Nov-2018

An overview of Honeywell’s layered optimisation solution

Advanced process control (APC) and real-time optimisation (RTO) tools have become necessary technologies for today’s process operating companies to compete and maintain profitable operations.

Honeywell

Article Summary

On a wide range of industrial automation projects, the implementation of these solutions can result in improved process stability, increased throughput and yield, decreased operating costs, improved product quality, and increased operating flexibility.

APC/RTO also forms the foundation for on-line optimisation, which typically increases advanced control benefits with accelerated project paybacks. Customers are now starting to look at a bigger prize; full gate to gate (plant-wide) optimisation, coordinating the Planning/Scheduling and APC/RTO layers.

The challenge for end users is to achieve a level of optimisation providing the best possible return on investment (ROI). This ROI includes not only the cost of implementing a project and the expected optimisation benefits, but also the cost of maintaining the system with available resources.

Advanced control and real-time optimisation
Honeywell’s best-in-class Profit Optimization Suite enables full plant-wide optimisation and control be performed simultaneously. With a layered optimisation strategy, the appropriate combination of technologies can be selected to solve specific problems instead of using the same technology for each problem. This solution achieves significant advantages with reduced risk, low lifecycle maintenance costs and sustained benefits over the long term.

Advanced control and real-time optimisation technology have become necessities for today’s process operating companies to compete and maintain profitable operations. Honeywell has continuously demonstrated the benefits of advanced process control (APC) technology in improving profitability through improved process stability, increased throughput and yield, decreased operating costs, improved product quality, and increased operating flexibility.

Another benefit of advanced process control is that it forms the foundation for on-line optimisation and gate to gate optimisation, which typically adds an additional 30 percent of the advanced control benefits with project paybacks often in less than six months.

Table 1 shows typical APC and optimisation benefits that Honeywell customers have experienced.

In the past, many companies underestimated the cost and complexity of implementing and maintaining real-time optimisation systems. Traditionally, companies involved in optimisation projects adopted a “top down” approach using detailed non-linear steady-state process models. These traditional, steady-state optimisers sent targets to multivariable controllers, which drove the plant toward optimum operation. However, unless plant personnel were committed to understanding and maintaining such systems, most of them went out of service within the first year.

Although there is a need for optimisation based on detailed modelling in some cases, a large portion of the optimisation problems in the process industries today can be addressed directly by leveraging MPC models already invested in and less extensive, targeted modelling that focuses on key units exhibiting significant non-linearity.

Detailed process modelling has proven useful to operating companies as a tool for process design, operator training, and operations planning, monitoring and optimisation. Detailed modelling is essential for performing tests which are not feasible on the real plant. For example, a model may be used to evaluate different reactor catalysts or a process retrofit. For use in on-line optimisation, however, the trade-off between process benefit versus added cost and ongoing maintenance effort is very real and must be considered carefully.

In many cases, a less extensive modelling approach can, and should, be used for on-line optimisation because process measurements are available to correct for model inaccuracies on an ongoing basis.

The challenge for the process industries is to provide a level of optimisation which realises the best return on investment (ROI) for the customer. This ROI includes not only the cost of implementing a project and the expected optimisation benefits, but also the cost of maintaining the system with the customer’s available resources.

To meet this challenge, Honeywell has developed a layered approach to optimisation including:
• Controller-based optimisation (ProfitTM Controller)
• Distributed dynamic optimisation (ProfitTM Optimizer)
• Non-linear gain updating (UniSim Optimization Suite with  ProfitTM Bridge)
• Gate to gate optimisation (ProfitTM Executive)

Honeywell is well positioned to provide the level of optimisation needed for a particular application, from robust multivariable control and dynamic optimisation to detailed first-principles modelling and non-linear optimisation. In addition, our process and optimisation consultants can analyse the process to determine the most appropriate technology.

This paper describes the components of our layered optimisation solution, the advantages and benefits of this approach, as well as methods for determining optimisation benefits of an application.

Layered optimisation solution
As shown in the figure below, the various advanced control and optimisation technologies within Honeywell’s Profit Optimization Suite are dependent on other enabling technologies such as advanced regulatory control and inferential modelling tools. In addition, optimisation occurs at several levels in the control and optimisation hierarchy, from unit control and optimisation to multi-unit optimisation to plant-wide optimisation. These levels represent increasing optimisation scope.

Honeywell’s layered optimisation solution also consists of different optimisation technology layers, from linear dynamic optimisation, to non- linear steady-state optimisation, to non-linear dynamic optimisation.

Honeywell’s optimisation technology can be divided into four categories. These categories are presented in the table below together with the relevant Honeywell products.

Overview of optimisation technologies
Honeywell’s industry-leading APC technologies, including Profit Controller and Profit Optimizer, contain embedded optimisation to enable optimisation and control to be performed simultaneously. The advantage of a layered optimisation strategy is that the appropriate combination of technologies can be selected to solve specific customer problems, rather than using the same technology for each problem.

Figure 2 shows various optimisation configurations that Honeywell has successfully implemented for customers.
The appropriate optimisation solution for a specific customer application is based on the following criteria:
• Expected ROI:
• Maintenance costs (e.g., rigorous models versus data models)
• Potential benefits (e.g., tangible and intangible)
•    Complexity of solution (affects application uptime)
• Process changes (magnitude, frequency, type)
• Process flexibility (degrees of freedom to adapt to changes)
• Level of customer technical expertise (less expertise may require a less complex solution)
• Implementation requirements
• Maintenance requirements