Advanced process control in the cloud

Engineering and IT Insight: When considering which applications to move to the cloud to reduce costs, consider moving advanced process control (APC) model building and APC model validation tools—usually lightly used and usually not mission critical. Learn three types of cloud-based services.

02/22/2013


Controls and IT Integration, Control EngineeringUsing the “cloud” is not something that is normally considered for process control, but the situation is slightly different when using advanced process control (APC). APC uses predictive models of a process to generate setpoints and control moves that are better than those generated by typical PID algorithms. APC is usually used where responses are nonlinear or even discontinuous and where multiple process values can be required to generate the correct control move. Implementing APC usually involves three steps: building the APC model, validating the APC model using data from a running system, and executing the model to generate setpoints and control moves.

Model execution is implemented in real-time control systems and is not suitable for moving to the cloud. Cloud response times are variable and accessibility is problematic, so this is not an environment for any real-time control. Model execution is often performed within a distributed control system (DCS), programmable logic controller (PLC), or an attached PC. Most DCS vendors provide APC elements in their control systems, but these are usually just model execution blocks that assume you have already created and validated the APC model.

APC model building, however, is different. If you are lucky to have very smart engineers, you may be able to build an APC model from first principles. First principles allow you to use knowledge of your processes to construct mathematical models, accounting for all required product quality attributes and all possible process parameters needed to achieve the attribute targets. Most people are not lucky enough to have the knowledge or resources to build first principle models. Even something as simple as a blending operation may require so much physical modeling as to make it impractical for general use. In these cases, you can use pattern analysis tools to discover the mathematical relationships between the process parameters and the quality attributes to develop an empirical model.

Pattern analysis tools take a lot of data and use a lot of computing time, but they are only occasionally needed, usually when the process changes or equipment is changed. Lots of data, lots of computing power, and only occasional needs are the sweet spot for a cloud-based solution. A purchased pattern analysis system would require servers and databases that would normally sit unused, tying up capital and consuming IT support resources.

Three types of cloud-based services

A cloud-based solution may be a system as a service (SAAS), in which the cloud service vendor provides an operating system environment, such as a Microsoft Windows Server, that runs your application. Another option is a platform as a service (PAAS), in which the cloud service vendor provides a bare-bones machine that you load with your operating system and application. A third option is an application as a service (AAAS), in which the cloud service vendor provides a full application and the environment. Any one of these options should provide a lower cost solution than locally hosting and maintaining an application that you will use only rarely. Model building for advanced process control that produces an empirically derived model is a good candidate for a cloud-based solution.

Many APC projects fail because the models are not maintained, and over time they fail to accurately reflect the real behavior of the system. There should be, but often is not, a schedule for regular validation of the model. Model validation compares the expected results from the model with the actual results from the system. When validation is not performed, small changes in material properties, or equipment changes due to aging or replacement of equipment, will eventually result in an invalid model. Invalid models will generate suboptimal or even wrong control moves for the process. Validation can be performed on every production run, but if the model was built from statistical data, then validation should be performed when only enough data is available to smooth out random variations.

Model validation is also a process that requires a lot of data and computing power, yet may be only occasionally run. APC model validation is also a good candidate for a cloud-based solution. One of the advantages of a cloud solution is the general ability to scale up the database space available and computing power available for short periods of time. Online retailers take advantage of this by scaling up during the main shopping seasons, then downsizing and saving money during off-seasons. You can use these same scale-up features to validate your APC models when the cloud vendor’s systems are being least used and are available at reduced rates.

When considering which applications you can move to the cloud to reduce your capital and support costs, consider moving your APC model building and APC model validation tools. These tools are usually lightly used and are usually not mission critical applications. So when your CIO says, “Everything to the cloud,” your manufacturing IT team will have a plan in place.

 

- Dennis Brandl is president of BR&L Consulting in Cary, N.C., www.brlconsulting.com. His firm focuses on manufacturing IT. Contact him at dbrandl(at)brlconsulting.com. Edited by Mark T. Hoske, content manager, CFE Media, Control Engineering and Plant Engineering, mhoske(at)cfemedia.com.

ONLINE extra - This posted version contains more information than the print / digital edition from March 2013 Control Engineering, including explanation of three types of cloud-based services



No comments
The Top Plant program honors outstanding manufacturing facilities in North America. View the 2013 Top Plant.
The Product of the Year program recognizes products newly released in the manufacturing industries.
The Leaders Under 40 program features outstanding young people who are making a difference in manufacturing. View the 2013 Leaders here.
The new control room: It's got all the bells and whistles - and alarms, too; Remote maintenance; Specifying VFDs
2014 forecast issue: To serve and to manufacture - Veterans will bring skill and discipline to the plant floor if we can find a way to get them there.
2013 Top Plant: Lincoln Electric Company, Cleveland, Ohio
Case Study Database

Case Study Database

Get more exposure for your case study by uploading it to the Plant Engineering case study database, where end-users can identify relevant solutions and explore what the experts are doing to effectively implement a variety of technology and productivity related projects.

These case studies provide examples of how knowledgeable solution providers have used technology, processes and people to create effective and successful implementations in real-world situations. Case studies can be completed by filling out a simple online form where you can outline the project title, abstract, and full story in 1500 words or less; upload photos, videos and a logo.

Click here to visit the Case Study Database and upload your case study.

Why manufacturers need to see energy in a different light: Current approaches to energy management yield quick savings, but leave plant managers searching for ways of improving on those early gains.
Electric motor power measurement and analysis: Understand the basics to drive greater efficiency; Selecting the right control chart; Linear position sensors gain acceptance
Protecting standby generators for mission critical facilities; Selecting energy-efficient transformers; Integrating power monitoring systems; Mitigating harmonics in electrical systems

Annual Salary Survey

Participate in the 2013 Salary Survey

In a year when manufacturing continued to lead the economic rebound, it makes sense that plant manager bonuses rebounded. Plant Engineering’s annual Salary Survey shows both wages and bonuses rose in 2012 after a retreat the year before.

Average salary across all job titles for plant floor management rose 3.5% to $95,446, and bonus compensation jumped to $15,162, a 4.2% increase from the 2010 level and double the 2011 total, which showed a sharp drop in bonus.

2012 Salary Survey Analysis

2012 Salary Survey Results

Maintenance and reliability tips and best practices from the maintenance and reliability coaches at Allied Reliability Group.
The One Voice for Manufacturing blog reports on federal public policy issues impacting the manufacturing sector. One Voice is a joint effort by the National Tooling and Machining...
The Society for Maintenance and Reliability Professionals an organization devoted...
Join this ongoing discussion of machine guarding topics, including solutions assessments, regulatory compliance, gap analysis...
IMS Research, recently acquired by IHS Inc., is a leading independent supplier of market research and consultancy to the global electronics industry.
Maintenance is not optional in manufacturing. It’s a profit center, driving productivity and uptime while reducing overall repair costs.
The Lachance on CMMS blog is about current maintenance topics. Blogger Paul Lachance is president and chief technology officer for Smartware Group.