State of Charge estimation of a Metal Hydride tank to state Hydrogen through Neural Network based modeling
This master’s thesis has been carried out during the second quarter of the 2023-2024 school year, autonomously. The ever-increasing need for non-contaminating and perpetual energy sources is becoming more critical as the global demand for energy continues to rise alongside growing environ- mental co...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/415179 |
| Acceso en línea: | https://hdl.handle.net/2117/415179 |
| Access Level: | acceso abierto |
| Palabra clave: | Hydrogen as fuel Hidrogen com a combustible Àrees temàtiques de la UPC::Energies |
| Sumario: | This master’s thesis has been carried out during the second quarter of the 2023-2024 school year, autonomously. The ever-increasing need for non-contaminating and perpetual energy sources is becoming more critical as the global demand for energy continues to rise alongside growing environ- mental concerns. Traditional fossil fuels, while historically reliable, contribute significantly to pollution and the depletion of natural resources. In response, there is a pressing shift towards sustainable and clean energy alternatives. All these renewable sources tend to be accompanied by a problem of unreliable production capabilities. To counteract this, hydrogen is emerging as a particularly promising solution in this landscape. As a versatile energy carrier, hydrogen can be produced from various renewable sources and has the potential to significantly reduce greenhouse gas emissions. Its ability to store and deliver energy efficiently makes it an attrac- tive option for a range of applications, from powering vehicles to generating electricity. The development and deployment of hydrogen technologies are gaining momentum, supported by advances in production methods and increasing investment. As the world moves towards a more sustainable energy future, hydrogen is poised to play a crucial role in meeting the dual challenges of reducing environmental impact and ensuring a reliable, long-term energy supply. Storing hydrogen poses several significant challenges. Hydrogen requires compression to very high pressures or cooling to extremely low temperatures to achieve a practical storage density, both of which involve substantial energy costs and technical complexity. Additionally, hydrogen is a small molecule that can easily escape through minute leaks, complicating the design and maintenance of storage vessels. Ensuring safe and efficient storage remains a key hurdle in the broader adoption of hydrogen as a mainstream energy carrier. A promising technology to overcome this are Metal Hydride (MH) tanks, where hydrogen is stored in solid form and does not require the high pressures and low temperatures as other techniques. The challenge this new technology poses is the estimation of its State of Charge (SoC), where an expensive flow meter is necessary to know it. In this thesis, the development of models to be able to estimate the flow going into an MH tank is covered. With only data about the pressure and temperature on the system. Some different architectures are tested and compared between them. The best performing techniques are com- bined and a best performing model is finally selected. With this model, the SoC is calculated and compared to the actual value. |
|---|