The Integration of Two Control Systems
The Integration of Two Control Systems

Author:

Publisher:

Published: 1995

Total Pages: 3

ISBN-13:

DOWNLOAD EBOOK

During the past year the Continuous Electron Beam Accelerator Facility (CEBAF) has installed a new machine control system, based on the Experimental Physics and Industrial Control System (EPICS). The migration from CEBAF's old control system, Thaumaturgic Automated Control Logic (TACL), had to be done concurrently with commissioning of the CEBAF accelerator. The smooth transition to EPICS was made possible by the similarity of the control systems' topological design and network communication protocol. Both systems have operator display computer nodes which are decoupled from the data acquisition and control nodes. The communication between display and control nodes of both control systems is based on making named requests for data, with data being passed on change of value. Due to TACL's use of a central communications process, it was possible to integrate both control systems' network communications in that process. This in turn meant that CEBAF did not require changes to any other software in order to support network communication between TACL and EPICS. CEBAF implemented the machine's control under EPICS in an evolutionary, controlled manner. 4 refs., 3 figs.

The CEBAF Control System
The CEBAF Control System

Author:

Publisher:

Published: 1995

Total Pages: 4

ISBN-13:

DOWNLOAD EBOOK

CEBAF has recently upgraded its accelerator control system to use EPICS, a control system toolkit being developed by a collaboration among laboratories in the US and Europe. The migration to EPICS has taken place during a year of intense commissioning activity, with new and old control systems operating concurrently. Existing CAMAC hardware was preserved by adding a CAMAC serial highway link to VME; newer hardware developments are now primarily in VME. Software is distributed among three tiers of computers: first, workstations and X terminals for operator interfaces and high level applications; second, VME single board computers for distributed access to hardware and for local control processing; third, embedded processors where needed for faster closed loop operation. This system has demonstrated the ability to scale EPICS to controlling thousands of devices, including hundreds of embedded processors, with control distributed among dozens of VME processors executing more than 125,000 EPICS database records. To deal with the large size of the control system, CEBAF has integrated an object oriented database, providing data management capabilities for both low level I/O and high level machine modeling. A new callable interface which is control system independent permits access to live EPICS data, data in other Unix processes, and data contained in the object oriented database.

Fast Feedback System for CEBAF.
Fast Feedback System for CEBAF.

Author:

Publisher:

Published: 1995

Total Pages:

ISBN-13:

DOWNLOAD EBOOK

A fast feedback system based on concepts of modern control theory has been implemented in the CEBAF Control System to stabilize various machine parameters. The continuous wave operation of CEBAF requires that parameters such as beam energy and position are stabilized against fast fluctuations. The beam energy must be stabilized against fast gradient and phase fluctuations in the RF accelerating system. This fast feedback system currently operates at 60 Hz rate and is integrated with EPICS. The mathematical model of the system for various feedback loops is expressed in state space formalism. The design of control law and simulation of closed-loop system response is performed using MatlabTM and SimulinkTM. This paper describes the process of designing control algorithms, implementation of the fast feedback system and operational experience with this system at CEBAF. The performance of this feedback system, while operating at much higher rates with high closed loop gain, can be enhanced by continually performing on-line identification of the system from the input and output data. System identification is the process of developing or improving an analytically derived mathematical representation of a physical system using experimental data. The current status of this feature is presented.