EPE Association: EPE 2023 - DS3n: Big Data and Artificial Intelligence in Energy Conversion

EPE 2023 - DS3n: Big Data and Artificial Intelligence in Energy Conversion
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2023 ECCE Europe - Conference > EPE 2023 - Topic 10: Data Analysis, Artificial Intelligence and Communication > EPE 2023 - DS3n: Big Data and Artificial Intelligence in Energy Conversion

   [return to parent folder]

   Data-Driven Component Cost Models for Power-Electronic Converters
By Carsten FRONCZEK, Niklas FRITZ, Rik W. DE DONCKER [View]
[Download]

Abstract: Multi-objective optimization significantlyshortens the design time for converters and improvesinitial design decisions. In addition to technical objectives,economic aspects such as cost play a role in almost everyapplication. This paper provides cost models for convertercomponents based on publicly available data from spring2023. These cost models are based on the largest datasetof any converter-component cost model in the literature.Besides physical properties of the components, they includeeconomic factors and allow their application fromprototype to series production. The cost models wereevaluated using statistical metrics. Further, the derivedcomponent cost models serve as a basis to a simplifieddesign optimization for a 10 kW dc-dc converter.

   Data-Driven Online Stability Diagnostics for Parallel Grid-Connected Converters in Weak Grid
By Caiyun QIN, Hui XU, Feng GAO [View]
[Download]

Abstract: This paper presents a data-driven online diagnostics method to identify the stability of parallel grid-connected converters. Specifically, the dataset consists of summed output current of all the parallel converters at the point of common coupling (PCC) and state label. Neural network algorithm is applied to infer steady state of parallel converters. The case study validates accuracy and efficiency of the proposed method.

EPE 2023 - DS3n: Big Data and Artificial Intelligence in Energy Conversion
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2023 ECCE Europe - Conference > EPE 2023 - Topic 10: Data Analysis, Artificial Intelligence and Communication > EPE 2023 - DS3n: Big Data and Artificial Intelligence in Energy Conversion
	[return to parent folder]

	Data-Driven Component Cost Models for Power-Electronic Converters By Carsten FRONCZEK, Niklas FRITZ, Rik W. DE DONCKER	[View] [Download]
Abstract: Multi-objective optimization significantlyshortens the design time for converters and improvesinitial design decisions. In addition to technical objectives,economic aspects such as cost play a role in almost everyapplication. This paper provides cost models for convertercomponents based on publicly available data from spring2023. These cost models are based on the largest datasetof any converter-component cost model in the literature.Besides physical properties of the components, they includeeconomic factors and allow their application fromprototype to series production. The cost models wereevaluated using statistical metrics. Further, the derivedcomponent cost models serve as a basis to a simplifieddesign optimization for a 10 kW dc-dc converter.

	Data-Driven Online Stability Diagnostics for Parallel Grid-Connected Converters in Weak Grid By Caiyun QIN, Hui XU, Feng GAO	[View] [Download]
Abstract: This paper presents a data-driven online diagnostics method to identify the stability of parallel grid-connected converters. Specifically, the dataset consists of summed output current of all the parallel converters at the point of common coupling (PCC) and state label. Neural network algorithm is applied to infer steady state of parallel converters. The case study validates accuracy and efficiency of the proposed method.