A smart current control for superconducting cable test
A smart current control for superconducting cable test Pasquale Arpaia1,2, Amalia Ballarino1, Vincenzo Daponte1,3, Domenico Maisto4, Giuseppe Montenero1,2, Cesare Svelto3 1CERN (TE-MSC-SCD, 30-04-030, CH-1211 Geneva 23, Switzerland, ph. , fax , ) 2Università del Sannio (Dept. of Engineering, Corso Garibaldi 102, 82100 Benevento, Italy, ph. , fax , ) 3Politecnico di Milano (Dip. Elettron. Inform e Bioing., Leonardo da Vinci 32, 20133 Milano, Italy, ph. , fax: , , ) 4CNR Institute for High Performance Computing and Networking (Via Pietro Castellino 111, 80131 Napoli, Italy, Ph. , fax ) Abstract- A smart current control system for enhancing performance of a superconducting cable test station is proposed. According to the Fuzzy Gain Scheduling logic, the controller parameters are modulated by the working conditions. An improved flexibility for the definition of the set point is achieved. The system effectiveness has been proved through several simulations and demonstrated experimentally on the Facility for Research on Superconducting Cables at European Organization for Nuclear Research (CERN). I. Introduction The superconductivity is widely exploited in many applications from applied physics to engineering. In all the applications, the characterization of the superconducting cables is a key point and has to be carried out thoroughly by means of ad hoc testing. In test facilities, often cables are tested by a superconducting transformer in order to cut operation cost [1]-[2]. In the Facility for Research on Superconducting Cables (FReSCa) at CERN, the current driven in the sample is controlled by a digital Proportional Integral (PI) control. However, bandwidth constraints require the definition of slow start and stop for current ramping. Furthermore, available control strategy limits the acceleration at 800 As-2. Finally, the system to be controlled, namely the chain current source and superconducting transformer, exhibits significant non-linear behaviour. Control performance can be improved by taking into account the system variations in the design of the control strategy. Adjusting control parameters according to the system behaviour is the main feature of the Adaptive Control (AC) strategies [3]. Assuming a deep knowledge on the system dynamic, the Gain Scheduling (GS) control strategy is the most suitable to perform this task [3], with a particular focus on the Fuzzy PI Gain Scheduler (FGS) [4]. The GS strategy divides the operating domain in several subspaces, according to the homogeneity of the system behaviour. In this paper, a smart current control system, for improving metrological performance of a superconducting cable measurement station, is proposed. The controller is based on an ad-hoc adaptive GS strategy, customized by means of a Takagi-Sugeno-Kan (TSK) Fuzzy inference [5], exploiting the representation capability of Fuzzy system to realize a suitable PI controller for each working condition. In particular, in Section II, state-of-the-art controllers of superconducting transformers are reviewed, and in Section III, a background on Fuzzy Gain Scheduling is reported. In Section IV, the proposed adaptive control system is outlined, and in Section V, the experimental validation results at FReSCa of CERN are illustrated. II. Digital control for superconducting transformer: a review In measurement stations based on superconducting transformers (Fig. 1), the cable is linked to the transformer secondary and the current Is flowing into the sample is sent through the feedback loop to the control system. 733 ISBN-10: 84-616-5438-2 | ISBN-13: 978-84-616-5438-3 19th IMEKO TC 4 Symposium and 17th IWADC Workshop Advances in Instrumentation and Sensors Interoperability July 18-19, 2013, Barcelona, Spain Figure 1 Architecture of a cable test station based on superconducting transformer. The measurement system integrates the voltage read from the Rogowski coils, by measuring the current on the sample; this value, subtracted from the current reference IRef,
Información del documento
- Subido en
- 27 de julio de 2024
- Número de páginas
- 6
- Escrito en
- 2023/2024
- Tipo
- Examen
- Contiene
- Preguntas y respuestas