Real-Time Data Analytics to Improve Operational Support
Peter Argus meets with Chuck Smith, the process control engineer, and Tim Olsen, the production manager, to develop strategies that enhance equipment performance, process control monitoring, and energy management and create production improvement strategies. Peter explains that integrating these new digital operational strategies enhance support systems. Chuck shares a diagram (Figure 6.1) presenting real-time operations and enabled support systems using the EIDI, which he calls the "business objectives pyramid."
"This is a good place to start the discussion. A system of record, storing realtime data and events, can serve many functions in our business. As we have seen, transforming data into operational insights for reducing plant problems is necessary. These hidden losses are there. We need to reduce abnormal situations to increase overall production effectiveness," acknowledged Tim.
Increasing the scope of industrial plant for process control and operational intelligence.
Business Objectives Pyramid
Proper application of today's technologies enable the process industries to develop competence centers to manage their industrial complexes, integrating operational and business objectives.
The Left Side of the Pyramid
The left side of the pyramid in Figure 6.1 shows traditional engineering control levels fed by sensors and data collected from the processes. You may think of it as operations technology or ОТ. At the lowest level is the instrumentation level, which consists of devices for acquiring data from sensors, field displays, and hardware safety interlocks for ensuring safe emergency shutdowns. The instrumentation level sends the data to the regulatory control level, which consists of traditional control hardware such as distributed control systems (DCSs) and programmable logic controllers (PLCs). This level integrates real-time data for the regulatory control level. It is one of the most important levels because it has to be extremely robust and responsive for industrial process continuity and operational safety.
Layer 1: Regulatory Controls
The regulatory controls maintain process variables at their prescribed set- point and stabilize variations caused by local disturbances occurring at a timescale of seconds to minutes. Various causes originate the disturbances, such as weather conditions, changes in raw material characteristics, and start-up and shutdown at other refinery sections of the supply chain. In addition, this layer allows the operator to take manual control of the plant in cases of emergency.
The regulatory control layer ensures safe operation, collects data at the original resolution from the sensors, and provides tools to configure process displays and control actions. The regulatory control layer transmits the streaming plant data to a dedicated industrial historian, akin to the black box of an airplane. The data stream feeds the layer on the right side for enhanced equipment condition-based maintenance (CBM) and control improvements. As such, the operational intelligence reuses the real-time data with the proper tools for time-series analysis.
Layer 2: Multivariate Controls
Layer 2 evolved because of advances in control algorithms, hardware, and computing capabilities. In a typical processing plant, the problems are generally classified as multivariable, with many control interactions caused by nonlinearities of the process, process equipment constraints, and unknown process disturbances. Because of the possible interactions among the variables, all control moves must be carefully coordinated. The control actions are taken to accommodate longer duration disturbances, usually minutes. This is also caused by slow processing times of online process sensors or instream process analyzers.
Proclndustries' refinery units are complex, presenting unconventional process dynamics. The process dynamic requires streaming data to develop process models for control design and maintenance.
Layer 3: Coordination Controls
The process coordination controls in layer 3 of the pyramid integrate all the process controls, quality process controls, constraint controls, and the environmental and safety controls for each process unit. These process controls require layer 3 right-side support to analyze the operational data and identify process and equipment constraints.
Layer 4: Supply Chain Optimization
The plant supply chain optimization is implemented to balance overall constraints to find optimal steady-state operating conditions of the plant, based on current production requirements and factors such as raw materials, energy and consumable costs, and production demand. Plant coordination activities are related to planning and scheduling activities reported on the right-hand side of the pyramid. "Imagine if we could improve our daily planning and our overall process coordination," said Chuck. "Monica and Peter showed us how real-time data can be used to detect opportunities for reducing minor losses in the refinery." (See Chapter 3.)
To remain competitive, the team must be vigilant about proactively maintaining equipment to achieve optimum availability and performance levels. Reusing data from the industrial data infrastructure is the most cost-effective way of achieving high productivity and performance levels. Ideally, operations and maintenance teams must collaborate, using the same data to keep the equipment online 24/7 for optimal production levels. Using the capabilities of a modern industrial data infrastructure, maintenance personnel can implement real-time condition-based and predictive maintenance strategies. Real-time analytics trigger alerts to troubleshoot problems prior to catastrophic failures.