Agile Plants, Agile Engineers

No one can deny that the world is getting ever smaller. In business, communication technology substitutes for travel whenever possible, reducing apparent distances to almost zero. E-mail, teleconferencing, and the Internet can instantly bring team members from around the world into the same "room" for planning and other "face-to-face" discussions at minimal cost.

06/01/2004


AT A GLANCE

 

  • Customer demands

  • Software-configurable hardware

  • Agile plants: Two versions

  • Multivariable control


No one can deny that the world is getting ever smaller. In business, communication technology substitutes for travel whenever possible, reducing apparent distances to almost zero. E-mail, teleconferencing, and the Internet can instantly bring team members from around the world into the same "room" for planning and other "face-to-face" discussions at minimal cost.


Goods and materials, however, do not permit such luxury. There is no alternative to moving tangible items from point of manufacture to point of use. Large facilities can take advantage of economies of scale and reduce wide demand swings by carefully scheduling high-quantity customer jobs that peak at different times. On the other hand, transportation costs, the logistics of international shipping, time lag from one end of the supply chain to the other amid rapidly changing needs at both ends, and possible misunderstandings between suppliers and customers make the challenges of large, centralized manufacturing difficult and complex.

While large-scale plants can be more efficient at making a lot of one or a few kinds of products, smaller ones can be more nimble in producing smaller or more customized orders. Bringing production operations closer to end-users involves building "boutique" plants that deliver precisely what is needed in a small geographic area without incurring throughput uncertainties and exorbitant shipping costs. In the extreme case, such plants serve a single customer. These designer plants must be smaller and more flexible than their more traditional siblings.

The idea of bringing manufacturing directly to customers is far from new. Chemical companies requiring nitrogen (as an inert gas for hydrogen-reduction reactions, for example) have for decades avoided the high cost of buying gas shipped in cylinders or tank trucks by having the supplier construct small unattended plants called nitrogen generators adjacent to their own facilities to furnish gas through pipelines in a continuous stream. High-volume plants can feed high-volume users in the same way.

Affect on implementation

Boutique plants require engineers to think differently about implementing control systems, according to Rahul Kulkarni, data acquisition and control manager at National Instruments. He contends that engineers want completely software-configurable hardware with built-in intelligence. Augmenting hardware intelligence with FPGAs (field-programmable gate arrays) has become increasingly common, allowing engineers to load programs dynamically for different configurations of the same basic production-line architecture. Tools have emerged that allow programming FPGAs even without knowing Verilog high-level design language (VHDL).

Emergence of Ethernet in industrial automation applications has made implementation easier as well. Engineers can piece together boutique plants like giant jigsaw puzzles on an Ethernet-based network. Standards have emerged to make the task easier as well. IEEE 1588 precision time protocol (PTP) standard allows synchronizing clocks in sensors and actuators on an Ethernet-based network.

Real-time control


Because suppliers typically establish a confined radius (the distance from a generating facility beyond which shipping product becomes prohibitively expensive), two types of designer plants have developed. One version serves situations where demand is well-known and production levels remain relatively constant—most of the time. For example, industrial gas companies build hydrogen-generating plants that supply petroleum refineries with gas for the cracking process (breaking larger molecules into smaller, more useful molecules). The plant supplies a constant stream of gas for that customer, but, through periodic increases to production, can provide containerized gas or liquid hydrogen for shipment elsewhere. However, the plant must be sufficiently agile to provide additional gas without compromising the refinery's needs.

The other type of plant is suitable when demand varies less predictably. Consider oxygen-generating plants that feed steel mills. A single mill may have 10 steel-making processes, each requiring a different amount of oxygen to achieve necessary production levels. Although the pipeline architecture is the same as with constant flow, here the mill's control system must pull the required amount of oxygen in real time. The generating plant maintains a minimum production level (which occasionally means venting gas when refinery plant requirements fall below that level), and then ramps up as conditions change.

Even with tight real-time control, gas supply disruption could still cause problems in each of the production examples described above. The system would detect a disruption and call for substitution from gas cylinders held in reserve against just such an eventuality. Although this solution would dramatically reduce the dip in the refinery's or the steel mill's output, it would require human intervention with all its accompanying uncertainties. In addition, bottled gas costs much more than locally generated gas —the impetus for constructing the plant in the first place. In competitive industries like petroleum refining or steel making, a manufacturer cannot easily pass site-specific cost increases on to customers, so at the very least, the episode would erode profit margins.


Schematic of a simpple nitrogen generator is typical of those used at a "designer" plant. Air goes in and nitrogen and waste gas results. Source: COntrol Engineering with data from Air Products.

Controlling flow

Controlling an agile plant presents its own kind of challenge. Varying production levels involve adjusting numerous process parameters. In the traditional gas plant, control requires a cadre of operators manually adjusting parameters, such as pressure, purity, and gas throughput.

As an alternative, some industrial gas producers use advanced multivariable process controls. Instead of adjusting process parameters individually, an engineer determines necessary production rates and required purity levels and sends that information to the control system. A multivariable control system looks beyond the instantaneous state of the process. Using its own knowledge of how the process works to manipulate all the appropriate parameters simultaneously, the process control system ensures sufficient flow and gas purity to meet customer specifications. By monitoring the process in real time, the control system can anticipate problems and correct them before product quality or production volume is adversely affected.

For example, suppose a multivariable control system is told that the steel plant will require an increase in oxygen production by a certain percentage at a particular time. Long before the appointed time, the process control system begins to manipulate the control parameters to accomplish the ramp up without compromising either production volumes or product quality during the transition. Its analysis might reveal, for example, that changing a particular parameter will cause purity to fall short of spec. It would take appropriate corrective action in advance by adjusting other parameters so that product quality never crosses that line. Such control operation, however, does have constraints. To avoid process or safety violations, parameters cannot exceed certain predetermined limits.






No comments
The Engineers' Choice Awards highlight some of the best new control, instrumentation and automation products as chosen by...
The System Integrator Giants program lists the top 100 system integrators among companies listed in CFE Media's Global System Integrator Database.
The Engineering Leaders Under 40 program identifies and gives recognition to young engineers who...
This eGuide illustrates solutions, applications and benefits of machine vision systems.
Learn how to increase device reliability in harsh environments and decrease unplanned system downtime.
This eGuide contains a series of articles and videos that considers theoretical and practical; immediate needs and a look into the future.
Sensor-to-cloud interoperability; PID and digital control efficiency; Alarm management system design; Automotive industry advances
Make Big Data and Industrial Internet of Things work for you, 2017 Engineers' Choice Finalists, Avoid control design pitfalls, Managing IIoT processes
Engineering Leaders Under 40; System integration improving packaging operation; Process sensing; PID velocity; Cybersecurity and functional safety
This article collection contains several articles on the Industrial Internet of Things (IIoT) and how it is transforming manufacturing.

Find and connect with the most suitable service provider for your unique application. Start searching the Global System Integrator Database Now!

SCADA at the junction, Managing risk through maintenance, Moving at the speed of data
Flexible offshore fire protection; Big Data's impact on operations; Bridging the skills gap; Identifying security risks
The digital oilfield: Utilizing Big Data can yield big savings; Virtualization a real solution; Tracking SIS performance
click me