Instituto de Plasmas e Fusão Nuclear Instituto Superior Técnico Lisbon, Portugal http://www.ipfn.ist.utl.pt B. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition ITER CODAC ITER CODAC Bruno Soares Gonçalves [email protected]http://ipfn.ist.utl.pt/EU-PhD/
ITER CODAC. Bruno Soares Gonçalves [email protected] http://ipfn.ist.utl.pt/EU-PhD/. CODAC. CO ntrol, D ata A ccess and C ommunication system. Large-scale experiments: control and data aquisition challenges. The next generation of physics experiments will be highly complex - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Instituto de Plasmas e Fusão NuclearInstituto Superior TécnicoLisbon, Portugalhttp://www.ipfn.ist.utl.pt
B. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition2
CODACCODAC
COntrol, Data Access and Communication system
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition3
Large-scale experiments:Large-scale experiments:control and data aquisition challengescontrol and data aquisition challenges
The next generation of physics experiments will – be highly complex– raise new challenges in the field of control and
automation systems
– demand well integrated, interoperable set of tools with a high degree of automation
– deliver and process data at a rate of up to hundred GBytes/s.
– deploy and integrate systems with different degrees of complexity and provenience
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition4
Commercial technology will likely meet the
basic requirements on which physics
experiments can leverage for building future
control systems
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition5
However…However…
Future systems are envisioned to be more
than an order of magnitude larger
than those of today
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition6
ITER hardware developers/providers should be ready to
–Providing well integrated, interoperable set of tools
–Deploy and integrate systems with different degrees of complexity and provenience
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition7
providing robust, fault tolerant,
reliable, maintainable,
secure and operable control
systems
More challenging will be… More challenging will be…
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition8
Large-scale experiments:Large-scale experiments:R&D on control and data aquisition R&D on control and data aquisition
R&D activities target
– Self-triggered front-end electronics with
adequate output bandwidth and data processing
– MIMO controllers with efficient resource sharing between control tasks on the same unit
– Massive parallel computing capabilities.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition9
Is ITER any different?Is ITER any different?
ITER CODAC is a challenging endeavour
• ITER will generate a huge quantity of experimental data
– 150 plant systems– 1 000 000 diagnostic channels– 300 000 slow control channels– 5 000 fast control channels– 40 CODAC systems– 5 Gb/s data– 3Pb/year data (e.g. 12 IR cameras
in a 10 minutes discharge: 1.728 Tbytes)
In addition...
ITER will require a far higher level of availability and
reliability than previous/existing tokamaks.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition10
Why do we need it?Why do we need it?
Real Time Measurement and Control of magnetically confined
plasmas is a critical issue for safe operation and
high performance of fusion reactors.
It ensures:
•Reliability: control the plasma dynamic over long times
•Reproducibility: avoid deviations from the reference scenario
•Plasma & machine protection: avoid of instabilities & disruptions
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition11investmentenvironment components
Protection to the…
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition12
What are the functions of the What are the functions of the control systems?control systems?
Fusion machines have a wide range of Plasma Control Systems
– plasma control tools: to reach and reproduce scenarios which cannot be programmed
– operation supervisor: to protect Plasma and Machine
Both thrusts will be steered by the need to satisfy regulatory requirements while effectively
controlling a burning plasma.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition13
What are the inherent constraints?What are the inherent constraints?
Fusion reactors sites will be a Nuclear Registered sites
• Similar to Nuclear power station in terms of quality, audit trails, ..... etc.
• Will need control system architecture within the licensing constraints
• Requires tools to guarantee safety, protection of investment and
guaranteed operation
• Hostile environment for measurements, networks, electronics – human
access will be restricted
• Will require a far higher level of availability and reliability unforeseen on
previous/existing tokamaks
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition14
Are there similarities with Are there similarities with fission reactors?fission reactors?
Operation supervisory tools will likely be similar to the ones
in use in fission reactors
– slow response time of the control system the order of hundreds of milliseconds.
– Typical functionality of an industrial SCADA (Supervisory Control And Data Acquisition)
• displays on mimic diagrams, • trending, • warning and alarm handling, • manual triggering of commands or changes to set-points.
However, fusion reactors are expected to explore more advanced operation scenarios capable of
– sustaining for a long duration a steady-state plasma
– suppressing plasma instabilities almost completely.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition15
ITER Challenges:ITER Challenges:Inherent ConstraintsInherent Constraints
ITER is assumed to be totally legacy-free for hardware and software• Methodologies will have to be tested and proved on existing fusion
devices before implementation on ITER• It will be necessary to take informed decisions based on technology
progress• Maintenance will be an issue and proliferation of technologies must be
avoided
ITER is an international project• In-kind procurement world-wide• Integration of Plant Systems from all participants• The implications of in-kind delivery of subsystems need to be recognised• Powerful remote access networks• Remote access security
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition16
ITER Challenges:ITER Challenges:Scientific ExploitationScientific Exploitation
ITER will generate huge quantities of experimental data – PBytes per year (but still less than LHC)(e.g. 12 IR cameras data resulting from a 10 minutes shot: 1.728 TBytes)
ITER will provide tools for continuously accessing and analysing data during a pulse - Requires data indexing by events
ITER will have a very strong flexible set of diagnostics and tools for optimising the performance during a pulse - Adequate tools and methodologies need to be developed
ITER will have a limited number of pulse cycles and an unlimited number of ideas to be tested– Will schedule and reschedule many activities during a pulse
ITER will evolve both equipment and ideas over 20 years– A lifetime of 30 years including procurement – evolution must be built
into CODAC
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition17
What is required for advanced What is required for advanced operation scenarios?operation scenarios?
For fusion burn control is essential to integrate simultaneously – multiple measurements from different sensors, – real-time plasma modelling from several tools – multiple actuators
Requires fast control loops with time constraints of the order of tenths of microseconds.
Furthermore, the control systems are mandatory to be robust, and fault tolerant
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition18
ITER CODACITER CODAC
CODAC provides the COntrol, Data Access and Communication functions
for ITER, allowing integrated operation.
This includes:
•continuously monitoring the Plant Systems;
•displaying their status to operators including alarms;
•preparing and automating scheduled operations (including plasma pulses);
•recovering data from Plant Systems;
•storing and making all the data available.
CODAC uses multiple logical and physical networks to segregate these disparate functions.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition19
ITER: ITER: as seen by CODACas seen by CODAC
• CODAC integrates ALL ITER Plant Systems
• Many networks: operation, interlocks,
safety
• CODAC functions are like present tokamaks
Typical CODAC Functions
Time & Synchronisation
Data Access
Gateway
Safety NetworkCODAC Network
I nterlock Network
Coil power supply & distribution
system
TF coil system
PF coil system
Correction coil system
CS coil system
Diagnostic systems
Additional heating & power supply systems
Gas inj. system
Pellet injector system
Remote handling system
Component cooling & chilled
water system
Tokamak cooling water
system
Heat rejection system
Tritium plant
system
Cryoplant system
Glow discharge system
Radiation & Environmental
monitoring system
Cryostat & VV pressure suppression
system
FPSS VVPSS
Access control
ECH&CD system
NBH&CD system
ICH&CD system
LHH&CD system
Experiment Sites
Central Interlock System
Steady state power supply
system
Vacuum pumping system
Magnet I&C
system
Miscellaneous instrumentation
systems
Liquid distribution
Gas distribution & compressors
Central Safety System
Plant Monitoring
Operation Scheduling
Plant Operation
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition20
ITER:ITER:Instrumentation and ControlInstrumentation and Control
I&C is in 3 clear tiers– Safety: protects
personnel, population and environment
– Interlock: protects ITER investment
– CODAC: operates ITER
I&C is in 2 layers
– Plant Systems: local responsibility
– Networks when responsibility lies across Plant Systems
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition21
ITER Instrumentation & Control ITER Instrumentation & Control System physical architectureSystem physical architecture
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition22
CODACCODAC
CODAC shall provide:
• Supervisory control functions as a project wide and interface specifications between CODAC and Plant Systems.
• Central data management functions i.e. data archiving, data monitoring and visualization functions.
• PSH functional profile and data exchange software for the asynchronous communication interface between CODAC and Plant System.
• Mini CODAC as a tool to carry out FAT (Factory Acceptance Test)
• Specifications for the Network Interface Units with their interface specifications.
• Self-description schema and tool for Plant System I&C designers.
• Functional mimic diagrams for Plant System data monitoring, trends, plasma discharge preparation, sequencing, and data display.
• Functions for global and plant operating states management, plasma control, data recording with time stamp, data marshalling, data archiving...
• Capability to log all commands and state transitions along with all the associated data with time stamping.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition23
ITER:ITER:Plant systems I&CPlant systems I&C
PSH
CODAC System / Mini-CODAC
Plant System Controller
Signal Interface
Sensors/ Actuators
Pla
nt
Sy
ste
ms
I&
C
PON
PSHPSHPlant System
ControllerPlant System
Controller
Signal Interface
Signal Interface
Sensors/ Actuators
Sensors/ Actuators
PON
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition24
Network Interface between plant Network Interface between plant system I&C and central I&C systemssystem I&C and central I&C systems
SDN
AVNTCN
CSN
PON
CIN
PSH
Plant System Controller
Signal Interface
PISController
PSSController
Plant Control System Plant Interlock System Plant Safety System
Plant System I&C
I/F
I/F
I/F
I/F
I/F
Network Panel
Signal Interface Signal Interface
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition25
CODAC and Plant Systems I&C CODAC and Plant Systems I&C architecturearchitecture
ITE
R_C
OD
AC
_P
CD
H_
Fig
ure
s_
Vis
io_0
00
3
CODAC Networks
Equipment Equipment Equipment
Equipment
Subsystem(PLC)Subsystem
(PLC)
Digitizers
Subsystem(PLC)
Subsystem(PLC)
Plant System - A Plant System - B Plant System - C
Subsystem(PC)
Subsystem(PLC)
Subsystem(PLC)
Digitizers
Digitizers
Digitizers Equipment
CODAC Systems
I&C Bridge I&C Bridge I&C Bridge
Plant System Host Plant System Host Plant System Host
CODAC Systems
Provides supervisory functions of ITER plant operation, plasma experiment, overall ITER plant operating status monitoring, data archiving and alarm handling, HMI interface and remote experiment handling functions.
Modular design with Dual Redundancy required
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition26
CODAC componentsCODAC components
The principal CODAC System is the Supervisory Control System
The Supervisory Control System:
• dynamically allocates any required resources to an ITER operation Task.
• manages a dynamically evolving set of concurrent activities, each of which is driven by an Operation Schedule.
The Operation Schedule
• is prepared by Schedule Preparation and each Operation Schedule requires Schedule Validation before becoming executable.
• is executed by Schedule Execution once the resources are made available by SCS.
There is a strong interface between scheduling and ITER operation planning
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition27
CODAC componentsCODAC components
The status of ITER is obtained from Plant Monitoring, which also generates a data stream for Data Logging.
•Maximum refresh frequency proposed is 3 Hz, corresponding to a human reaction.
•Minimum rate is 0.1 Hz to ensure a continuous record and continuous functionality checking.
Monitoring data are available in the Experiment Sites to enhance contact with operation.
The functionality is typical of an industrial SCADA (Supervisory Control And Data Acquisition)
•displays on mimic diagrams,
•trending,
•warning and alarm handling,
•manual triggering of commands or changes to set-points.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition28
Central data management functionsCentral data management functions
• CODAC shall provide data logging, data monitoring, data archiving and data visualisation functions within the Main Control Room.
• Data are generated by different parts of the ITER plant at different sampling rates according to the information itself and according to the operational state of ITER.
• The notion of pulse archive also disappears, replaced by the data in a particular pulse being and identifiable time interval in a continuous data retrieval stream.
• Storage strategies required for efficient recovery of data taken at different sampling rate (0.1 Hz to 1 MHz). Separating the total data flow into suitable streams to optimize the retrieval of archived information.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition29
•the operation should be considered continuous and the ageing ‘shot’ paradigm shall be replaced•Event-driven support where data is acquired or actions performed only when relevant events occur
– data indexing using events and time-stamps with synchronism of all digitizer and generator endpoints (absolute time)
•Provide tools for continuously accessing and analysing data during a pulse
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition30
CODAC and Plant Systems I&C CODAC and Plant Systems I&C architecturearchitecture
ITE
R_C
OD
AC
_P
CD
H_
Fig
ure
s_
Vis
io_0
00
3
CODAC Networks
Equipment Equipment Equipment
Equipment
Subsystem(PLC)Subsystem
(PLC)
Digitizers
Subsystem(PLC)
Subsystem(PLC)
Plant System - A Plant System - B Plant System - C
Subsystem(PC)
Subsystem(PLC)
Subsystem(PLC)
Digitizers
Digitizers
Digitizers Equipment
CODAC Systems
I&C Bridge I&C Bridge I&C Bridge
Plant System Host Plant System Host Plant System Host
CODAC Networks
Provide ITER wide physical and logical interconnections between CODAC and the Plant Systems.
The roles and functions of networks are defined according to their performance requirements.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition31
CODAC NetworksCODAC Networks
The CODAC architecture is based on distributed systems connected
over a set of complementary asynchronous and synchronous networks
Each Plant System and CODAC Systems can communicate over one or more networks
•Asynchronous general purpose Plant Operation Network (PON) provides the backbone of CODAC communication for most CODAC data traffic.
•General ITER Networks (GIN) used to connect between the Plant Operation Zone and external Experiment and Analysis Sites.
Networks used will depend on the required functionality, volume of data, bandwidth and latency.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition32
High Performance NetworksHigh Performance Networks
Some functions of CODAC require deterministic, hard real-time communication and synchronization between distributed nodes.
These requirements are addressed by the CODAC High Performance Networks
•Synchronous DataBus Network (SDN)
•Time Communication Network (TCN)
•Event Distribution Network (EDN)
•Audio Video Network (AVN)
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition33
Issues:Issues:Timing, Synchronization, events and Synchronous Timing, Synchronization, events and Synchronous data Transport Networksdata Transport Networks
Issues Actions
Distribution of timing signals from central unit to sub-system timing decoder cards where timing signals are subsequently connected to digitizers and control processors
•Include timing, synchronization and event decoder directly on digitizers and control processors
– Higher reliability
– Improve commissioning and maintenance efforts
– Possibility of performing the timing and synchronization of control processes over the synchronous network (IEEE 1588)
Absolute time concept •CODAC shall provide absolute time broadcast; triggers are timing events; clocks are recovered from the synchronous data (optical) links used for events transport.
– Autonomous local absolute time clock units on the digitizers synchronized through a fast network
New diagnostics and plasma controllers may require an updated real-time control and monitoring hardware infrastructure
•Specify a new synchronous network infrastructure
– Lower latency distribution of plasma variables and events (< 2 us)
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition34
CODAC and Plant Systems I&C CODAC and Plant Systems I&C architecturearchitecture
ITE
R_C
OD
AC
_P
CD
H_
Fig
ure
s_
Vis
io_0
00
3
CODAC Networks
Equipment Equipment Equipment
Equipment
Subsystem(PLC)Subsystem
(PLC)
Digitizers
Subsystem(PLC)
Subsystem(PLC)
Plant System - A Plant System - B Plant System - C
Subsystem(PC)
Subsystem(PLC)
Subsystem(PLC)
Digitizers
Digitizers
Digitizers Equipment
CODAC Systems
I&C Bridge I&C Bridge I&C Bridge
Plant System Host Plant System Host Plant System Host
Plant Systems provide data acquisition, operation & control, status/alarm monitoring, and data communication functions with the CODAC systems.
Also have local autonomous operation control independent from the CODAC.
Plant Systems communicates only through CODAC.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition35
CODAC and Plant Systems I&C CODAC and Plant Systems I&C architecturearchitecture
ITE
R_C
OD
AC
_P
CD
H_
Fig
ure
s_
Vis
io_0
00
3
CODAC Networks
Equipment Equipment Equipment
Equipment
Subsystem(PLC)Subsystem
(PLC)
Digitizers
Subsystem(PLC)
Subsystem(PLC)
Plant System - A Plant System - B Plant System - C
Subsystem(PC)
Subsystem(PLC)
Subsystem(PLC)
Digitizers
Digitizers
Digitizers Equipment
CODAC Systems
I&C Bridge I&C Bridge I&C Bridge
Plant System Host Plant System Host Plant System Host
Plant System Hoststandard image of Plant System to the CODAC.
It is a single point entry for the asynchronous communication with CODAC (data exchange)
Controls data flow, interprets all the commands and passes them to the Subsystem Controller for necessary actions.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition36
Plant System supplier’s Plant System supplier’s responsibilityresponsibility
Plant System suppliers shall:
– Provide self-description data of their Plant System I&C and shall receive interface requirements from the CODAC.
– Provide and implement applications for their Plant System monitoring, data acquisition, autonomous operation and control functions.
– Provide Plant System simulators/Plant subsystem simulators.
– Carry out Factory Acceptance Tests using mini-CODAC as a testing tool. Also Plant System is responsible to carry out installation, commissioning and SAT at the IO site.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition37
Mini-CODACMini-CODAC
Development of a mini-CODAC mandatory • required to test all CODAC concepts prior to full development• provide a development and test tool for Plant System designers
Mini-CODAC is a CODAC emulator for development of Plant Systems
• Tool for carrying out functional testing of the Plant System to certify Plant System functional integration.
• Scalable functionality to achieve limited performance testing of the Plant System interfaces with CODAC.
• Used to carry out Provisional Acceptance Tests (and for repeating the FAT if needed) to prove and verify that all plant cabling and connections have been terminated correctly and that the input and output schedule is as required by the design.
• Does not define the technical functionality and test processes of Plant Systems but defines and provides environment with limited performance to facilitate integration testing of Plant Systems with CODAC.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition38
Plant System SimulatorPlant System Simulator
PSH
Mini CODAC Plant System # 1 Simulator
Plant System # 2 Simulator
Plant System simulator
ITER
_CO
DAC_
PCDH
_Fig
ures
_Vis
io_0
005
Plant Subsystem simulator
Used to test the behaviour of Plant Systems during different phases of integration (Factory Acceptance Tests, Commissioning and Site Acceptance Tests and also during plant operation).
Plant System/Sub-system simulator should be based on self description data from Individual Plant System/sub-system supplier.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition39
CODAC and Plant Systems I&C architectureCODAC and Plant Systems I&C architecture
ITE
R_C
OD
AC
_P
CD
H_
Fig
ure
s_
Vis
io_0
00
3
CODAC Networks
Equipment Equipment Equipment
Equipment
Subsystem(PLC)Subsystem
(PLC)
Digitizers
Subsystem(PLC)
Subsystem(PLC)
Plant System - A Plant System - B Plant System - C
Subsystem(PC)
Subsystem(PLC)
Subsystem(PLC)
Digitizers
Digitizers
Digitizers Equipment
CODAC Systems
I&C Bridge I&C Bridge I&C Bridge
Plant System Host Plant System Host Plant System Host
EquipmentMay not communicate in a project-wide standardised form.
Configured hierarchically according to the individual Plant System design.
Not procured for direct interface with CODAC but subject to an integration procurement arrangement to deliver as a Plant System component.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition40
Sub-systems and equipments Sub-systems and equipments
Control System Interface Definition Optics alignment controller
interfaces Laser alignment unit interfaces Lasers controller interfaces Detection system controller
interfaces Windows monitoring unit
interfaces Inner wall monitoring unit
interfaces Control and monitoring software
specification Local controller hardware
specification
Radial Port Labyrinth
Primary Vacuum Window
Cryostat Window
Biological Shield Labyrinth
Beam Combiner
532 nm Lasers
Calibration Laser
1064 nm Lasers
PLASMA
VACUUMVESSEL
CRYOSTAT
PIT
M4
M5
LinearServomotors
Linear Servomotors
Alignment Optics Feedback Controller
CCCD
PIT RELAYS
Laser Alignment Feedback Controller
CCCD
CCCD Blanket
penetration
Primary imaging mirror
Laser Controller
Filter Spectometer
Gated Fast Detectors
(MCP PMTs)
Data Acquisition 20 GSPS, 10-bit,
6 GHz, 500 ps
Collected LightTransmission Line
Laser Transmission
Line
Detection System
Controller
LOCAL CONTROL
Windows Monitoring Unit
Inner-Wall Monitoring Unit(to be defined)
7 beams1 beam
7 beams
CCCD Back wall
2 cameras
XY axisXY axis
Laser control, triggering, fault detection and monitoring lines
40LIDAR CONTROL AND DATA ACQUISITION as example
Diagnostic systems will have local controllers and data acquisition which may have to be developed to meet the requirements
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition41
Issues:Issues:Data AcquisitionData Acquisition
Issues Actions
The size of the data collected in each shot is increasing(ITER will generate huge quantities of data. e.g 12 IR cameras data resulting from a 10 minutes shot: 1.728 TBytes)
• Implement faster data transport to comply with cycle-time
• Higher-speed real-time pulse processing both during and after shot
High sampling rates and data bandwidth
• Data reduction techniques directly in the sub-system, preferably on the digitizer modules
– Increment the transient recorder time window
– Solve local bandwidth bottlenecks
– improve the CODAC performance since data reduction tasks will be parallelized, lessening data management burden on CODAC infrastructure thus helping meeting the operation delay constrains
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition42
Multitude of different hardware platforms to maintain
• ‘generic’ sub-system able to perform one or more of the (fast) local control operations, feedback and data acquisition tasks
• Develop for easy deployment and maintenance
Higher processing power requirements for local data reduction techniques, fast real-time local control or other specialized algorithms on the digitizers
• Use of processors with parallel processing capabilities (multicore, FPGAs,…), interconnected through a (multiple-) full-mesh topology low-latency high bandwidth network for transport of variables and data streams
Ease the programming of local real-time processing algorithms by non-specialized users
• Development of a set of helper tools for easier programming, simulation and deployment of code on reconfigurable devices (Field Programmable Gate Array - FPGA) and multi-core processors
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition43
Improve systems reliability •Standards based instrumentation with inherent redundancy and mechanical/ thermal characteristics (e.g. Advanced Telecommunications Computer Architecture - ATCA)
Local and global management of hardware operation is required to improve maintainability
•Implementation of a standards based improved hardware management interface (e.g. Intelligent Platform Management Interface (IPMI), Shelf (crate) management (ShM)
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition44
Easier installation/replacement of hardware modules
•Hardware interface designed for ‘Plug-and-Play’, and ‘Hot-swap’e.g. ITER will evolve both equipment and ideas over 20 years therefore evolution must be built into CODAC
Improve the speediness of sub-system commissioning and integration on global CODAC
•Design the local sub-system control and data acquisition software for autonomous operation•Preliminary test operations during development can be performed autonomously
Scalability
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition45
CODAC componentsCODAC components
Plasma Control is implemented as a specific Operation Schedule to maximise reuse of automation and plasma control tools.
•General feedback control, including Plasma Control, uses a Synchronous DataBus to communicate data converted to physics units, including an estimate of the error on each signal and its status.
•Evaluation of plasma diagnostic information is local in the diagnostic Plant System if this is straightforward.
•Information is collected over the Synchronous DataBus for analysing data from multiple Plant Systems and finally transmitted over the Synchronous DataBus to the actuators.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition46
JET has a wide range of Plasma Control Systems
Aim of Control Systems in fusion:Aim of Control Systems in fusion:JET as an exampleJET as an example
Magnetics
R-T Signal Server
R-T Controller
plasma
CXS Ti (R)
MSE pitch (R)
Flux surfaces EQX
Confinement
VUV impurities
Shape & Current Control (PPCC)
ECE Te (R)
q profile
Neutron X-ray etc.
Interferom/Polarim
NBI
ICRH
LHCD
GAS + Pellets
PF Coils
Vis H/D/T
Vis Da, Brem, ELM
Comms network ATM, some analogue
X-ray Ti (0)
LIDAR Ne&Te(R)
Simulink codeEQX kinetic map
Pale blue = Diagnostic, Sky blue = Analysis, Red = Heating / Fuelling / Magnets & Power, Yellow = PPCC (XSC), Green = RTMC
TAE / EFCC
Wall Load
Coil Protection
To reach and reproduce scenarios which cannot be programmed• Quasi-Steady State Experiments• Magnetic and Kinetic Profile & ITB Experiments• Mode Conversion Experiments• Radiation and Impurity Experiments• MHD Experiments
To protect Plasma and Machine• NB Shinethrough (WALLS, PEWS and NBLM)• LHCD Launcher (Monitor Iron and Radiation)• Avoid Disruptions• Mimize waste (Neutrons, Tritium)
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition47
High Maintenance•Proliferation of interfaces•Should converge on modern instrumentation, computer, etc. standards
Lack of commonality and functionality between different devices RTMC systems•e.g. JET PPCC not simply exportable to other devices and vice-versa
Lack of flexibility•Integration of new equipment and physics into the existing infrastructure is time-consuming.
Lack of good transport and integrated models and tools•Integrated development environment and interchange formats
Future developments should acknowledge these issues
Limitations of existing Control Systems: Limitations of existing Control Systems: JET as an exampleJET as an example
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition48
Issues:Issues:Control SystemsControl Systems
Issues Actions
Higher algorithm complexity and higher number of input signals for plasma control
• Use of MIMO plasma controllers• Higher real-time processing power
Lower loop delays for time-critical real-time control
• Faster, lower latency real-time local synchronous network for variable sharing among processors
Faster design cycle of RT control systems
• Develop an improved integrated framework for code development, simulation and deployment
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition49
How important are the control systems?How important are the control systems?Vertical Stabilization as exampleVertical Stabilization as example
•Loss of control if plasma reaches the vessel protecting tiles.
•Dedicated MIMO systems are designed to make the plasma vertically stable so that other controllers can successfully control the plasma position and shape.
From F. Sartori, IEEE CONTROL SYSTEMS MAGAZINE, APRIL 2006
Elongated plasmas are vertically unstable
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition50
How important are control systems?How important are control systems?ITER Vertical Position ControlITER Vertical Position Control
• Loss of vertical plasma position control in ITER will cause thermal loads on
Plasma Facing Components of 30-60 MJ/m2 for ~0.1s.
– PFCs cannot be designed to sustain such (repetitive) thermal loads
• Vertical Displacement Events also generates the highest electromagnetic loads
– A phenomenological extrapolation of horizontal forces from worst JET cases implies horizontal loads ~45MN on ITER vacuum vessel.
– MHD wetted kink model developed to simulate the horizontal loads predicts ~20MN
– Vertical loads ~90MN
Plasma vertical position in ITER must be robust & reliable to ensure a vertical plasma position control
loss is a very unlikely event
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition51
JET Vertical Stabilization:JET Vertical Stabilization:Specifications of the MIMO architectureSpecifications of the MIMO architecture
• Reduction of – loop delay on the signal acquisition/generation endpoints (down to 10 µs)
– on the data interconnect links from and to the processing unit
– on the analogue signal path (analogue filters);
• High processing power– on the acquisition/generator endpoints
– on the system controller;
– improvement of MIMO algorithm performance
• Synchronization of all digitizer/generator endpoints;
• Architecture designed for maintainability, upgradeability and scalability;
• Low cost per channel;
• Low risk of implementation and testing.
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition52
Does it work?Does it work?ATCA @ JET Vertical StabilizationATCA @ JET Vertical Stabilization
Front view Rear view
192 input signals
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition53
• 192 signals acquired by ADCs and transferred at each cycle
• 50 s control loop cycle time with jitter < 1 s
• Always in real-time (24 hours per day)
•1.728 x 109 50 s cycles/day
•Crucial for ITER very long pulses
Does it work?Does it work?ATCA @ JET Vertical StabilizationATCA @ JET Vertical Stabilization
192 input signals
Front view
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition54
ITER relevant technology:ITER relevant technology:ATCAATCA
ATCA is the most promising architecture to substantially enhance the performance and capability of existing standard systems
• It is designed to handle tasks such as – event building, – feature extraction – high level trigger processing.– TeraOPS of processing power in a single sub-rack.
• First commercial open standard designed for – high throughput – High availability
High interest to data acquisition physics and attractive for
experiments requiring a very high up-time
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition55
Demanding solutions require robust hardware:Demanding solutions require robust hardware:ATCA is an ITER relevant solutionATCA is an ITER relevant solution
ATCA is the most promising architecture to substantially enhance the performance and capability of existing standard systems
•It is designed to handle tasks such as – event building, – feature extraction – high level trigger processing.– TeraOPS of processing power in a single sub-rack.
•First commercial open standard designed for – high throughput – High availability
High interest to data acquisition for physics and attractive for
experiments requiring a very high up-time
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition56
Why ATCA?Why ATCA?
ATCA platform is gaining traction in the physics community because of • Advanced communication bus architecture (serial gigabit replacing parallel buses)
• very high data throughput options and its suitability for real-time applications
• Agnostic backplane that accepts several serial switch network protocols
• Scalable shelf capacity to 2.5Tb/s
• Scalable system availability to 99.999%
• Robust power infrastructure (distributed 48V power system) and large cooling capacity (cooling for 200W per board)
• High levels of modularity and configurability
• Ease of integration of multiple functions and new features
• The ability to host large pools of DSPs, NPs, processors and storage
• The ability to host multiple controllers and storage on a shelf
• High security and regulatory conformance
• Reliable, full redundancy support
• Reliable mechanics (serviceability, shock and vibration)
• Hardware management interface
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition57
ATCA @ JET Vertical Stabilization:ATCA @ JET Vertical Stabilization:“the star of the show”“the star of the show”
IPFN’s ATCA-MIMO-ISOL
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition58
Does it work?Does it work?
VS controller in action - Vertical kicks experiment
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition59
And when something fails…And when something fails…
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition60
CODAC milestonesCODAC milestones
• First plasma assumed to be 2019
• All functionality available – end 2015
• Control room and site distribution operational – end 2013
• First integration – end 2013
• Mini-CODAC operational – end 2010
• Plant System Host –end 2010
These dates may be revised (and certainly not for earlier)
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition61
“These days, building the best system isn’t enough. That’s the price of entry.”
Ann Livermore, HP
For fusion reactors the remaining price should be paid on…
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition62
SummarySummary
ITER CODAC is a challenging endeavour
Author’s name | Place, Month xx, 2007 | EventB. Gonçalves | Lisbon, February 8, 2010 | Diagnostics & Data Acquisition63
A professor is only as good as the A professor is only as good as the questions he raises in his pupils minds...questions he raises in his pupils minds...