Chris Stearns describes the implementation of effective operations monitoring solutions for process industry facilities.
Many automated industrial plants have now implemented some type of operations monitoring program. However, the effectiveness of these programs—the next logical step after alarm management—can be limited by the use of ad-hoc or standalone tools such as spreadsheet applications to evaluate process variables against operating limits, conduct plant data analysis and perform stewardship reporting.
Today’s operational challenges
For process plant owners, it is important to support control engineers, optimisation engineers and operators who are implementing best practices for operational excellence aimed at meeting the plant’s business and safety goals.
Personnel must monitor a wide range of measurements and key performance indicators (KPIs) from plant and equipment at a production site, as well as maintain the required values of variables to meet operating objectives such as maximum yield, utmost efficiency and minimum emissions.
An operating envelope is a collection of constraints, boundaries and operating limits in an industrial facility that, when exceeded, put the integrity of assets at risk. These limits are typically based on combinations of factors such as process unit capacity, equipment constraints and safety concerns. They can also be implemented for alarm systems and operating targets.
According to the Abnormal Situation Management Consortium (ASM), ensuring operations remain within correct limits is central to avoiding many of the root causes of abnormal situations.
To maximise the life of an asset in an industrial facility, it must be operated according to design parameters and not simply within process safety limits. That means extending operating strategies beyond operator visibility to the entire operations team and all those interacting with the process. Without a comprehensive limit management solution, operators lack the insight needed to run plants within operating envelope boundaries.
Industrial sites typically employ multiple types of process control applications, each of which can be used to independently enter and control respective targets, constraints or limits. Although these applications may relate to the same process measurements, the limits they use are sometimes inconsistent or conflicting. This situation can result in inefficient operation, costly process upsets and unplanned shutdowns.
Various groups within the plant are responsible for maintaining safe operating limit information. As these variables are often system configuration parameters entered by humans, there is the possibility values may fall outside of the safety and compliance envelope. Additionally, some processes have dynamic safe operating limits that are continually changing, which is challenging for plant operators to manage. As these limits are adjusted for safety, reliability and optimisation reasons, staff across the facility must have current and updated exceedence reporting to effectively manage site performance.
Plant owners and operators are under continual pressure to optimise their facilities and processes. This means achieving greater productivity more efficiently with fewer resources. Data about plant performance is key to making smart operational decisions, but in most cases, operators have access only to piecemeal information about their units and processes—examining performance often in a vacuum.
Process industry facilities typically devote considerable resources to rationalising their alarm systems so operators can effectively manage the process and not just respond to alarms throughout the shift. Alarm rationalisation involves reconciling individual alarms against the principles and requirements of the alarm philosophy. It is important that the relevant data for each alarm is documented to support the other stages of the lifecycle. This includes the alarm description, settings, causes of an alarm, consequence of no action, required operator action, response time, consequence rating, and so on.
A properly designed and well functioning alarm system is imperative to operational excellence initiatives, but it is not enough to simply operate within alarm boundaries. Operations managers need to know if units are running in a range that will assure production plans are met while staying within limits, which include (but are not limited to) equipment constraints, economic targets, environmental standards, safety system regulations and advanced process control strategies.
Many automated industrial plants have now implemented some type of operations monitoring program. These programs provide the tools for an operations department to establish and manage engineering limits and constraints, monitor performance to plan and limits, and to follow-up on performance problems.
Operations monitoring is meant to address questions such as:
* Are operating plans being met?
* What are the safety, process, design, reliability and environmental limits, and are these limits in effect consistently?
* If plans or limits are being violated, why?
* How can process performance and unit reliability be improved?
In many cases, operations monitoring programs make use of ad-hoc or standalone tools, such as spreadsheet applications or a combination of e-mail and printed reports, to evaluate process variables against operating limits, conduct plant data analysis and perform stewardship reporting.
Because personal spreadsheets are generally not subject to the same rigid control standards as other IT applications, errors and omissions can occur, impacting the accuracy of information used to develop planning targets and identify environmental constraints. Without a central data repository, different individuals may apply different data as the basis for reporting and decisions. Spreadsheets may also limit access to daily operating information for the rest of the organisation.
In addition, ad-hoc tools linked to plant historians can be a headache for IT to support. Spreadsheets are often inconsistently applied and difficult to keep up to date when the process or historian changes, or when their owner moves to a different job or site. Plus, they may not be well suited to following through on problems once identified. Although an historian itself captures a wealth of vital data, plant optimisation efforts will struggle without feedback from operations to put information regarding process limits, excursions, upsets and other activity in proper context.
Operations monitoring has evolved into an ongoing process employing advanced applications to proactively leverage fewer experts—using better technology—to focus on overall performance, often with the help of external vendors and partners. Today's virtual environment allows the enterprise to monitor each plant in real-time to achieve continuous learning and sustained improvement.
In a typical process plant, operations monitoring can be used to monitor measured and calculated process tags against operating, safety and corrosion limits, as well as other indications of reliability. Such engineering limits typically don’t change often and may have safety, environmental, or maintenance implications if they are violated.
Another common use for operations monitoring is to evaluate process data and KPIs against planning limits. Planners frequently adjust operating ranges when production strategies, product grades, or feeds change in a process unit. These limits usually change frequently and can have economic implications. Violating planning limits can mean reduced product quality, the wrong production rate, missed shipments to customers, etc.
Operations monitoring helps automate tracking actual process performance every shift. Many plants benefit from improving how routine issues are handled, before they grow into problems. For example:
* A de-salter in a crude unit is designed to operate at up to 350 degrees, but the corrosion rate increases noticeably when operated above 300 degrees. Systematically tracking excursions above 300 degrees and fixing the root causes of the deviations will extend the life of the equipment.
* A reboiler gradually fouls, reducing heat transfer and eventually limiting production. An anti-foulant is available but expensive, and the ideal injection rate is poorly understood. Monitoring the energy efficiency can help determine when an operator should look at the injection rate.
* A purge rate needs to be temporarily increased to remove impurities from a column. Monitoring the purge rate and the yield helps ensure the purge valve will be reset at the right time, which will prevent an undesirable loss of production.
Plant operations departments are re-thinking their approach to operational excellence in order to realise the maximum benefit from ongoing technology developments. Instead of simply managing the effects of operating outside established boundaries, they are seeking to expose the operating envelope to all appropriate plant stakeholders and ensure it is well understood across operations and related organisations.
Technology providers have historically offered operations monitoring applications as part of an overall operations management portfolio. Too often, however, this technology required users to accommodate a large hardware footprint, complex and costly server infrastructure and licensing, and extensive programming effort. This situation drove up the cost of operations monitoring programs and forced plant engineers to rely on less complicated 'homegrown' monitoring techniques.
Alternative to spreadsheets
The current breed of operations monitoring solutions helps industrial organisations transition from labor-intensive, legacy plant performance spreadsheets to an automated and standardised system for facility-wide data collection, analysis, and reporting. This allows the enterprise to move beyond disjointed spreadsheets and difficult scripting languages, non-standard and inefficient processes, and inconsistent calculations requiring significant manual input.
New software tools are intended to systematically monitor process plant performance data and summarise deviations from the operating plan. These tools are well suited for tracking operating performance against targets and highlighting problem areas. They are designed to fit into existing work processes and help operations teams institutionalise those procedures.
Plant data sources
Today’s operations monitoring infrastructure may reside at level 3 or 4 of the plant network hierarchy, utilising industry-standard OPC data access to establish connections for retrieving real-time data from historians and various other data sources. Monitoring solutions employing browser-based displays can provide plant-wide access to monitoring results.
Operations monitoring to enable better decision-making is a growing necessity based on current plant operational demands. Even experienced operators may not know the best operating range for throughput or may fail to realise the consequences of operating outside of targets. Furthermore, operations monitoring helps industrial facilities move to the next level of operational excellence by leveraging the inherent benefits of alarm management initiatives.
Operations monitoring benefits come from a variety of sources, including:
* Reduced number and severity of incidents;
* Reduced operating and maintenance costs through increased asset reliability;
* Better safety and environmental compliance;
* Increased operating margins through better fidelity to the operating plan.
An effective operations monitoring solution delivers these benefits by supporting a structured, systematic monitoring program. Engineers can utilise the latest software technology to monitor process values and record anything outside of the normal range, as well as scan, filter and consider these deviations in context. Operators and engineers can then assign reasons and comments to the deviations, while managers assess actual performance and set priorities based on associated reports.
In addition, IT professionals can take advantage of operations monitoring as part of an integrated plant information system allowing them to minimise administration costs, access process data from plant historians, and reduce capital and implementation costs with a common architecture across business applications.
Improving operational performance and reliability requires a team effort by operators, engineers, and various other specialists within the plant. These people will benefit from operations monitoring solutions that build on alarm management efforts and improve their ability to monitor the performance of processes and operating assets to make profitable operational decisions for both the short and long term.
Chris Stearns is with Honeywell Process Solutions.