Methodology of rapid evaluation of capital and operating cost of biorefineries with applications in multi-scale problems

Τσαγκάρη, Μιρέλα; Tsagkari, Mirela

dc.contributor.author	Τσαγκάρη, Μιρέλα	el
dc.contributor.author	Tsagkari, Mirela	en
dc.date.accessioned	2017-11-30T10:06:22Z
dc.date.issued	2017-11-30
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/46017
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.2840
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	biorefinery	en
dc.subject	cost estimate	en
dc.subject	investment	en
dc.subject	CAPEX	en
dc.subject	OPEX	en
dc.subject	conceptual estimate	en
dc.subject	early-stage estimate	en
dc.subject	short-cut model	en
dc.subject	heuristic model	en
dc.subject	cost uncertainty	en
dc.subject	cost method	en
dc.subject	βιοδιυλιστήριο	el
dc.subject	βιο-οικονομία	el
dc.subject	κόστος επένδυσης	el
dc.subject	κεφάλαιο	el
dc.subject	bio-economy	en
dc.title	Methodology of rapid evaluation of capital and operating cost of biorefineries with applications in multi-scale problems	en
dc.title	Μεθοδολογία ταχείας εκτίμησης κεφαλαιουχικού και λειτουργικού κόστους βιοδιυλιστηρίων με εφαρμογές σε προβλήματα κλίμακας	el
dc.contributor.department	Industrial Process Systems Engineering Unit	el
heal.type	doctoralThesis
heal.classification	ΜΗΧΑΝΙΚΗ ΧΗΜΙΚΩΝ ΔΙΕΡΓΑΣΙΩΝ	el
heal.classification	Estimates	en
heal.classification	Engineering--Estimates	en
heal.classification	Industrial engineering	en
heal.classification	Process engineering	en
heal.classification	CHEMICAL PROCESS ENGINEERING	en
heal.classificationURI	http://data.seab.gr/concepts/e193822c2c109b72b92877998fe2b614e3cd6bf4
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh00005641
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh85043181
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh85065864
heal.classificationURI	http://zbw.eu/stw/descriptor/15677-1
heal.classificationURI	http://data.seab.gr/concepts/e193822c2c109b72b92877998fe2b614e3cd6bf4
heal.dateAvailable	2018-11-29T22:00:00Z
heal.language	el
heal.language	en
heal.access	embargo
heal.recordProvider	ntua	el
heal.publicationDate	2017-07-13
heal.abstract	The thesis outlines a framework for rapid capital and operating cost estimation of evolving biorefineries. The cost models are addressed to the biorefinery community and aim to assist chemists in selecting economically viable biorefinery routes and engineers in avoiding laborious flowsheeting with uncertain economic analyses at process conception. The models assist in early-stage decision making for budget allocation in a portfolio of different projects and estimation of budget overrun risks. Biorefineries are society’s alternative to fossil-based production and aim to produce bio-fuels and chemicals from a wide variety of biomass sources by optimising production and resources use. The emerging field of integrated biorefineries offers an abundance of possible products, technologies and production routes. The systematic screening of capital expenditure of each biorefinery pathway still relies on cost heuristics regressed from petrochemical processes under the assumption they could be applied to biorefineries. The thesis studies several established cost methods on biorefinery processes and concludes that most require a high level of process detail not suitable at the early R&D stage and at times, they produce questionable results. A cost estimating methodological framework is proposed to develop short-cut cost models and cost estimates for biorefineries. The framework sets the basis for the systematic cost modelling and estimating of biorefineries and is reproducible and extendable to other process groups. To address the uncertainty around the term “early-stage estimation”, technology development levels are benchmarked against cost estimation standards: Technology Readiness Levels (TRL) 1 to 4 correspond to Class 5 cost estimates and Technology Readiness Level 5 corresponds to Class 4 cost estimates, each level requiring dedicated cost models within - 50 to + 100 % accuracy. Heuristic and reproducible cost models are developed for TRL 1-5 by mapping cost drivers from cost data inventories. The cost models produce probabilistic base estimates, i.e. point estimates that take into account the cost model’s uncertainty within 95 % certainty. The cost estimate is completed with uncertainty models that takes into account both statistical and technical challenges to produce a joint distribution of possible cost outcomes. The analysis is assisted by cost inventories that include commercial biorefinery costs gathered to that end. Three major biorefinery classes emerge from the data to classify biorefineries by type of employed technology: biochemical, chemical and thermochemical. Historical trends show that for the period 2000 – 2020, the US has been a pioneer in biorefineries construction (43 % of biorefineries), followed by France (9 %), Spain and Brazil (7 %), whereas bioethanol in the US and biodiesel in Europe dominate bio-production supported by favourable economic policies. At TRL 1, cost-capacity curves are developed for 11 biorefineries that allow estimation by power law relationships and statistical metrics to evaluate uncertainties within 95 % certainty range. New jobs vs capital cost curves are presented to estimate the social impact of future biorefineries in respect to job growth. The analysis shows that 1st gen ethanol remains more cost-efficient than advanced biofuels, whereas synergies among existing plants can largely reduce capital costs. 2nd gen ethanol biorefineries make the most positive impact by creating 1 new job per million euro invested. At TRL 2-3, biorefinery development is at a phase that allows application of the “Process Blocks Build-Up” estimating method; it is a modular approach which proposes probabilistic cost exponents and reference costs for segments found in 11 biorefinery types. The method addresses estimation of biorefineries beyond the state-of-the-art under the assumption that a new biorefinery is composed of blocks that resemble those found in state-of-the-art biorefineries. Differences in performance are accounted for with an efficiency factor for each block. Validation of the method shows it has an absolute average error of 27 %. At TRL 4, process development justifies a conceptual flow diagram and thus, the capital cost can be estimated from “Functional Unit” relationships, which postulate the capital cost as a function of the number of functional units and basic process parameters. The concept of the functional unit - established on petrochemical processes – is re-defined through examples to include unit operations commonly found in biorefineries. The percentage standard error of the estimate is used to model the 95 % certainty around the base estimate. Despite the advanced phase of development, the new correlations report an absolute average error of 50 %. The discrepancies are due to the uncertain and limited amount of data used in regression analysis. At TRL 5, installation factors (Lang factors) under uncertainty are proposed assuming the delivered equipment cost can be estimated. Factorial estimating is at the verge of early-estimating and reports an average absolute error of 12 %. The proposed installation factors (solid-fluid: 5.9, fluid: 9.5) are higher than the ones reported in common engineering handbooks, thus, confirming the research hypothesis that the offsites in biorefineries are capital intensive. The thesis introduces “Production cost estimating with uncertainty”, which assumes that the production cost elements can be expressed as a function of the raw materials, the utilities, the operating labour and the fixed capital investment and proposes modelling factors under uncertainty for each element. The cost models are validated across TRLs on two case studies; PHB from methane fermentation and FAME from in planta transesterification of castor seeds. The modular approach assists in estimating processes at low TRL based on little information and provides reasonable results. The short-cuts show good agreement with detailed techno-economic results from common engineering software (Aspen Process Economic Analyser). The cost models can be used interchangeably for rapid cost estimations and modelling work for various purposes.	en
heal.sponsor	The EC (Reneseng-607415 FP7-PEOPLE-2013-ITN) is gratefully acknowledged for supporting this work.	en
heal.advisorName	Dubois, Jean-Luc	en
heal.advisorName	Κοκόσης, Αντώνης	el
heal.committeeMemberName	Διακουλάκη, Δανάη	el
heal.committeeMemberName	Καλογήρου, Ιωάννης	el
heal.committeeMemberName	Λεμονίδου, Αγγελική	el
heal.committeeMemberName	Δέτση, Αναστασία	el
heal.committeeMemberName	Μαυρωτάς, Γιώργος	el
heal.committeeMemberName	Κροκίδα, Μαγδαληνή	el
heal.academicPublisher	Σχολή Χημικών Μηχανικών	el
heal.academicPublisherID	ntua
heal.numberOfPages	422
heal.fullTextAvailability	true