Mario Batz
European Organization for Nuclear Research (CERN), Geneva, Switzerland

            A major challenge in process operation is to reduce costs and increase
            system efficiency whereas the complexity of automated process
            engineering, control and monitoring systems increases continuously. To
            cope with this challenge the design, implementation and operation of
            process monitoring systems for control room operation have to be treated as
            an ensemble. This is only possible if the engineering of the monitoring
            information is focused on the production objective and is lead in close
            collaboration of control room teams, exploitation personnel and process
            specialists. In this paper some principles for the engineering of monitoring
            information for control room operation are developed at the example of the
            exploitation of a particle accelerator at the European Laboratory for Nuclear
            Research (CERN).

Many similarities exist between CERN and production industries concerning the processes, process
control and control room operation. CERN’s main production objective is related to particle beams
and luminosity, for which various traditional industrial processes are necessary, such as water-cooling
and electrical systems. Still today production processes cannot by automated completely. The remote
monitoring from a control room is an essential element of the total effectiveness to meet the
production objective. The process information that is available for control room operators determines
amongst other things the correctness and speed of their decisions and actions.
       In preparation of the LHC operation the SPS/LHC accelerator control room (PCR) and the
Technical Control Room (TCR) have launched a project to study the recovery of the Super Proton
Synchrotron (SPS) after major system breakdowns. The activities of both the control rooms shall be
focused on the main production objective. The documentation of operation activities, as well as the
specification of process monitoring tools and the needs of operator training will be established to
minimize the down time of the SPS.
      The monitoring information presented to the control room operators has to be defined in close
collaboration between control room teams, exploitation personnel and process specialists. For the
basic principles and requirements are identical for many processes, a method shall be available to
rationalize the engineering of the monitoring information that is applicable to the future LHC
operation and other CERN accelerators.

During the engineering of the TCR monitoring information the particle beam production has not been
considered systematically and the design approach has been different for different processes. In the
same way the monitoring of the technical infrastructure has not sufficiently influenced the monitoring
information of the PCR [1].
       In the PCR and the TCR the workload, knowledge and background of the operators are
different and the operation teams are hosted in different control rooms. Even more important is that
the work objectives after major incidents of the control rooms are not identical:
 - For the TCR: the restart of the technical infrastructure for the whole laboratory where the
   accelerators do not have exclusive priority
 - For the PCR: the restart of the accelerator equipment and finally the particle beam
      The monitoring of the technical infrastructure of CERN’s accelerators by the TCR is based on
alarm list displays, synoptic diagrams and trend curves of process values. The processes under control
 - electrical distribution and supply,
 - ventilation and air-conditioning,
 - water distribution and cooling, demineralized water production and distribution,
 - beam vacuum and compressed air,
 - monitoring and control system infrastructure,
 - fire and gas detection equipment,
 - communication equipment, such as telephones and intercom
       The state of the main process elements (pumps, valves, switches, analog values etc.) are
represented on synoptic diagrams for each system individually. From a general process overview the
operator can navigate to the details of sub-processes. Alarms are transmitted in parallel to the
synoptic diagrams and to an alarm list display, where the operator has an overview of all the alarms
active in any of the monitored systems. A simple indication of the consumers on the accelerator site
exists on the synoptic diagrams.
      The monitoring applications for each process are independent, so that navigation between
synoptic diagrams of dependent processes is difficult. The alarm list display does neither permit to
group alarms of the same or different processes together, nor to apply any filter criteria. Thus, the
control room operators need a lot of background information to exploit the monitoring information
and to quickly assess failures. Furthermore, complementary information has to be looked-up in
operation instructions, which are not part of the monitoring system.
       The TCR and the PCR share the same data logging tools and the PCR uses some of the TCR
process synoptic diagrams. However, the operators do not have enough background information to
efficiently use the monitoring tools that are technically available in both the control rooms.

The engineering of the monitoring information shall be based on a proper identification of the
production objective, the production processes and sub-processes, see figure 1.
       To meet the main production objective, the different processes of the main systems, such as
electricity, cooling, vacuum, have to be identified. The processes will consist of sub-processes, such
as cooling towers, deminaralized water, which on their turn can be split into operations, such as
pressure control, temperature control. Any of these elements has time dependencies and dependencies
with other sub-processes or operations. The combination of those dependencies determines the
critical paths for any restart operation [2].
       The entire information has to be available for the control room operators in form of
visualization diagrams for the critical paths, process diagrams and alarm list displays; not to forget
the operation instruction, see figure 2.

                                        Main Production Objective

                          Main System Processes          Main System Sub-Processes

                            Process Operations              Sub-Process Operations

                                         Process Dependencies

                                           Time Dependencies

                                                Critical Paths

                             Figure 1 Monitoring Information Identification Process

      In order to have a complete picture of all the influences and constraints the exploitation
personnel and process specialists have to contribute their know-how during the engineering phase [3].
This exercise has to be made for different phases in the production process such as start-up, process
studies, production optimization, physics, maintenance (stand-by) etc.



                                     Dependencies         Critical Paths

                               Instructions        Synoptic        Alarm List
                                                   Diagrams         Displays

                               Figure 2 Representation of Monitoring Information

The work objectives of the PCR and the TCR after a major incident have to be identical: it has to
serve exclusively the main production objective. Major incidents have to be treated from the
managerial point of view as a crisis. Thus, the information concerning faults and failures has to be as
transparent as possible for both the control rooms, as far as the main production process is concerned.
Details of each of the processes have still to be adapted to the particular needs of each control room.
The engineering of the monitoring information has to be optimized for the restart of the main
production process and not for the restart of each individual system process.
      The control room operators have to be enabled to find the best restart strategy based on process
dependencies, process functions and nominal operation values. The following information shall be
available, so that decisions and actions can rapidly be taken:
 - availability of the main production process and the main system processes,
 - state of the critical paths and the dependencies,
 - unavailability of process equipment and causes (fault states),
 - actions to be taken to reestablish the availability of the processes,
 - detailed information on the process equipment to verify the correctness and coherence of process
   information, to verify the correctness of the standard restart procedure and to establish alternative
   procedures, if necessary,
 - trend information of process operations that are part of the critical paths,
 - nominal process values and process limitations.
       Depending on the type of information the representation method and the degree of availability
of this information can be determined. As far as possible this information shall be delivered “in real-
time” via a computerized monitoring system. Navigation to the information shall be orientated to the
control room operator’s tasks. Navigation between trend displays, synoptic diagrams, visualization
diagrams and alarm lists shall thus be seamless.
       Furthermore, the distribution of control actions between an automatic system and the control
room operators has to be documented in a transparent way. Especially when several control rooms are
in charge of different processes, the faults, failures and the advancement of the restart of processes
shall be transparent. Field operators as well as crisis teams have to be kept up to date in the best
possible way, too.

5.         CONCLUSION
The key to the rationalization of process monitoring by control room operators is the focus of
monitoring information engineering to the main production objective. The proper analysis of the
processes and activities needed to keep the availability of the production leads to a task oriented
design of the monitoring tools. In addition the increase of transparency of information that is
exploitable without too much background information, is important to improve communication and
information flow after major breakdowns. The management of such crisis situation will profit from
this and the reduction of down time becomes easier.
       This paper has shown the major principles that shall be applied at the example of the
exploitation of the SPS by the TCR and PCR. However, this analysis is still on going and the final
results will be published later and shall be applied to the engineering of monitoring information and
systems for the LHC.

 [1] F. Dinonis, Audit de l’activité de remise en service des services techniques depuis la salle de
     contrôle TCR, Réf. HP-31/00/019, EDF R&D, 2000
     [2]    D. Galara, J.P. Heenebicq, Process Control engineering Trends, Annual Reviews in Control
            23 (1999) 1-11, Pergamon
     [3]    M. Pillet, L'AMDEC, conférence ASPQ, IUT Annecy 1994

To top