Process Energy Systems:
Control, Economic, and
Sustainability Objectives
Jeffrey J. Siirola
Thomas F. Edgar
FOCAPO/CPC 2012
Savannah, GA
1
Outline
Elements of sustainability
New emphasis on greenhouse gas emissions
Carbon management by energy reduction
Smart manufacturing, process control, and
operations optimization
• Dynamic energy minimization
• Next generation power systems (smart grids)
• Thermal energy storage and process control
•
•
•
•
2
Elements of Sustainability
•
•
•
•
•
•
Health and safety
Environmental protection
Materials and energy efficiency
Product stewardship
Corporate citizenship
Triple bottom line
3
Sustainability Issues Addressed
During Design
•
•
•
•
•
Inherent safety principles
High yield reaction chemistries
Material recovery and recycle
Heat integration
Multi-effect separation
• Carbon management remains particularly
difficult and expensive
4
Proposed Legislatively
Mandated US GHG Reductions
http://www.wri.org/climate/topic_content.cfm?cid=4265
5
CO2 Policy Alternatives
• Regulated CO2
– Recent EPA announcement on reporting
requirements
• Cap and Trade
– Establishes firm but decreasing limits on CO2 emissions
– Auctioning/trading of emissions permits
• Carbon Tax
– Price predictability
– Favored by large chemical companies
– Apply to all carbon sources
6
CO2 Absorption/Stripping of
Power Plant Flue Gas
Use 30% of
power plant output
Flue Gas
With 90% CO2
Removal
Stripper
Absorber
Flue
Gas In
Rich
Solvent
CO2 for
Transport
& Storage
LP Steam
Lean
Solvent
7
Base Case Carbon Capture and
Sequestration Technology
• Post combustion monoethanolamine
absorption
– 30% parasitic energy requirement for coal-fired
powerplant
– >70% increase in electric power cost
• Chilled ammonia alternative
• DOE Carbon Capture Simulation Initiative to
address and reduce commercialization risks
8
U.S. Industrial/Building Sector
• Industrial energy usage = 35 quads (total = 100
quads)
• This sector accounts for about one-third of total
U.S. GHG emissions
• By 2030, 16% growth in U.S. energy
consumption, which will require additional 200
GW of electrical capacity (EIA)
• Energy efficiency goals of 25% reduction in
energy use by 2030 (McKinsey and National
Academies Press reports)
9
Reducing Carbon Footprint in
Process Plants
• Fuel swapping (natural gas for coal)
• Conversion to non-fossil energy sources (nuclear,
solar, or biomass)
• Reduce energy requirements
– Use less energy-intensive chemistry/unit operations
– Increase heat and power integration
– Retrofits difficult to justify economically unless
accompanied by capacity expansion
– Operate processes with additional objective to minimize
energy consumption
10
Perspective of this Presentation
• Most carbon dioxide emission comes from
fossil fuel combustion
• Maximize energy efficiency ≡ minimize
carbon footprint
• Focus on process operation and control (not
design)
• Assume use of existing infrastructure to
maximize thermal efficiency
• Progress requires a systems approach
11
Optimization of Operations
•
•
•
•
•
Reduce energy consumption
Improve yields
Reduce pollutants
Increase processing rates
Increase profitability
12
Some Observations
• Most plants do not monitor energy consumption
on an individual unit operations basis, but only
total plant usage for accounting purposes
• Processes may be designed for energy efficiency,
but do not include degrees of freedom and
manipulated variables to minimize energy
utilization during operations
• Schemes control for desired throughput and
product fitness-for-use attributes (composition,
purity, color, etc.), but use utilities (energy) to
achieve these goals and to reject disturbances
13
21st Century Business Drivers for
Process Control (Edgar, 2004)
• Deliver a product that meets customer specifications
consistently
• Maximize the cost benefits of implementing and
supporting control and information systems
• Minimize product variability
• Meet safety and regulatory (environmental) requirements
• Maximize asset utilization and operate the plant flexibly
• Improve the operating range and reliability of control and
information systems and increase the operator’s span of
control
15
16
Transformation of Variation from the
Temperature to Flow for a Reactor Feed
Preheater (Downs et al., 1991)
17
More Observations
• Most multivariable algorithms (like MPC or LQG) do not
assign an economic value to the manipulated variable
moves, although some research efforts have been oriented
towards “economic” MPC
• Energy reuse adding heat and power integration will create
unit and control loop interactions and new disturbance
patterns, making control strategies more complex. Integer
(on-off) variables for equipment such as chillers will need
to be optimized
• Swapping thermal and electrical forms of energy can have
unexpected utilities systems impacts (dynamics and
control)
• Attempting to control carbon emissions as well as energy
usage will require new research investigations in PSE
18
Addition of Sensors and Manipulated
Variables to Minimize Dynamic Energy Use
• In a distillation column, maximize efficiency
by operating near the flooding point
• Balance yield improvement vs. energy use
• Add MV’s with multiple feed points,
bypasses
• Add hard and soft sensors for improved
real-time modeling (e.g., Dzyacky flooding
predictor based on pressures, temperatures,
levels, flow rates)
Predictive Modeling Needed to Manage
Dynamic Energy Use – Refinery Example
• Increased throughput to a crude distillation unit
must consider operating variables for crude tankage,
pumps, preheat trains, and distribution of cuts from
the tower
• Open up valves and let all equipment ramp up? Is
there an optimum way that incorporates energy
use? Perhaps a given ramp rate will result in more
energy efficient performance of downstream units
• If an abundance of fuel gas will be available in one
hour, will that facilitate a much more energy efficient
ramp up, rather than sending the excess to flare?
What is a Smart Grid?
• Delivery of electric power using two-way digital
technology and automation with a goal to save
energy, reduce cost, and increase reliability
• Power will be generated and distributed optimally
for a wide range of conditions either centrally or at
the customer site, with variable energy pricing
based on time of day and power supply/demand
• Permits increased use of intermittent renewable
power sources such as solar or wind energy and
increases need for energy storage
21
Electricity Demand Varies
throughout the Day
Source: ERCOT Reliability/Resource Update 2006
22
Today’s Grid
Smart
Grid 1.0
23
Tomorrow’s Grid
Smart
Grid 2.0
24
Three Types of Utility Pricing
• Time-of-use (TOU) – fixed pricing for set periods
of time, such as peak period, off peak, and
shoulder
• Critical peak pricing (CPP) – TOU amended to
include especially high rates during peak hours on
a small number of critical days; alternatively, peak
time rebates (PTR) give customers rebates for
reducing peak usage on critical days
• Real time pricing (RTP) – retail energy price tied
to the wholesale rate, varying throughout the day
25
26
Future Industrial Environment
• Stronger focus on energy use(corporate
energy czars?)
• Increased energy efficiency and decreased
carbon footprint
• Energy use measured and optimized for each
unit operation
• Increased use of renewable energy(e.g., solar
thermal and biomass) and energy storage
• Interface with smart grids
27
28
Thermal Energy Storage
• Thermal energy storage (TES) systems heat or cool a
storage medium and then use that hot or cold medium for
heat transfer at a later point in time
• Using thermal storage can reduce the size and initial cost of
heating/cooling systems, lower energy costs, and reduce
maintenance costs; if electricity costs more during the day
than at night, thermal storage systems can reduce utility
bills further
• Two forms of TES systems are currently used
– A material that changes phase, most commonly steam, water or ice
(latent heat)
– A material that just changes the temperature, most commonly
water (sensible heat)
29
TES Economics are Attractive
• High utility demand costs
• Utility time-of-use rates (some utilities
charge more for energy use during peak
periods of day and less during off-peak
periods)
• High daily load variations
• Short duration loads
• Infrequent or cyclical loads
30
Energy flows in a combined heat and power system with thermal storage
(Wang, et al. 2010)
Thermal Energy Storage Operating
Strategy with Four Chillers
(a)
(b)
-Chillers 1& 4 are most efficient, 3 is least
efficient
-Chiller 1 is variable frequency
(a) Experience-based (operator-initiated)
-No load forecasting
-Uses least efficient chiller (Chiller 3)
(b) Load forecasting + optimization
-Uses most efficient chillers (avoids Chiller 3)
(c) Load forecasting + TES + optimization
-Uses only two most efficient chillers
(c)
32
Conclusions
• Many opportunities to improve energy efficiency in the
process industries
• Energy efficiency ≡ sustainability (carbon footprint)
• Smart grids and energy storage will change the power
environment for manufacturing
• Development of new real-time modeling, control, and
optimization tools will be critical to deal with this
dynamic environment
• A focus on energy comparable to the current emphasis on
safety would yield significant improvements in energy
efficiency