Soft computing techniques are important tools that significantly improve the performance of energy systems. This chapter reviews their many contributions to renewable energy hydrogen hybrid systems, namely those systems that consist of different technologies (photovoltaic and wind, electrolyzers, fuel cells, hydrogen storage, piping, thermal and electrical/electronic control systems) capable as a whole of converting solar energy, storing it as chemical energy (in the form of hydrogen) and turning it back into electrical and thermal energy. Fuzzy logic decision-making methodologies can be applied to select amongst renewable energy alternative or to vary a dump load for regulating wind turbine speed or find the maximum power point available from arrays of photovoltaic modules. Dynamic fuzzy logic controllers can furthermore be utilized to coordinate the flow of hydrogen to fuel cells or employed for frequency control in micro- grid power systems. Neural networks are implemented to model, design and control renewable energy systems and to estimate climatic data such as solar irradiance and wind speeds. They have been demonstrated to predict with good accuracy system power usage and status at any point of time. Neural controls can also help in the minimization of energy production costs by optimal scheduling of power units. Genetic or evolutionary algorithms are able to provide approximate solutions to several complex tasks with high number of variables and non-linearities, like optimal operational strategy of a grid-parallel fuel cell power plant, optimization of control strategies for stand-alone renewable systems and sizing of photovoltaic systems. Particle swarm optimization techniques are applied to find optimal sizing of system components in an effort to minimize costs or coping with system failures to improve service quality. These techniques can also be implemented together to exploit their potential synergies while, at the same time, coping with their possible limitations. This chapter covers soft computing methods applied to renewable energy hybrid hydrogen systems by providing a description of their single or mixed implementation and relevance, together with a discussion of advantages and/or disadvantages in their applications. Â© Springer-Verlag Berlin Heidelberg 2011.

Use of soft computing techniques in renewable energy hydrogen hybrid systems / Zini, Gabriele; Pedrazzi, Simone; Tartarini, Paolo. - 269:(2011), pp. 37-64. [10.1007/978-3-642-22176-7_2]

### Use of soft computing techniques in renewable energy hydrogen hybrid systems

#####
*ZINI, GABRIELE*^{};Pedrazzi, Simone;Tartarini, Paolo

^{};Pedrazzi, Simone;Tartarini, Paolo

##### 2011

#### Abstract

Soft computing techniques are important tools that significantly improve the performance of energy systems. This chapter reviews their many contributions to renewable energy hydrogen hybrid systems, namely those systems that consist of different technologies (photovoltaic and wind, electrolyzers, fuel cells, hydrogen storage, piping, thermal and electrical/electronic control systems) capable as a whole of converting solar energy, storing it as chemical energy (in the form of hydrogen) and turning it back into electrical and thermal energy. Fuzzy logic decision-making methodologies can be applied to select amongst renewable energy alternative or to vary a dump load for regulating wind turbine speed or find the maximum power point available from arrays of photovoltaic modules. Dynamic fuzzy logic controllers can furthermore be utilized to coordinate the flow of hydrogen to fuel cells or employed for frequency control in micro- grid power systems. Neural networks are implemented to model, design and control renewable energy systems and to estimate climatic data such as solar irradiance and wind speeds. They have been demonstrated to predict with good accuracy system power usage and status at any point of time. Neural controls can also help in the minimization of energy production costs by optimal scheduling of power units. Genetic or evolutionary algorithms are able to provide approximate solutions to several complex tasks with high number of variables and non-linearities, like optimal operational strategy of a grid-parallel fuel cell power plant, optimization of control strategies for stand-alone renewable systems and sizing of photovoltaic systems. Particle swarm optimization techniques are applied to find optimal sizing of system components in an effort to minimize costs or coping with system failures to improve service quality. These techniques can also be implemented together to exploit their potential synergies while, at the same time, coping with their possible limitations. This chapter covers soft computing methods applied to renewable energy hybrid hydrogen systems by providing a description of their single or mixed implementation and relevance, together with a discussion of advantages and/or disadvantages in their applications. Â© Springer-Verlag Berlin Heidelberg 2011.##### Pubblicazioni consigliate

I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.

In caso di violazione di copyright, contattare Supporto Iris