There's more than one way to skin a cat, but there are thousands of ways for process applications to get more efficient and save energy. Because gains are possible in so many production stages, settings and components, they require thoroughly investigating individual applications, focusing on requirements needed to implement potential savings, and coordinating staff to maintain them.
Of course, the search for efficiency unfolds against a backdrop of rising and falling energy prices, logistics and availability issues, as well as standards and regulations mandating increased energy conservation, such as ISO 50001 on site energy management and the European Union's 2020 energy strategy for reducing greenhouse emissions. However, even though fuel and power have been less expensive lately, most users now seem to have it permanently on their do-more-with-less radar. Rules aside, this is likely because they have the right software and other tools to make managing energy a true contributor to their bottom lines.
"Most energy efficiency efforts were historically driven by price shocks, but it's different now because we've reached a point where metering is part of the whole manufacturing process. There are more—and more accurate—meters, not only on pump controls and flowmeters, but also on more electrical equipment," says Bruce Billedeaux, senior consultant at system integrator Maverick Technologies, a Rockwell Automation company. "Previously, there was just an electric meter when power came into the building, but now the whole energy stream is metered.
"For example, there used to be many unmetered valves, but now every variable-speed drive (VSD) has a meter that can show energy consumption. Onsite and close-to-real-time process controls have advanced enough that they can be used for energy monitoring and management, too. Previously, orifice plate flowmeters with DP sensors were only ±5% or ±10% accurate, and they weren't accurate enough for energy metering. But now we have magnetic and transit-time ultrasonic flowmeters with better than 1% accuracy, and they can be used to measure energy, too."
Billedeaux adds, with the belief that natural gas prices will remain historical low for a decade or more even if demand grows, there's increasing interest in using it instead of electricity. "This is obvious for thermal applications, but there's also interest in using natural gas for prime movers and onsite generation," he explains. "With low cost gas, and coincident thermal demand, co-generation has great prospects."
Michael Robinson, national marketing manager for projects, solutions and services, Endress+Hauser, adds that, "Implementing energy efficiency in process automation can be a difficult business decision to champion because new and added measurement devices are often needed, and that means increased technology migration and installation costs, which is why many projects don't get funded during customers' budget processes. This is unfortunate because in energy-efficiency projects with advanced process control (APC), for example in product dehumidification, these types of projects yield double benefits, such as consumption of less BTUs of heat and providing users more direct control of the moisture content in the produced product. For instance, if the customer contract states that a specified product will have no more than 20% moisture, we can now help them get to 19.9% instead of the 17% they are currently operating at. As a result of often prohibitive installed costs, we typically assist with a lot of micro projects to prove out the assumptions of energy efficiency programs, but often these projects fail to produce the necessary internal rate of return on the capital invested versus what other capital investments can yield, and these projects find no internal champion in the organization and get shelved."
No doubt one of the most popular times to seek and add energy-saving solutions is during periodic renovation and migration projects. As part of its own ongoing savings program, MOL Group recently improved energy efficiency and process stability at its up to 12-million-m3/d gas distillation plant in Algyõ, Hungary, by upgrading its distributed controls to Emerson Automation Solution's DeltaV process automation system (Figure 1). MOL also implemented DeltaV's embedded APC by adding its SmartProcess Distillation Optimizer software that enabled model-predictive control for the plant's distillation columns, simplified installation, and eased operator adoption. This improved control and performance of Algyõ's gas fractionation section, and reduced MOL's energy costs by €1.2 million annually.
"Our experience shows distillation columns are highly interactive, multivariable processes that are difficult to control, and so DeltaV was customized for us," says Attila Bodócs, production chief at MOL. "DeltaV, its embedded APC tools, and its engineering/consulting services were instrumental in helping MOL achieve such a significant efficiency improvement.”
Bodócs reports that SmartProcess automatically adjusts key flows on the plant's column, which enables MOL to make product more reliably and safely. In fact, as soon as APC was commissioned on the first column, MOL began generating benefits, including ensuring that product quality more closely matched customer specifications without using excess energy. So far, Emerson’s distillation solution was applied to five of the plant’s six gas distillation columns. Hot oil consumption was reduced 40%, and operational stability improved as the process automatically corrects variations. Also, operator workloads were reduced because there's is no longer a need to make small manual adjustments, which were previously necessary to keep the column operating at product quality limits.
Bob Sabin, consulting engineer with Emerson's Industrial Energy Solutions Group, adds that, "Many energy efficiency tools have existed for a long time, but lately their costs have been coming down, which means energy doesn't have to be looked at in a vacuum anymore, but can be just another key performance indicator (KPI) in a good manufacturing operation. More of our customers are considering energy along with their other KPIs, and it's one that can be easily affected by local operating decisions, more so than items like raw material costs. If your site has good reliability, availability and energy control, then it's more likely you're going to outperform your industry peers."
Sabin adds that most energy-intensive applications like boilers, fired heaters and distillation columns operate with single-loop input controls and computers, which keep them running fully automatic for optimal throughput, stability, quality and safety. "What's changed is that control blocks and functions are now more straightforward to use, so engineers need to know their controls can benefit from standardized hardware, PCs, networking buses and more mature software, and that their closed loops can talk fast enough to incorporate energy, too," he says.
Likewise, Robin Moulder, senior engineer at NRG Energy, and Ewa Michalowski, senior project manager at Schneider Electric just began replacing the oil igniter controls on the coal-fired Unit 1 at NRG's Big Cajun II power plant in New Roads, La. This igniter routinely starts Unit 1 in response to local electricity demand. The 35-year-old power plant has three units and is rated at 1,708 megawatts (MW).
All three units were designed to burn Powder River Basin (PRB) sub-bituminous coal. Unit 1 and Unit 3 presently burn PRB coal, but Unit 2 was converted burn all natural gas and is no longer coal-capable. The igniter on Unit 3 had been upgraded to using natural gas a year earlier. Replacing the igniter's controls requires custom cabinets and I/O, new logic for controllers, factory acceptance tests (FAT), installation, testing and commissioning.
"We had to convert Unit 1's igniter so it could burn natural gas," says Moulder. "It used to burn only diesel fuel oil, and it was controlled by Foxboro I/A and some PLCs for local functions, such as outputs to valves and lights. We needed to go to natural gas as the start fuel, and control it with Foxboro I/A with remote I/O, so we wouldn't have to wire as many field devices and I/O points back to the relay room. Because our plant is 35 years old, this room is under the control room, and all devices were originally wired to it. Now we only need wires going to our valve-control racks, which is a more efficient use of our factory-assembled skids and racks. We also typically use fiberoptic cabling to reach more remote systems like our burner management system (BMS) and distributed control information system (DCIS)."
Dr. Peter Martin, vice president of business value consulting and Edison master at Schneider Electric, explains, "When you go back 15 years, the price of power was consistent over six months to a year. However, deregulated grids and competition varied supply and demand over those connections, and this created a relationship between price of electricity and consumption. This created a need for combining process and power control, and adding them to energy pricing models."
Martin adds that Schneider Electric's combined control solution is its year-old Integrated Power and Process Management (IPPM) program, which manages energy consumption by balancing the three critical variables of energy, raw materials and production value, which is the actual value of an end product at the time it was produced. “All three variables are closely related, which means there can be great variation in production value based on when products are made," says Martin. "The right discussion becomes what are we doing in real time to be effective, and what can still be transactional? Monthly data isn't good enough anymore, so we're also moving to real-time accounting and data analysis."
Power joins control gang
Despite all the economic and technical forces that continue to swirl around energy, what's clear is that power is breaking out of its old, static, accounting-based, pay-the-utility-bill silo, and becoming a full-time, dynamic process control variable that can contribute to profits.
"Some facilities have implemented more advanced controls as they drive for increased production and better yield from raw materials, and often there's an energy objective, but in many cases it's included as just one dimension within a larger picture," says Perry Nordh, product manager for advanced control, optimization and monitoring at Honeywell Process Solutions. "Energy efficiency is quite complex and covers production plants, utility plants, planning and scheduling. In some cases, utilities such as steam and power production were optimized separately, but focus has started to shift to overall efficiency and optimization, and new methods are required."
Nordh adds that there are two sides to bringing energy efficiency into process control—produce energy more cheaply, and consume less of it. "More efficient production of steam and/or electricity means optimization of utilities (MWh, steam, fuels, H2) according to production plans and schedules. Optimizing equipment can be done with applications like Honeywell’s Advanced Energy Solutions, which can help coordinate energy demand with efficient production," he says. "Energy and steam demand is managed with multivariable predictive controls (MPC) that predict future errors and correct them before exceedance occurs. On the energy consumption side, many process units are already managed by multivariable predictive controllers (MPC), such as Honeywell’s Experion Profit Controller, which allows multiple objectives to be optimized simultaneously, including production throughput, yield and energy usage. This way, energy efficiency simply becomes part of the objective function and Profit Controller will drive to the true economic optimum based on the optimized variables and unit costs."
Profit Controller resides in Honeywell's Experion 500 DCS, and runs on its C300 controller and in its Application Control Environment (ACE) node, which is an application server that sits on an Experion network and is tied to its server. "This lets younger and less experienced users learn the intricacies of optimization because it shows and makes performance curves easier to understand, and helps them manage their high and low points and other control problems," adds Nordh. "We've also established an Optimization Center of Excellence (OCE) to advise users on next steps because even though energy should be part of production control, it can be very hard to separate and handle one curve without affecting others."
Software streamlines, simplifies
Probably the biggest aid to bringing energy efficiency under the process control umbrella is the emergence of increasingly real-time-capable optimization software combined with APC techniques.
"The most important change in the state of energy efficiency is that just about every engineer is asking, 'How much is that piece of machinery or widget going to cost me in energy use?' And process control engineers are taking it a step further, and analyzing things like power quality to make sure they can increase uptime on their production lines," says Matt Newton, director of technical marketing at Opto 22. "APC can dramatically improve energy efficiency in process control applications. Compared to traditional proportional-integral-derivative (PID) controllers, APC offers the ability to manage applications that require multiple inputs and outputs to run. More importantly, an APC system can leverage: external data sources such as KPI targets, recommended best practice settings from OEMs, and historical production databases to more effectively monitor and control a given process."
In fact, during a recent modernization project at its refinery in Mumbai, Bharat Petroleum Corp., Ltd., (BPCL) reports it updated the plant's diesel hydro-desulfurization/hydrogen-generation (DHDS/HGU) units by implementing APC. BPCL adds its Mumbai refinery is one of India's most flexible, versatile and productive, and processes about 12 million metric tons of raw material per year into 61 types of crude. However, it needed to increase throughput of the crude and hydrocracker units; improve stability of column profiles; maintain reformer steam-to-carbon ratio and temperature profile; maintain first- and second-stage reactor temperature profiles; and maximize production of liquified petroleum gas (LPG) and heavy diesel.
To those ends, BPCL implemented Profit Controller and Profit Sensor Pro software from Honeywell, and accessed its engineering support services to design, install and maintain multiple input/output (MIMO) applications of APC in BPCL refinery units, including crude distillation, vacuum distillation, naphtha stabilizer, HGU, hydrocracker unit (HCU) and lube oil separation.
“Profit Control and Profit Sensor Pro let BPCL to improve profitability with increased production, reduced operating costs, increased energy efficiency, and improved product quality,” says V. Suresh, deputy GM for technology at Bharat Petroleum.
Based on patented Robust Multivariable Predictive Control Technology (RMPCT), Profit Controller and Profit Sensor Pro enabled the refinery's DHDS/HGU units to reduce variation in methane slippage; increase reformer optimization due to skin temperature inferential; reduce variation in diesel sulfur and reduce fuel gas required by the feed heater; and reduce steam in the diesel stripper due to reduced inferential diesel flash variation. BPCL also documented an overall reduction in the plant's fuel gas and steam use, which added financial benefits.
Elsewhere in India, Tata Power's Coastal Gujarat Power Ltd. has been implementing its 4,000-MW (800 MW x five units), coal-fired, supercritical Ultra Mega Power Project (UMPP) near the port city of Mundra. However, Tata also wanted to implement fleetwide monitoring and diagnostics of the health and performance of its plants, and turn their data into real-time insights for more efficient operations and proactive maintenance.
Tata eventually selected Schneider Electric’s Avantis Prism predictive asset analytics software for its remote monitoring and diagnostics center, and has been completing a three-phase deployment of Prism for two units of CGPL's UMPP plant. Phase 1 used Schneider’s Maintenance Diagnostics and Services Center (MDSC) for model building, tuning and training on two of the 800-MW units and their boilers, steam turbines, cooling water pumps, coal pulverizers, fans, boiler feed pumps, generators and transformers, while Tata is deploying Phases 2 and 3, which include three more 800-MW units at Mundra, two units at Tata's Trombay plant and its other facilities.
Beyond continuously monitoring health and performance of UMPP's critical assets, Prism provides early warnings of equipment that's performing poorly or is likely to fail. Prism is based on an algorithm called Optics, which uses advanced pattern recognition (APR) and machine learning to determine an asset’s unique operating profile during loading, ambient and operational process conditions, and help with root-cause analysis and provide fault diagnostics. Tata also uses Prism's web application to deliver and manage alerts, and it built an Advanced Center for Diagnostics and Reliability Enhancement (ADoRE) for remote monitoring and diagnostics, dispatching personnel, and enabling knowledge sharing and collaboration.
So far, Tata has deployed more than 300 models with monitoring and diagnostics, and used Prism to catch equipment degradations early. In one catch, UMPP found a bypass valve on a low-pressure heater was partially open when it should have been closed, which caused condensate ﬂow to bypass through the heater, resulting in higher and inefficient extraction temperatures. In another catch, top thrust and guide-bearing temperatures of circulation water pumps were rising above expected levels. Each unit has two pumps for handling an 800-MW load, and circulation water pump bearings are used for external sealing and cooling using clariﬁed water. Thrust and guide bearing temperatures are monitored by three redundant RTDs with trip/alarm values at 195 °F and 185 °F, respectively, using two out of three (2oo3) protection. In this case, Prism showed the rise was 50-65 °F above predicted values. The pump was later inspected for suspected clogging in the bearing cooling water line. After clearing the block, the bearing temperature and generation potential were normalized. Identifying and investigating these issues before they caused serious equipment damage resulted in substantial savings, as well as performance and reliability improvements. Estimated cost savings on this catch was $270,000.
“We've found Avantis Prism to be an effective predictive diagnostics tool for detecting functional deviations and impending failures at an early stage for initiating suitable prioritized maintenance actions for enhanced reliability of critical power plant equipment,” says Praveen Chorghade, chief of core technology and diagnostics at Tata Power.
Getting efficiency ducks in a row
To meld power and process control successfully, developers and organizers report they need some organizing principles to help coordinate their roles and collaboration.
"The first task in energy efficiency is to look at all your streams, and determine where your power is going now and what it's worth. Many facilities know they're burning gas and consuming electricity, but they don't know, for example, that their 500 °F drying process is wasting all kinds of heat," says Maverick's Billedeaux. "Every application and plant should have an energy balance that includes what energy is coming in, where is it leaving, and identifying the value of each. Fortunately, part of evaluating that energy value means accessing data from controls that are already in place."
In its "Sustainable Energy Efficiency Users Guide," Emerson defines an energy management information system (EMIS) as "fundamentally, a cascaded target setting and reporting structure for energy data and operating variables (see sidebar). EMIS starts from high-level performance measures within the site manager’s portfolio and cascades through operational areas and structures to short-term control parameters at the plant operator level, such as boiler/furnace/gas turbine firing conditions. At all levels, KPIs, frequency of review and corrective action loops need to be defined for energy management system (EMS) processes. Generally, an EMIS will group units at a site by their commonalities, such as a common: utility supply, operating objectives or operation management for line responsibility. Typically, an energy balance is made around these units and KPIs are established and calculated online. KPIs could include energy index at a site, area or unit, total stack energy loss, energy/feed ratio, etc." (Figure 2)
Emerson's Sabin reports its solution in this area is its Energy Advisor EMIS software, which sits alongside historians, aggregates data, builds performance models, and generates timely alerts. It also offers Smart Process Energy software configured in DeltaV for closed-loop, real-time monitoring and control of steam, electricity and other generating applications for improved stability and responsiveness. "Smart Process Energy employs standard software templates and blocks to create customized, first-principal models in DeltaV," explains Sabin. "It solves best-cost positions within constraints every 10-15 seconds, directs load allocations on steam or electricity production units, and enables better decisions on making, buying or selling power."
Interfaces go long and short
To view data from combined and more efficient process and power controls—and make better decisions—users require more flexible HMIs and other interfaces. For instance, the City of Ann Arbor, Mich.'s Wastewater Treatment Plant (WWTP) is four years into a five-year project to renovate and expand its 30-million gallons per day (mgd) facility, which has two main sections, including its old, 9.5-mgd west plant and its newer, 20-mgd east plant. WWTP serves about 130,000 residents in Ann Arbor and nearby. Average daily processing is about 18 mgd, which helps WWTP maintain its required 24/7 operations.
"This latest project meant we had to replace a lot of old and obsolete equipment, and we also needed a new aeration system that required blowers and variable-frequency drives (VFD)," said Earl Kenzie, P.E., manager of WWT services, City of Ann Arbor Water Utilities. Because the blowers in Ann Arbor WWTP's aeration system are crucial for providing air to microorganisms or "good bugs" consuming dissolved solids at the plant, Kenzie reports that upgrading them was essential.
"These blowers were previously continuous speed, and so it was important to move them to variable-speed drives (VSD) and take up less space," explains Kenzie, who adds the migration allowed the blowers to save 30% on power, which saves WWTP $85,000-$90,000 per month.
Besides migrating to VSDs, the utility also adopted Rockwell Automation's PlantPAx control system, scalable ControlLogix controllers and redundant servers. Because WWTP also wanted to adopt some virtual tools, Kenzie checked on using pre-built virtual machines (VM) from Rockwell Automation on virtual servers. Overall, the HMI side of WWTP's upgrade project included development of 425 screens, which display data from 15,000 tags and 36 wastewater processes managed by four older PLCs and 30 ControlLogix controllers and CompactLogix controllers. Kenzie added that the VMs reduced the upgrade's deployment time, and made the plant's infrastructure more robust, while their graphics effort also reduced configuration time.
"This lets operators view profiles for several processes at once, and click on areas such as clarifiers and sludge pumps to see if they're discharging the right number of gallons, or read airflows for the aeration tanks," adds Kenzie. "Plus, WWTP can use the displays to train operators and SCADA supervisors."
Phil Kaufman, energy technology manager at Rockwell Automation, adds that, "What's new to the energy efficiency game is that we can bring smarter drives and circuits, plus data we've collected for years that wasn't used before, and employ it externally to subtract one type of consumption from the other loads. So, we're not just aware of an improvement, but we know what it is. "
Kaufman adds that Rockwell Automation also uses Common Industrial Protocol (CIP) networking like EtherNet/IP, and its recently released CIP Energy standardized protocol can instantly take energy data and 'odometerize' it to show power and flow metering, as well as drive overloads. "Data from CIP Energy can be used in algorithms and dashboards for HMIs, stored in historians, or moved to the cloud for analysis. Fuel is also an ingredient in many applications, which is why we want to view it as a process variable. We're seeing more people who want to go here because if they can save even five cents on energy, then it goes right to their bottom line."
Beyond assistance from software and other tools, another primary force aiding energy efficiency is, you guessed it, the networking and simplification granted by the Industrial Internet of Things (IIoT).
"Process control and power solutions are continuing to merge under the umbrella of software, IT and Internet-based systems," says Opto 22's Newton. "We’ve seen a number of players in the cloud-based energy monitoring market, and I expect this to increase as demand-response systems become more mainstream and the existing grid continues to transition towards a smart grid."
Emerson's Sabin explains the industrial Internet is just another way to leverage measurement and sensing to aid efficiency, which is fortunate because its technology has also mushroomed as its costs have come down. "This gives users a new vision of their operations, lets them capture real-time state-of-performance across whole sites, and see their energy use almost in real time," he explains. "So, after capturing data when their application is running well on a good day, users can run subsequent data and compare it to their historical ideal. This lets them find out in five minutes if they have a plugged line, wrong ingredient or other abnormal situation, and fix it immediately."
Maverick's Billedeaux adds, "We've been following the demand response movement, in which customers work with energy providers, usually an electric utility, to better match demand with generation capacity," he says. "With the backing of the Federal Energy Regulatory Commission, most large utilities have a demand response program. Demand response is entirely dependent on Internet connectivity and integration."
Future optimization and optimism
Logically, because of the benefits they deliver, ties between process and power control are only going to get closer and more closely intertwined.
"In the future, energy efficiency is going to be even more integrated with real-time control. All the tools are already there," says Emerson's Sabin. "Users can have good energy control and performance at no extra cost, along with meeting other goals of throughput, quality and safety. At present, many users run extra boilers just in case, use more costly fuel because cheaper ones don't fire as well, or generate high-pressure steam and use pressure reducing valves—many times sacrificing costs to achieve reliability. But with today's software and tools, this is obsolete, and users no longer have to give up energy efficiency and lower costs to achieve reliability and proper response."
Honeywell's Nordh adds that, "On a larger scale and with a securely connected system, very complex problems can be addressed. For example, in a complex reactive process unit like a UOP CCR Platformer for high-octane, low-sulfur gasoline, it's been impossible to be an expert in all facets of the process, equipment, operational details and control system. With the IIoT, experts can be a virtual part of onsite teams with a connected view into the performance and operation of the unit, and give recommendations through programs like UOP Connected Performance Services. So, when users want to know if it's worth starting their process application, they'll know for certain that they're making the right product at the right time."
Maverick's Billedeaux adds, "As we move from coal to natural gas, cogeneration, micro-grids, local and self-generation, and a mix of renewables—process controls can help with all of them."
Schneider Electric's Martin concludes, "I recently asked Greg Shinskey and others if process control was becoming a lost art and if its heyday was over, and they argue that control's heyday is actually on the horizon because it can be applied in real time to so many new areas like energy efficiency and others to aid reliability, maintenance, safety and profitability."