Author(s):
Prof. dr. sc. Nedjeljko Perić -
Department of Control and Computer Engineering, Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia
Prof. dr. sc. Željko Ban -
Department of Control and Computer Engineering, Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia
As the importance of renewable energy sources rises, the importance of making them efficient and readily available also rises. As part of the global effort to develop and use renewable energy sources, we established the Laboratory for Renewable Energy Sources (LARES) in the Faculty of Electrical Engineering and Computing at the University of Zagreb. LARES conducts research specifically related to controlling wind and solar energy sources, as well as various forms of energy storage.
LARES Overview
We designed LARES as a microgrid consisting of a custom made wind turbine, an electrolyser for hydrogen production, a hydrogen fuel cell stack, and photovoltaic panels (see Figure 1). The purpose of the laboratory is to investigate and develop microgrid control algorithms, as well as to design algorithms to control specific energy sources. The laboratory researchers study ways to increase energy conversion efficiency for renewable energy sources using advanced control and estimation techniques.
It’s challenging to develop these control algorithms because renewable energy sources often display nonlinear, stochastic behavior; the development requires many control loops; and it requires real-time hardware operation on a millisecond time scale. The nature of the renewable energy system requires complex control algorithms in addition to classical algorithms. We needed hardware that was fast enough to execute these complex algorithms in real time and a simplified way to program them.
To fulfill these requirements, we considered several hardware platforms, including programmable logic controllers (PLCs), industrial PCs, dSpace hardware that can be programmed with The MathWorks, Inc. Simulink® or MATLAB® software, and National Instruments hardware programmed with LabVIEW software. We based our requirements on computational speed, number and diversity of input and outputs, programming simplicity, and cost-effectiveness ratio. We chose National Instruments CompactRIO and PXI hardware.
Wind Turbine Energy Source
The LARES wind energy source consists of a wind turbine placed in an air chamber with artificial wind produced by a powerful blower (see Figure 2).
The electrical energy produced by a generator-driven wind turbine is conditioned by a controlled AC/AC converter.
The control system hardware is located in a control room and consists of a PC powered by NI LabVIEW and an NI PXI-1033 chassis. An NI cRIO-9014 with an NI 9219 universal analog input module and an S.E.A. WLAN communication module are housed in the wind turbine rotor hub. Because we can program the amount of wind the blower produces and control the individual pitch of each turbine blade, we can investigate problems related to nonhomogenous wind speed and tower vibrations. The equipment in the wind portion of the laboratory is dedicated to investigating the behavior of the system in nearly stochastic wind conditions, as well as developing a control algorithm for the system.
Hydrogen Plant
The hydrogen part of the laboratory (see Figure 3) consists of the electrolyser for hydrogen production, metal hydride hydrogen storage containers, a fuel cell stack, voltage converters, and a cooling system. The ON/OFF valves establish the paths of the hydrogen and oxygen flow. The control valves in the fuel cell system obtain the desired hydrogen pressure and hydrogen/oxygen flow ratio and regulate the stack temperature. The fuel cell stack output voltage is conditioned by a DC/DC boost converter.
The control system of the hydrogen-based energy source in LARES is built on a PC program powered by LabVIEW and an NI cRIO-9024 controller with field-programmable gate array (FPGA) circuits and I/O modules in an NI cRIO-9118 reconfigurable chassis for real-time hardware control. The I/O subsystem uses an NI 9425 digital input module, an NI 9476 digital output module, a 24-bit NI 9219 universal analog input module, an NI 9206 fast analog input module, and an NI 9265 fast analog output module. Figure 4 shows the cRIO-9024 controller and the I/O modules.
We developed the control software using the LabVIEW graphical development environment. The control software performs sequential and feedback control of the hydrogen plant, controls the voltage converters, and controls the safety and monitoring systems. The reconfigurable FPGA circuits in the NI 9118 chassis and fast analog I/O are especially useful in boost converter control because all responses must be fast.
Photovoltaic System
The photovoltaic system is planed as the third plant in the laboratory microgrid. It will be based on photovoltaic panels placed on the building roof. The electrical energy from the photovoltaic panels will be transfered to the laboratory microgrid by controllable voltage converters.
Laboratory Functions
We built the laboratory to design control algorithms for specific laboratory energy sources, which includes advanced control algorithms such as adaptive control and model predictive control.
In addition to controlling the specific energy sources in the laboratory, we will develop algorithms for microgrid control. The algorithm will control the energy flow in the microgrid to obtain optimal energy production and storage by monitoring the availability of each specific energy source.
The microgrid based on hydrogen, wind, and sun power plants supports experimental research of the control paradigms needed for optimal synergy with other microgrid parts to reach specific objectives on a microgrid level (for example: maximum efficiency, maximum component lifetime, and maximum profit). Our research could be a good foundation for virtual power electric plant control.
Conclusion
We chose National Instruments software and hardware to create the LARES control systems because of its unique qualities of reliability, availability, and robustness. In addition, the modularity of the equipment gives us the option of future system expansion. The processor speed augmented with FPGA circuits makes complex control algorithm execution possible in real time. And last, but not least, the graphical programming interface simplified programming complex control algorithms and gave us fast prototyping.
MATLAB® and Simulink® are registered trademarks of The MathWorks, Inc.
Author Information:
Prof. dr. sc. Nedjeljko Perić
Department of Control and Computer Engineering, Faculty of Electrical Engineering and Computing, University of Zagreb, Croatia
Zagreb
Croatia
Print
PDF
