of the flatness based method and the Input/Output (I/O) linearization technique is presented to control a low power electrical hybrid system supplied by a fuel cell (FC) and a supercapacitor pack (SCP). This hybrid system feeds three independent loads through the DC/DC buck converters. The first method is used to manage the energy between FC, SCP and loads, while the second one is employed to control the buck converters. This leads to benefit advantages of the two methods: prediction of the system behavior even in the transient state and decoupling the control variables of the buck converters. The simulation results prove efficiency of the control strategy of the multi sources/multi loads electrical hybrid system.