A major European oil company upgraded their refinery as part of a strategy to enhance plant profitability, increase throughput capacity, upgrade product yields and improve on-stream reliability. One of the upgrades performed at the refinery was the integration of an Automated Tank Gauging system (ATG). Traditionally, the operation collected data manually from their farm of tanks.
Twice a day a group of people was deployed to visit every tank, climb to the top, measure tank levels using a stick, record the information and then enter it into the refinery’s Tank Information System (TIS). Climbing the tanks and exposing operators to potentially inclement weather posed considerable safety concerns. Then, there is the logistics of coordinating data collection and the time delay. The data collected was not accurate because temperature can significantly impact the volume of a tank.
Consider, for example, that a change in temperature requires corrected level values that account for the thermal expansion of the tank walls. A manual reading out by 1 mm and multiplied by 74 tanks can cause a significant discrepancy between actual and recorded inventory levels. Also consider the impact of a tank that leaks. This causes serious repercussions not just for the industry, but for the environment as well. In addition, the time consuming manual gauging process meant information was not available in real-time to plant operators and key decision makers. Last, manual data entry errors would sometimes mean that inventory was incorrectly calculated.
The ATG system enabled terminal operators to have easy real-time access to all measured and calculated parameters, such as level, temperature, flow, density, gross and net standard volumes. The continuous accurate monitoring of tanks makes it possible for the company to deal with tank level uncertainties representing significant potential losses and maintain good overall inventory management. How did they do it?
The foundation
Part of the company’s strategy involved upgrading their plant’s control system with a Yokogawa Centum Control System (CS). The old Bailey control system was retained because it had proprietary software tools still used by operators. The company’s control engineer explains, “The ATG project installed field instrumentation to the tanks spread across the Refinery. Field IO (input/output) consisted of Foundation Fieldbus fed into a Yokogawa Centum CS. However, for operators to access the data through the Bailey-dependent Order Management System (OMS) and TIS, they needed to take the accurate real-time field data captured in the new Yokogawa Centum CS and transfer it to the Bailey DCS.”
Standards
The oil company needed to interface between two disparate control systems. One option was to write proprietary interfaces; however this would require weeks of custom coding, delay implementation and drive up costs. Another would be to use OPC standards based connectivity technology. OPC is a technology used to transfer data among devices and applications which, compared to the proprietary method, is an inexpensive, easy to use, and commercially off-the-shelf solution. Based on industrial connectivity standards, it is quick to integrate and provides a stable, secure and reliable environment.
For these reasons, the oil company opted to integrate the OPC-standards-based connectivity solution.
The refinery faced three main integration issues:
- The Bailey Control System was old and not equipped with an OPC
interface.
- They needed a way to facilitate the transfer of tag data from the Yokogawa Centum CS to the Bailey system
- The Yokogawa Centum CS and the Bailey system were on different Windows domains, which complicates Windows DCOM configuration.
To enable the Bailey system with OPC, The control engineer opted to integrate the Matrikon OPC Server for Bailey. "OPC gives us the flexibility to expand; it enables us to add data from any other OPC-compliant system”, commented the integrator. “In this case it was the Yokogawa Centum CS, but it can be from other control systems as well. So we can extend the life of our legacy applications without sacrificing the new features of the latest control systems.”
Integrating the new server meant tackling a difficult interface problem. Since the tank gauging hardware was interfaced with the Yokogawa but the visualization tools were connected to the Bailey system, they needed a way to transfer data from one control system to the other (Figure 1). Matrikon OPC’s Data Manager is an off-the-shelf solution that transfers data between control systems. But because the control systems were on different Windows domains, there were DCOM issues.