TNK-BP's refineries are in the process of upgrading instrumentation and control systems, as part of the five-year strategy. For technical and commercial reasons this investment program must be made at each refinery in stages. In order to fulfil the ultimate aim of automation, which for us is to deliver optimum value to our refining assets, we have to determine which technologies to invest in, at which refineries and on which processes to implement them, how to implement and support these (i.e. manpower issues) and when to perform all the above mentioned actions.
There are five refineries in TNK-BP, ranging from small and simple to large and complex. There has been some prior investment in modern control systems, but there is a significant presence of obsolete equipment (e.g. pneumatic, rather than electronic, field instrumentation) and there are several problems associated with this.
First, the old equipment does not allow "tight" control, which makes operating a process at its performance limits difficult; it also prevents the use of modern advanced control technology. Second, due to the age of the equipment and the scarcity of spare parts, the low availability of the controls can seriously compromise the safe and reliable operation of a whole process. Third, these reliability problems mean that a great deal of maintenance attention and expense is required to keep them running. Last, and most important, there are regulatory issues which must be addressed especially regarding the cover provided by our emergency shut-down systems.
Plan Development
In 2005, the Automation Expert Group, which comprises chief metrologists from each refinery, developed a five-year strategy to address the problems detailed above and so remove the obstacles to delivering optimum value by the implementing of advanced technology. The model used as the basis for the strategy resembles a pyramid, and is shown in Fig. 1 – in relation to this the following general objectives were developed:
Upgrade field instrumentation to modern, electronic type with "smart" potential (e.g. self-diagnosis, fault reporting, etc.);
Replace obsolete control and emergency shutdown systems with modern variants;
Implement, at relevant refineries, gasoline and fuel oil blending optimization;
Implement advanced control technology on key units (e.g. FCCs at Ryazan and LINIK, AVTs everywhere), supported with performance analyzers;
Provision of appropriate interfaces to information systems (e.g. PI, LIMS);
Support for Company-wide policies, e.g. mass metering, furnace safety, etc.;
Development of in-house expertise in these identified technology areas;
Change appropriate business processes to allow full value recovery from new technologies.
Practical and commercial considerations mean that the timing of the implementation often has to be coincident with the turnaround schedule of a particular refinery. In addition, the value-adding nature of the plan means that a refinery may progress up the pyramid in some processes ahead of others. Also noteworthy is the long-term focus on developing our own specialists, to implement and support these technologies in-house rather than relying on external contractors.
The most important objective in the whole list, however, is the last one – the need to change business processes is vital to successful implementing of virtually all these technologies. For example, to achieve the potential value-added in one of the proposed gasoline blending optimization projects we will need to effect change in at least four major business processes; production planning, the blending process itself, quality control plus the project process.
Results So Far
The Company's refineries are steadily implementing the plan as detailed above, and in the next two to three years some major reinstrumentation projects are planned.
The re-instrumentation projects include the installation of "smart" instrumentation – that is, field devices (for example, measuring and/or controlling flow rate, pressure, etc.) that have some built-in intelligence for diagnosis of faults. At Saratov Refinery, the Visbreaker unit has "smart" control valves which can report problems early, so supporting "predict and prevent" maintenance. This is now the standard approach across TNK-BP refining for new instrumentation.
The more advanced automation technologies, which are widespread in Western Europe and North America, are also beginning to appear in our refineries. For example, in 2005 Ryazan Refinery implemented the first advanced control application of its kind in Russia. And recently Lisichansk Refinery has commissioned a fuel oil blender, using a viscosity analyser in the pipeline for feedback control, which has dramatically reduced excess quality (or "giveaway") cost in the product.
As a first step toward developing our own in-house expertise in these new technologies, we have started to train our automation engineers from each refinery on new courses within the corporate training program. The Automation Expert Group has designed a skills matrix that is intended to address those areas where we believe we will experience a skills shortfall in the future, and these courses are therefore a critical part of the whole upgrading strategy.
The Future
By the end of 2010, all our refineries should have virtually full control system cover with adequate emergency shut-down capability. A significant proportion of the field instrumentation of our process units will be "smart", and will transmit reliability data automatically to our maintenance staff. Our major products will be blended using advanced analyser and optimization technology. Our major upgrading and separation process units will feature advanced control and analyzer technology.
The value added by this upgrading of automation includes improvements to HSE performance, energy consumption, reduced product quality giveaway costs and delivering better equipment availability. And all of this, of course, will depend on our migrating to appropriate business processes. Without this crucial step, our ability to deliver long-term, sustainable value from these investments will be very limited.
