Back To Top
Digital Transformation

Enabling Advanced Automation for Pipeline Operations - Part 1

November 17, 2022

Author: Kamaal Mahmud, P.Eng., Jayson McAllister, M.Sc., P.Eng.

Introduction
Pipeline infrastructure and supporting operations are responsible for transporting critical energy supplies throughout the globe. Recently, there have been frequent political blockades which have resulted in increased difficulties, costs, and time to build out new infrastructures. These events have resulted in a new look at how existing pipeline infrastructure can be improved and optimized with the use of advanced automation. This post focuses on the energy industry specifically but has implications for any industry with pipeline operations.

What is the challenge, and why does it matter?
Due to the geographic reach of pipeline infrastructure and assets, the control of pipelines is generally handled by Programmable Logic Controllers (PLCs), which control assets located in close proximity. Many PLCs are typically required for larger pipelines. The PLCs monitor and control local processes, mostly irrespective of how other assets upstream or downstream are operating. Due to the interconnectivity of the pipeline process, this siloed control method can result in large inefficiencies in transporting product from one location to another as the local PLCs are not working together.
 
In order to help solve this issue, pipeline operators deploy a Supervisory Control and Data Acquisition (SCADA) system to consolidate information from all the local islands of automation. These isolated automation data sources operate separately from the rest of the facility and feed their information to a central operations center to allow for basic overarching control of the entire system. However, SCADA systems are not well suited to the implementation of advanced automation. Advanced Automation that is described here includes control applications that use Model Predictive Control (MPC), Scheduling Optimization, Real-time Optimization (RTO), Soft Sensors, Cross-PLC Procedure Automation or cross-PLC regulatory control.
 
The Process Industry Automation Hierarchy standard, Figure 1, describes the different layers of automation that should exist for any industrial control system. The Process Industry Automation Hierarchy takes on a pyramid form where information gets aggregated as it moves from the bottom (instrument/sensor level) to the top. Each layer enables the layer above and allows for progressively more complex and advanced automation aimed at achieving process control objectives and ultimately delivering improved process performance, efficiency, reliability, and safety.

 

Figure 1 Automation Hierarchy
Figure 1 Automation Hierarchy
The instrumentation layer of the pyramid encompasses the physical devices that provide measurements of the process and give the ability to control the process. Instrumentation is the largest layer of the hierarchy because no automation can be done without access to these devices. These devices must remain in good working condition to enable the upper layers of the automation hierarchy to function correctly.
 
The second layer of the automation hierarchy is the Basic Process Control System (BPCS). BPCS refers to the systems that are responsible for controlling the basic portions of the process automatically. An example of basic control includes starting a pump, opening a valve, or controlling a flow measurement to a setpoint by manipulating a valve open percent through a computer system such as a PLC. Automated logic, such as motor control, PID, and ratio control loops, exist at this level and makeup 90% or more of the controllers in a typical control system.
 
Procedure Automation is the third layer of the automation hierarchy. This layer is responsible for simplifying the operational workload, reducing time spent performing monotonous tasks, improving equipment reliability, and streamlining the operation of automation infrastructure.
 
The dynamic optimization layer of the automation hierarchy is the first layer in which mathematical optimization algorithms are used. This layer refers to applications such as model predictive control and steady state optimization. These applications are typically responsible for larger multivariate processes, processes that require constraint handling, processes that exhibit large deadtimes, and/or simple economic optimizations.
 
The top layer of the automation hierarchy is Planning and Scheduling. Planning and Scheduling can occur at different time frames such as planning deliveries for each month of the year or planning deliveries for each hour of the day. This layer incorporates pipeline-related applications such as optimizing short-term schedules for product-gathering pipeline systems.
 
The first three layers can be implemented effectively in each local PLC for any application where the PLC has access to all the information it needs (i.e. assets that are physically located in close proximity to the PLC). SCADA can help the operator change SPs, open/close valves, and perform basic operations. This is where pipeline operators are situated due to the challenges of engineering effort, lack of APC knowledge/resources, and/or the inability of the existing systems to efficiently enable advanced automation applications. 
Modern Distributed Control Systems, such as the Emerson DeltaV DCS, are well suited to provide a platform for implementing all levels of automation hierarchy.
Kamaal Mahmud & Jayson McAllister
What is the Solution?
Modern Distributed Control Systems, such as the Emerson DeltaV DCS, are well suited to provide a platform for implementing all levels of the automation hierarchy. DeltaV comes equipped with standard IEC61131-3 control language function blocks for analog and discrete control. The DeltaV control library also has capabilities to easily configure operator-independent time/event variant actions through sequential function charts, state transition diagrams, sequencers, and the cause/effect matrix. In addition, function blocks with Structured Text allow for building complex logic expressions using conditional and iterative structures, and a wide array of algebraic and trigonometric functions and operators. Custom function blocks and control module classes/templates may also be created for quicker deployment of control strategies that can be applied to many instances. DeltaV also has built-in advanced control function blocks like Model Predictive Control with LP Optimizer, Neural Network, Fuzzy Logic Control among others.
 
In addition to the extensive control libraries, DeltaV supports various industrial network communication protocols like OPC DA, OPC UA, Modbus TCP, Ethernet-IP, Profibus, Foundation Fieldbus, and HART. 
 
For integrating with existing pipeline operations, the most logical location is to connect the DCS to the SCADA system. The DeltaV system sits on top of the SCADA system and has access to information from all local PLCs. Spartan Controls recently connected the Emerson DeltaV DCS System to communicate with a SCADA system which enables the implementation of Advanced Automation at a large pipeline customer.
 
The DeltaV system architecture was configured to ensure that redundancies in the existing system design were maintained. Communication between DeltaV and SCADA is accomplished through redundant OPC DA communication networks with the ability to operate from both the primary and backup operations control centers.
 
There are future plans to upgrade to the latest OPC UA communication standard, as well as deploy DeltaV Mobile which will allow operations and engineering teams to view real-time process data, trends, and notifications securely from their mobile devices.
Figure 2 DeltaV-SCADA System Architecture
Figure 2 DeltaV-SCADA System Architecture
An Advanced Automation Audit was completed to uncover opportunities for advanced automation to improve the pipeline operations. To perform the audit, Spartan Controls’ APC engineers reviewed pipeline operations with panel operators, engineers, and other stakeholders from the customer over the span of several weeks. The interviews resulted in a 98-page report that uncovered more than 100 opportunities for process control improvements (potential cumulative value of $12 M per year) which are now enabled through the use of DeltaV. The opportunities spanned the entire automation pyramid presented above and were grouped to be implemented over several phases. Opportunities were classified based on relative payback versus implementation effort. A few opportunities implemented by Spartan Controls APC team during the first phase are below: 
 
  1. A Reid Vapour Pressure Kalman Filter which improved the utility surrounding an existing analyzer measurement.
  2. A Line Scraper Position Estimation soft sensor to track position as it travels through the pipeline.
  3. Procedure automation application for automating pipeline start-up, shutdown, and product switches.
  4. Procedure automation for automating shutdown and restart of assets as line scrapers pass by. Line scraper position estimate from the previous project was used to enable this automation.
  5. Model Predictive Control application to maximize butane injection for a pipeline, while respecting crude specification constraints for density and RVP. The Kalman Filtered RVP from the previous project was used to provide much improved control performance.
Advanced automation is the natural next step in the evolution of pipeline operations. Reducing operator workloads, improving process reliability, pushing constraints to maximize profits, and coordinating siloed assets will help streamline pipeline operators’ activities and drive improved bottom lines. DeltaV can be interfaced into any existing SCADA system via OPC and facilitate easy implementation and maintenance of advanced automation solutions over the entire life cycle of the assets.

Further details on the automation projects will be shared in Part 2 of this blog.
Advanced Process Control Engineer
Kamaal Mahmud
Advanced Process Control Engineer
Advanced Process Control Engineer
Jayson McAllister
Advanced Process Control Engineer

Related Stories