Industrial processes generate large amounts of heat, which often cannot be utilised immediately. A key reason for this is that supply and demand for heat often do not coincide in terms of timing. Thermal energy storage systems can bridge this gap by absorbing surplus heat and making it available when needed at a later time. The recovery and storage of industrial waste heat therefore offers great potential for increasing energy efficiency and reducing greenhouse gas emissions. So-called packed-bed thermal energy storage (PBTES) systems are of particular interest. These work, for example, by allowing hot air to flow through a storage material consisting of solid particles, thereby transferring the heat directly to the material. During discharge, this process is reversed.
Researchers at TU Wien have now developed a new method that allows the state of such a thermal energy storage system to be precisely determined during operation. The work, which was recently published in the journal Applied Thermal Engineering, combines several fields of research: the modelling of distributed physical systems, mathematical model reduction, optimal sensor placement, control-theoretic state observation and online parameter identification.
Knowing what is happening inside
A key challenge in operating a packed-bed thermal storage system is knowing its actual internal state. A relatively narrow transition zone, known as the thermocline, forms between the hot and cold regions. Its position, shape and stability have a direct influence on how efficiently the storage system operates and how much of the stored energy can actually be utilised.
"A complete measurement of the spatial temperature distribution would require a large number of sensors inside the storage system. Particularly given the high temperatures and harsh conditions found in industrial plants, this is technically complex and expensive," says Stefan Jakubek, explaining the problem.
This is precisely where the new method comes in: the researchers have developed an estimation method which, based on a mathematical model of the process and a handful of strategically placed temperature sensors, can infer the entire spatial temperature distribution within the storage tank. "The mathematical model thus, in a sense, supplements the information that cannot be measured directly by sensors," explains René Hofmann.
Complex physics for ongoing operations
The researchers are using a mathematical method to simplify the representation of the complex processes within the heat storage tank. This enables them to calculate more quickly how the temperature boundary between hot and cold material within the tank shifts. In doing so, the properties of the original model that are crucial to this movement are retained. They are also investigating at which positions temperature sensors provide a particularly wealth of information about the overall state of the storage unit.
The new method not only reconstructs the temperature distribution, but also estimates the heat transfer between the gas flowing through it and the storage material. "This parameter enables us to describe how effectively heat is transferred between gas and solid," says Martin Kozek. "This opens up another interesting avenue of research." Changes in heat transfer can thus provide indications of changes within the storage unit. In industrial applications, for example, dust-laden gas flows can lead to deposits over the long term, thereby impairing heat transfer. Continuous parameter estimation could therefore not only support energy-efficient operation in future, but also provide information on the condition of the plant and enable condition-based maintenance.
When control engineering and energy engineering come together
The work also demonstrates the added value of collaboration between different disciplines: the Institute of Energy Technology and Thermodynamics at TU Wien contributes expertise in thermal energy systems and heat storage, whilst the Institute of Mechanics and Mechatronics at TU Wien complements this with methods from mathematical modelling, control engineering and process automation. Both institutes are conducting joint research as part of the Christian Doppler Laboratory for Digital Twins of Distributed Parameter Systems .
