Cristin-resultat-ID: 1828921
Sist endret: 11. september 2020, 08:51
Resultat
Faglig kapittel
2020

Artificial Intelligence Models Used for Prediction in the Energy Internet

Bidragsytere:
  • Cristina Viorica Heghedus og
  • Rong Chunming

Bok

Energy Internet
ISBN:
  • 978-3-030-45453-1

Utgiver

Springer Nature
NVI-nivå 1

Om resultatet

Faglig kapittel
Publiseringsår: 2020
Sider: 321 - 352
ISBN:
  • 978-3-030-45453-1

Klassifisering

Fagfelt (NPI)

Fagfelt: Informatikk og datateknikk
- Fagområde: Realfag og teknologi

Beskrivelse Beskrivelse

Tittel

Artificial Intelligence Models Used for Prediction in the Energy Internet

Sammendrag

Decision-making in the energy field, especially in recent years, is highly data-driven. The crucial information extracted from historical data tells a lot about future behaviour and expected events. The extraction of this information is traditionally performed by statistical analysis and predicting future trends. However, in recent years, the amount of collected data has increased substantially and has become harder to handle by traditional means. The processing power and storage capacity of computers have also increased and have opened new horizons for data analysis and prediction tasks. One of the great advantages of modern technology is the possibility for automation in the execution of some tasks; thus, less human interaction and higher efficiency can be achieved. Therefore, in this chapter, the energy field is combined with the field of computer science that contributes to automation, that is, artificial intelligence (AI). The subject of artificial intelligence for the Energy Internet is analysed in this chapter mainly from the point of view of future consumption prediction (electricity, wind speed and solar radiation), and prediction models, efficiency measurements and implementation techniques are described in detail. Models are analyzed in terms of accuracy and design (stand-alone and hybrid models). Implementations include descriptions of different libraries used for AI models and characteristics of programming languages. Recently developed data processing and prediction models are numerous and highly efficient; thus, applying these models to data from the energy field has various advantages, including cost-effective resource management, asset management and energy efficiency.

Bidragsytere

Cristina Viorica Heghedus

  • Tilknyttet:
    Forfatter
    ved Institutt for data- og elektroteknologi ved Universitetet i Stavanger
Aktiv cristin-person

Rong Chunming

  • Tilknyttet:
    Forfatter
    ved Institutt for data- og elektroteknologi ved Universitetet i Stavanger
1 - 2 av 2

Resultatet er en del av Resultatet er en del av

Energy Internet.

Zobaa, Ahmed; Cao, Junwei. 2020, Springer Nature. TU, BULærebok
1 - 1 av 1