29th Symposium on Fusion Technology (SOFT 2016)

P4.042 Preliminary Exception Handling Analysis for the ITER Plasma Control System

8 Sep 2016, 14:20

1h 40m

Foyer 2A (2nd floor), 3A (3rd floor) (Prague Congress Centre)

Board: 42

Poster C. Plasma Engineering and Control P4 Poster session

Speaker

Gerhard Raupp (Tokamak Scenario Development E1)

Description

To operate ITER and control long and finally thermonuclear discharges with very complex physics and a limited set of actuators requires a sophisticated Plasma Control System (PCS). To provide the required control functionality, the PCS will include many control loops to keep parameters within operation envelopes. These must be backed by exception handling functions, to optimize continuous control performance, autonomously perform controlled shutdown in critical situations, or assist interlock systems in pulse termination for investment protection. The Conceptual Design Review (CDR) for ITER PCS held 2013 covered all ITER operation phases. It provided a comprehensive set of system requirements and a preliminary functional analysis, and proposed a conceptual PCS design compatible to these. To prepare for the Preliminary Design Review (PDR) in late 2016, the PCS analysis and design is currently being extended and detailed. The goal is to show that candidate designs meet ITER requirements and assumptions, with focus on control of 1^stst plasma and early H/He phases of ITER operation (but keeping future high-performance phases in mind). Further aspects to be covered are integration of PCS exception handling with plasma state forecasting and with ITER CIS (Central Interlock System) protection mechanisms. Capturing and analysing the required exception handling capability follows a formal approach (see Treutterer et al., this conference): Separate research groups collect functional aspects needed to control (and handle exceptions during) nominal discharge phases (magnetic and kinetic control of plasma initiation, formation, ramp-up, flattop, ramp-down), or when disruptions and runaways occur (includes interfacing to CIS).  In this contribution we show how such results enter a database and how this information is used to find out how to organize exception handling, find recurring patterns, synthesise re-usable handling schemes, develop standardised methods to categorize and escalate exceptions and manage a rule-based decision taking throughout the system.