Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Design and simulation of a decentralized control system for a PEM water electrolyzer

Dapontis Stavros

Full record

---

Summary

We are living in an era of constant search for clean energy, i.e. energy that is as environmentally friendly as possible. The aim is sustainable development, creating a carbon-neutral society, and in this context, we are introducing hydrogen as an energy carrier. Hydrogen represents a clean and, at the same time, flexible energy carrier and for this reason it is at the centre of the scientific-research community and of politics. Hydrogen, as the energy carrier that it is, has the ability to store surplus power from renewable energy sources by converting it into hydrogen. The most widespread way of producing hydrogen is the electrolysis of water through an electrolysis plant. Thus, this thesis revolves around the process of modeling and designing a control system of a PEM type electrolysis plant.First, the theoretical and conceptual framework around the types of electrolysis units that exist is studied, with a greater emphasis on the PEM type electrolysis unit. Once the parts that make up the PEM type unit have been analysed, the process of mathematical modelling and design of the control system is started. The process of this system is described in a nonlinear model and then with the help of Taylor expansion, linearization of this is carried out. The purpose of this modelling is to create a state-space model together with the corresponding output-input transfer functions. The state-space model is used to design feedback controllers for the purpose of regulating hydrogen production by manipulating the electric current.Finally, controllers are evaluated and compared with each other to find the optimal one. Some important evaluation criteria are: significant reduction of disturbances fast response and minimization of potential errors as the value of the electric current is changed or not. Efficiency criteria such as ISE, IAE, ITAE and ITSE helped towards a better controller selection. A key point in finding the optimal controller is to try to keep the current usage low (hence low power) while maintaining high efficiency (H2 generation per power).

Available Files

Services