1575 muhlack et al wvj 28 6 2013

2.1 MB·

a W R I
Optimising fermentation
through simulation
By Richard Muhlack*, Neil Scrimgeour, Eric Wilkes, Peter Godden and Dan Johnson
The Australian Wine Research Institute, PO Box 197, Glen Osmond SA 5064
Managing director Dan Johnson
*Email: [email protected]
What if winemakers had advance warning of sluggish fermentation? What if winemakers could predict
when a wayward ferment might arise and take steps to prevent it, before it happened? the aWRI ferment
simulator is a powerful new weapon in the winemaker’s arsenal. It can be used to predict ferment
behaviour, test and evaluate alternative ferment management strategies, and monitor refrigeration and
electricity demand so that corrective action can be taken on time and on target.
fERMENtatION: a CRItICal stEp Optimal fermentation performance efficiency and productivity of an entire
requires significant resources, typically winery can be affected by stuck and
Demanding, resource-intensive involving daily sample collection, sluggish fermentation.
and sometimes unreliable: laboratory analysis and winemaker Australian winemakers can reduce
fermentation is, arguably, the tastings. It also significantly affects such demands on their resources
most important and critical step in the equipment availability, energy and water through process automation control
winemaking process. use, and refrigeration capacity. The strategies. These are commonly used
in other industries such as beer,
at a glaNCE pharmaceuticals and dairy, where quality
can be assessed and controlled through
online monitoring in real-time. In the
The AWRI’s Ferment Simulator is now available for download from the AWRI website:
wine industry, problem fermentation
• T wo versions are available – a lite version that provides basic simulation functionality,
behaviour could be predicted through
and a full version that models up to 100 concurrent ferments together with
computer simulation, giving producers
refrigeration demand profiles
• P roblem fermentation behaviour can be predicted earlier (with both versions) and advance warning of pending issues
controlled more effectively through computer simulation, giving winemakers advanced before they happen.
warning of pending issues before they occur
sIMulatION: a NEW WEapON
• ‘ What if?’ analysis provides winemakers with the ability to simulate, evaluate and fine-
tune alternative ferment management strategies
The AWRI Ferment Simulator is the
• T he full version also calculates refrigeration load profiles and peak electricity so that
new weapon in the winemaker’s arsenal.
wine producers are better informed to manage site electricity use more effectively and
This novel tool gives winemakers
minimise electricity demand tariffs
the ability to predict fermentation
• T he AWRI Ferment Simulator is a unique and powerful resource for real-time ferment
performance and behaviour, test and
management and maintaining product consistency. It helps winemakers achieve
evaluate alternative management
greater productivity and profitability.
strategies, monitor refrigeration and
Air to Water Liquid Chillers
A new generation of efficient and intelligent chilling
• High energy efficiency, even on part load
• Low refrigerant charge for reduced carbon footprint
• eDrive™ variable water flow for energy savings
• Intelligent climatic control that continuously optimises power consumption
• Lightweight, durable construction with a compact footprint Neosys™
• European designed and made
• Glycol option to -10°C
Ecolean™
• Up to 46°C ambient 20kW 1000kW
13 23 50 heatcraft.com.au
38
www.winebiz.com.au Wine & Viticulture Journal NOVEMBER/DECEMBER 2013 V28N6
For an explosion
of fruit flavour
Result of a natural crossing between 2 Saccharomyces cerevisiae strains selected for specific winemaking
properties, Uvaferm eXencetm is an excellent yeast to accelerate the natural process in the release of
intense thiols aromas such as exotic fruits, peach, apricot and grapefruit for white and rosé wines.
The research and selection were conducted by the Institute for wine biotechnology, University of Stellenbosch (Republic
of South Africa) and Lallemand.
Lallemand Australia Pty Ltd | 23-25 Erudina Ave, Edwardstown, South Australia 5039 | Australia | Tel: [phone]
Lallemand Oenology: Natural Solutions that add value to the world of winemaking / www.lallemandwine.com
Lallemand Oenology: Natural Solutions that add value to the world of winemaking / www.lallemandwine.com
a W R I
figure 1. particle swarm analysis allows the simulator to rapidly converge on the optimal parameters, as shown here after
one, 10 and 20 calculation steps, respectively, to enable rapid and accurate prediction of ferment status.
electricity demand, and flag potential Simulator puts this theory into practice The AWRI Ferment Simulator
problems so that corrective action can be by using biochemical equations to model incorporates new elements to address
taken early, exactly when needed. fermentation performance. In this regard, heat transfer and hydro-dynamics –
Two versions of the tool are now it is similar to other models that have factors that are important in commercial
available for download from the AWRI been presented in scientific literature: winemaking. One such innovation is
website: a lite version that provides basic see Boulton (1980), Caro et al. (1991), particle swarm analysis (PSA), which is
simulation functionality and ‘what-if?’ Cramer et al. (2002), Sainz et al. (2003), used to adapt the simulation to changing
capability, and a full version. The full Malherbe et al. (2004) and Coleman et al. commercial ferment conditions such as
version offers all the features of the lite (2007). temperature, yeast, wine type, nutrient
version and also allows modelling of up There are drawbacks to these other levels, agitation regime and tank size.
to 100 concurrent ferments together models, however. They require users to PSA is a type of artificial intelligence
with the ability to generate refrigeration estimate a large number of parameters, algorithm that has been used for a range
demand profiles across the entire tank which can make them difficult to apply of problems across a number of industry
farm. in a commercial situation. Some use sectors including mining, finance,
artificial intelligence models to generate mapping, aerospace and defence. It is
hOW DOEs It WORK? designed for modelling complex, non-
relationships between variables; others
linear problems and is well suited to
The tool applies scientific principles use mathematical models without
commercial wine fermentation. It allows
relating to classic engineering process taking physical processes fully into
the simulator to quickly and efficiently
control theory. In order to make correct account. Above all, their fundamental
determine the optimal parameters
decisions on how to control a process limitation is that, generally, they assume
that match the real-world ferment
(in this case, a wine fermentation), perfect mixing and uniform conditions
being modelled. It generates this key
information on process conditions throughout the fermentation vessel.
information with minimal computation,
must be combined with knowledge of Such assumptions may be valid under
as shown in Figure 1.
how a process behaves under different laboratory conditions in a fermentation
circumstances. From an engineering shake-flask or micro-fermentor, but What CaN It DO?
perspective, this knowledge comes not they will rarely apply in a large-scale
only from human experience, but from commercial ferment. Although these The AWRI Ferment Simulator
process modelling and simulation. models can offer general insights into was developed over several vintages,
Developing a model can help to the conditions affecting fermentation, incorporating input from industry
promote better understanding and their ability to predict fermentation collaborators to determine its performance
visualisation of the system that users performance in a commercial requirements and features. To test and
would like to control. The AWRI Ferment environment can be limited. evaluate the model, a wide variety of
ISECO Engineering Services Pty Ltd 723 Burwood Road iseco.com.au
Hawthorn East email: [email protected]
Specialist Refrigeration Consultants Victoria 3123 tel: [phone]
Australia fax: [phone]
With over 12 years of specialist winery engineering experience
We provide reliable and impartial engineering design and review services for a range of winery size and type:
• Energy Audits OEH Energy Saver panel member • Brine and glycol system design • Advice on alternative refrigerants and how to
• Winery & tank farm refrigeration capacity • Cold rooms, grape & must cooling reduce winery refrigeration carbon emissions
• Australian Refrigeration Standard • Hot water systems design • Engineering support for Clean Technology
AS 1677 Compliance audits • Compressed air design Investment Program grant applications
40
www.winebiz.com.au Wine & Viticulture Journal NOVEMBER/DECEMBER 2013 V28N6
a W R I
commercial fermentation data – The full version of the simulator What aRE thE BENEfIts?
across multiple production sites and allows wine producers to follow multiple
vintages – was used to assess its concurrent ferments across a tank The primary benefit of computer
performance under different operating farm, with a visual ‘traffic-light’ display simulation is that problem fermentation
conditions. Its performance has been indicating the fermentation status of behaviour can be predicted earlier and
evaluated on two fronts: the ability each active ferment. Winemakers can controlled more effectively than would
to consistently fit a full fermentation also follow refrigeration load profiles otherwise be possible. Early warning gives
profile and its capability to reliably (see Figure 2, see page 42) by tank or winemakers the opportunity to take action
predict fermentation performance across all active ferments based on and prevent problems before they occur.
from initial conditions. The risk of a wayward ferment is reduced
fermentation progress and expected
There has been a deliberate effort as a result.
weather patterns. Warnings are given
to minimise the information required If a potential issue is identified, ‘what
when total refrigeration capacity is
to run a simulation. The result is a if?’ analysis provides winemakers with
expected to be exceeded.
model that simply requires the user the ability to simulate, evaluate and fine-
Both the lite and full versions
to supply tank volume together with tune alternative ferment management
include a ‘what if?’ analysis capability.
initial temperature and Baume data. strategies. When using the full version,
This allows winemakers to assess
Users can also supply additional users can calculate refrigeration load
the impact of management strategies
information: initial cell mass and profiles and develop appropriate plans
such as temperature adjustment, yeast
nitrogen levels, information on DAP to mitigate the risk of exceeding peak
nutrient addition and tank agitation. For
additions, any external agitation electricity demand. This information can
applied, the addition of nutrients, and example: make a significant difference to wine
the inoculation date. Information on “What if I increase the temperature producers’ operating costs and profitability:
wine or yeast type is not required. by 2°C on day 3?” or “What if I add DAP it offers early warning so that wineries
During industry evaluation, the on day 4?” can manage electricity usage effectively
ferment completion times predicted This capability means that changes (particularly on hot days) and take steps to
by the simulator were found to be to processing can be simulated before schedule winery operations appropriately
accurate. Using only two to three days they are implemented, to ensure that to avoid punitive electricity costs.
of input data, the difference between the optimal strategy is chosen (Figure 3, The tool provides a unique and
actual and simulated ferment times see page 43). powerful resource for continuous quality
was one to 1.5 days or less, in most improvement and product consistency from
cases. ferment to ferment.
DiPel
DiPel® DF has been widely used for many years in grape
BIOLOGICAL INSECTICIDE
production and continues to provide good cost effective
control of lightbrown apple moth and vine moth.
Resistance No re-entry No WHP, Safe to all Stops feeding
management lnterval no residues beneficial immediately
rotation option insects
DiPel – Be Biorational
Scan this QR code
for more information
www.sumitomo-chem.com.au
about DiPel DF: DiPel® is a registered trademark of Valent BioSciences Corporation, Libertyville, IL, USA. ALLOWED INPUT 10332AI
41
V28N6 Wine & Viticulture Journal NOVEMBER/DECEMBER 2013 www.winebiz.com.au
a W R I
figure 2. predicted refrigeration demand in relation to plant capacity over the coming 10 days.
WhERE CaN I aCCEss thE aWRI Both versions can be downloaded as A LibreOffice version (compatible
fERMENt sIMulatOR? Excel files. The lite version (compatible with v3.5 onwards) of the AWRI Ferment
with Office 2003 onwards) provides basic Simulator – full version is also available
The two versions of the AWRI Ferment simulation functionality; the full version for Linux operating systems. The source
Simulator are now available from the (compatible with Office 2007 onwards) code has been unlocked, allowing
AWRI website. offers more advanced features. wine producers to adapt the simulator
For further information, please contact Kauri
NZ Tel: 0800 KAURIWINE AUS Tel: [phone]
NZ Fax: [phone] AUS Fax: [phone]
Email: [email protected] Web: www.kauriwine.com
42
www.winebiz.com.au Wine & Viticulture Journal NOVEMBER/DECEMBER 2013 V28N6
a W R I
figure 3. after 3.5 days a sluggish ferment (of 14.5 days duration) is predicted (left). an increase in ferment temperature
together with a Dap addition and a pump-over is shown to bring the ferment back ‘on-course’, with ferment now complete
after eight days (right).
functionality to interface directly with their matching funding from the Australian Coleman, M.C.; Fish, R. and Block, D.E. (2007)
onsite LIMS/PLC/SCADA systems for Government. The AWRI is a member Temperature-dependent kinetic model for nitrogen-
limited wine fermentations. Appl. Environ. Microbiol.
automatic data acquisition if desired. of the Wine Innovation Cluster. The
73(18):5875-5884.
To access and download the authors thank Sharon Mascall-Dare
Cramer, A.C.; Vlassides, S. and Block, D.E.
simulator and obtain further details, and Ella Robinson for their editorial (2002) Kinetic model for nitrogen-limited wine
visit www.awri.com.au assistance. fermentations, Biotechnol. Bioeng. 77(1):49-60.
Malherba, S.; Fromion, V.; Hilgert, N. and
aCKNOWlEDgEMENts REfERENCEs Sablayrolles, J-M. (2004) Modelling the effects
of assimilable nitrogen and temperature on
This work is funded by Australian Boulton, R.B. (1980) Prediction of fermentation fermentation kinetics in enological conditions.
winemakers and grapegrowers behaviour by a kinetic-model. Am. J. Enol. Vitic. 31: Biotechnol. Bioeng. 86(3):261-272.
through their investment body, 40-45. Sainz, J.; Pizaro, F.; Perez-Correa, J.R. and
Caro, I.; Perez, L. and Cantero, D. (1991) Development Agosin, E. (2003) Modelling of yeast metabolism and
the Grape and Wine Research and
of a kinetic-model for the alcoholic fermentation of process dynamics in batch fermentation, Biotechnol.
Development Corporation, with must. Biotechnol. Bioeng. 38:742-748. Bioeng. 81(7):818-828. WVJ
40’ Refrigerated Container

Proven Carrier Thinline Call us today for
69NT40 machinery expert advice and
reliable service
Sales and hire
Temperature range +20°C / -25°C
11 Bellchambers road,
Suitable for 1165 x 1165 x 150
Edinburgh North
Pallets X 8 Call [phone]
www.irssa.com.au
21’ Palletwide Reefer

Proven Carrier Thinline
69NT40 541-403 machinery

Double stacked Deck Beam
storage capability with cargo
securing devises (load locks)

(500 kg even load distribution)
Suitable for 1165 x 1165 x 150
pallets x 10
43
V28N6 Wine & Viticulture Journal NOVEMBER/DECEMBER 2013 www.winebiz.com.au