Università di Pavia, Facoltà di Ingegneria

1. Università di Pavia, Facoltà di Ingegneria SISTEMI DI TECNOLOGIE ENERGETICHE E MODELLI DI PROGRAMMAZIONE ECONOMICA G.C. Tosato – <giancarlo.tosato@efda.org> Pavia, 13 Maggio 2002- Dipartimento Ingegneria Elettrica, Aula Seminari, Piano C, - via Ferrata, 1

2. ARGOMENTI • SETTORE ENERGETICO: PROBLEMATICHE - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - • DESCRIZIONE QUANTITATIVA: ELEMENTI E VARIABILI DEL SISTEMA, INVENTARI • SISTEMA: IDENTIFICAZIONE DELLE CORRELAZIONI, SPIEGAZIONE DEGLI EVENTI • MODELLI: RAPPRESENTAZIONE DEL SISTEMA, PROIEZIONI E VALUTAZIONI - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - • GENERATORI DI MODELLI TECNOLOGICI • DOMANDE; ESEMPI: Modelli e scenari, Confronto di opzioni tecnologiche, Valutazione di politiche e misure

3. 1 - Settore energia: domande strutturali • Che problemi ci riserva nel suo sviluppo? Riserve sufficienti? Costi compatibili? Inquinamento? Sviluppo sostenibile o no? Equità intergenerazionale? • Quanto siamo lontani dalla sostenibilità? • Quanto costa mitigare i cambiamenti climatici? • Che distanza fra prezzi attuali e costi sostenibili? • Quanto e come investire per cambiare rotta? ricerca? • Quale equilibrio di lungo termine tra sistema energetico e resto del sistema economico? Tra paesi più e meno sviluppati? Equità internazionale?

4. 2a - Nodi: tecnologie energetiche Quale spazio per migliorare l’efficienza / il costo di produzione? (ingegnere, tecnico) • Settori • Parametri di caratterizzazione • Fonti di DB tecnologici • Esempio: IKARUS (Germany) • Essempio: EM (World Bank)

5. 5 Settori ENERGY TECHNOLOGIES RESOURCES PRODUCTION (Supply) ENDUSE (Demand) Mining Residential/Service Transformation Industry Conversion Transportation Transport & Distr Cross - sectors

6. Parametri di caratterizzazione

7. Esempio di caratterizzazione

8. Data Bases IKARUS - Germany: + LP optimization / sectoral simul.models CO2DB - IIASA, Vienna: linked to simulation models EM - World Bank: linked to simulation models IPCC/TI - USA: mitigation technologies inventory DECADES - IAEA, Vienna: elc tch + energy chains GREENTIE - IEA, Paris: DB on technology providers MARKAL country DBs, linked to optimisation models other in UK, NL, USA, etc.

9. 9 Struttura dei dati in IKARUS Esempio: IKARUS, Germania Processo (tecnologia) Schema di impianto Dati relativi alla tecnologia selezionata

10. 10 EM, World Bank Data structure in EM Products (fuels) Processes (technologies) Emissioni Composition (fuels) Costs (technologies) Costs (fuels)

11. 2b - Flussi: statistiche e bilanci energetici Quanto costa l’energia alla nazione? Quanto inquina? (statistico, economista) • Statistiche energetiche nazionali (enr_ita2.xls) • Statistiche energetiche (ex: IEA, ENERDATA) • Bilancio Energetico Nazionale (MAP/DGERM) • Bilanci Energetici di sintesi (ex: IEA) • Bilanci energetici di dettaglio (ex: IEA) • Matrici Input / Output (generali, energetiche)

12. DB statistiche energetiche • base content of most energy data bases: • socio - economic - land use data, maps (?) • energy balances, flows data, prices, indicators, • balances, capacity and reserves, etc. • energy related environment data, indicators, • by country (or region), year, fuel, sector, unit, etc • main providers (free or at some cost) • International Energy Agency, Paris • Energy Information Administration, Washington • EUROSTAT, Luxemburg • ENERDATA s.a., Grenoble, France

13. 3 - Correlazioni nel sistema Quale fenomeno spiega la situazione attuale, in termini di flussi, tecnologie, etc.? (economista dell’energia) • Analisi econometriche • Macro • Micro • Analisi dei fattori • Globale • Settoriale • Reference energy system: • Bilanci dei flussi di energia e materiali • Bilanci delle tecnologie

14. 3a -Relazione econometrica base • Ln(TPES(t)) = • A + • B * Ln(GDP(t)) - • C * ln (Pr(t)) + • D * t • TPES = Consumo naz. Di energia primaria eq. • GDP = prodotto interno lordo • Pr = prezzo medio dell’energia • T = anno; B, C = elasticità; D = miglioram. tecn

15. 15 3b -Analisi dei fattori: identità di Kaya G=Pop*(Gdp/Pop)*(En/Gdp)*(Fos/En)*(CO2/Fos) = P * W * E * F * C EnergyService Carbon Intensity G : CO2 Emissions P : Population W : per capita Gross Domestic ProductE : energy intensity of the GDP F : share of fossil fuels on Total Primary Energy Supply C : carbon intensity of fossil fuel mix

16. 16 3b -Analisi dei fattori settoriale (IEA,99) G = k(wk*Ak * i (Si * Ii *  jFij)) EnergyService Carbon Intensity G : CO2 Emissionsw : Weighting factor: 1990 emissions in sector k A : Activity in Sector k S : Structure in Sector k (sub-sector i share of sector activity) I : Energy Intensity in sub-sector iF : Carbon Content of fuel j used in sub-sector iRepresents Changes in: 1) supply efficiency, 2) supply fuel mix 3) end-use fuel switch

17. 17 3b - Analisi dei fattori, esempio (IEA,99)

18. 3c - Reference Energy System Imports Exports Other regions Industrial Electricity Residential O&Gas Proc. & Refining Commercial Other Processing Transport Domestic Non-Energy Sources Processing (*n) EndUse-Tech EU-Demands

19. 4 – Modelli di previsione Modelli qualitativi/ quantitativi (story lines vs. scenarios) Caratteristiche distintive dei modelli energetici quantitativi: - contenuto(scope) - materia: settoriale o globale; geografico: locale, nazionale, regionale, mondiale - approccio teorico (tipo di variabili e di equazioni) top-down (macro) vs. bottom-up (tecnologico) - orizzonte temporale: breve, medio, lungo

20. 20 4 - Alcune categorie di modelli Top-down, econometric Short term Long term General equilibrium Auto- regressive Sectoral Macro- economic Sectoral/ technology End use models Simulation Optimization Bottom-up, engineering

21. 5 – MARKAL: Diagramma di funzionamento Technological database Economic Scenario Environmental Scenario • Base Case Demands for energy services • Demand Elasticities • Oil Price MARKAL • Cap(s)-&-Trade • Taxes, Subsidies • Sectors’ Measures Equilibrium • Technology Investments and Market Shares • Emission Trajectories • Adjusted Demands for energy services • Marginal Values of Energy Forms (Prices)

22. 5 – MARKAL: Variabili principali • Mining/Import/Export (r, t, c, ts) • Investment (r, t, p) • Capacity (r, t, p) • Operation (r, t, p) • Demand Loss/Increase (r, t, dm, k)

23. 5 – MARKAL: Equazioni principali, 1 • Demand (r, t, dm) • Production from all “related” end-use technologies + Elastic variables End-use demand • Commodity balance (r, t, c) prices • Production  Consumption • Process activity (r, t, p) • Operation  Capacity  Availability Factor • Capacity transfer (r, t, p) • Capacity = Investments + “Residual” capacity

24. 5 – MARKAL: Equazioni principali, 2 • Electricity sector • Time-sliced balance (r, t, ts) • Production  Consumption • Seasonal reservoir management • Essentially lets you specify seasonal plus an annual availability factor • Peaking • Total ELC capacity  (1+ERESERVE)  Capacity needed to meet the energy requirement • Base load • Total night production of Base-Load-Techs  Base-load-fraction  total night demand

25. 5 – MARKAL: Equazioni principali, 3 • User defined constraints • Any of the variables can be used to define a new constraint • Salvage • The investment cost of “unused” technology stock is refunded • Objective function • NPV of (tech costs + mining/import costs - export revenues + taxes - subsidies)

26. 5 – MARKAL - MACRO Energy Costs Energy Y Consumption MACRO MARKAL Labor Capital Investment

27. 5 – Variabili MACRO • Utility • Consumption (t) • Investment (t) • Energy costs (t) • Production (t) [Excluding energy sector] • Capital stock (t) • End-use demand (t, dm)

28. 5 – Equazioni MACRO • Utility •  Disc-fact Log(Consumption) • Use • Production = Consumption + Investment + Energy Costs • Production • Production = f(Energy, Capital, Labor) • Capital accumulation

29. 6a – Scenari nazionali: energia

30. 6a – Scenari nazionali: emissioni

31. 6b - Costi complessivi di mitigazione in Italia(GDP 0.5-1.0% al 2010, tassa eq. 80-250 $/tCO2)

32. 6b -Opzioni di mitigazione ottimali: sintesi

33. 6c - Politiche e misure di mitigazione: naz. Economic Instruments: Liberalised Internal Market in electricity and gas, Carbon tax, Subsidies (-, +), Legal instruments: Codes to improve the thermal insulation of buildings, Minimum Efficiency standards for end use devices including cars, IPPC, traffic restrictions, Voluntary agreements: RUE in industry/municipalities, Energy audits in industry, services, buildings, Phase out of the less efficient end use devices, .. Diffusion of information: Monitoring mechanism, Energy Labels of end use devices, Direct investments: Energy RD&D, procurement,… Other

34. 6c - Potenziale dei meccanismi flessibili Evaluation tool: Markal, an energy technology based shadow price generator The model of the Italian energy system (15000 eq.) is run first with the Indian model (+5000 equations), then with the model of USA + Canada (+60 Keq) In the base scenarios, the total systems costs are minimised. Other scenarios: with/without the Kyoto target to 2030, Italy can/cannot purchase emission permits from India, excluding / including purchases from USA, Canada. Project based analysis (Clean Development Mechanism) calculating the marginal system cost (strategic elements to guess which might be the price of Emission Permits)

35. 6c - FlexMex:punto di vista del venditore CO2 emissions in India in 1995: 760 MtCO2/y, and a strong increase is expected: +100% in 2015, +200% in 2030 If the trade of emission rights is permitted, these emissions reduce by 10% in 2010 and by 20-25% in 2020-2030 at a marginal price of 25-35 US$’96/tCO2, by investing in Gas CC power plants instead of Coal PP (the concepts of baseline and additionality) The extra energy system costs are about 1.7% (investments), the economic surplus of selling on the emission trade market these CO2 reduction units is double (3.6% of the total s.cost) CDM or Trade of Emission Permits?

36. 6c - FlexMex: punto di vista del compratore Which part of the national commitments is worth buying? 67% at 10-12 US$’96/tCO2 (India->Italy exchange) 52% at 25-35 US$’96/tCO2 (India->USA-Canada-Italy) 33% at 40-80 US$’96/tCO2 (Eu proposed ceiling) 0% at about 200 US$’96/tCO2

37. 6 - Studi recenti • Impact of different schemes for international flexible mechanism for mitigation (Can-US-India-It, CH-Columbia, NL-CH-SW, Nordic Eu countries) • Environmental effects of reducing/removing energy subsidies or of adding a Carbon Tax (Italy, Australia) • National MARKAL models contribute to identify climate change mitigation options and the evaluation of climate change policies (Can, National Communic. to FCCC of Aus, Be, Cz, It, Latvia, NL, Sw, Us) • Effect of including in the energy system materials, full fuel cycle analysis and endogenous technology learning for mitigation strategies

38. 6 - Modelli globali multiregionali • US MARKAL+MERGE for other regions (DWI-Stanford) • The Global Markal Macro Trade model with Endogenous Technology Learning (PSI-IIASA) • The SAGE Project at the US Energy Information Administration (System to Analyse Global Energy markets) with Time Stepped Technology Learning • The IEA Energy Technologies Perspectives Project adds technological insight to 2002 World Energy Outlook with a bottom-up multiregional global model • Multi-regional Global TIMES (NRCanada, GERAD) • Long time horizon multiregional global model for SERF3 (Socio Economic Research on Fusion)

39. 6 - Energia e ambiente su scala locale Under ETSAP and ALEP (Advanced Local Energy Planning, IEA IA on Energy Conservation in Buildings and Community Systems) District energy grids expansion, waste management, local pollution vs global mitigation (local Agenda 21), energy conservation policies in buildings, public vs private transportation, local tax/subsidies • Germany (Mannheim, • Italy (Bologna, Torino, Aosta, Basilicata, • The Netherlands (Delft, • Sweden (Jancheping, Norcheping, • Switzerland (Geneve, • China (Hong Kong, etc.

40. 7 - Il progetto ETSAP/IEA • Present participants: Australia, Austria, Belgium, Canada, EU, Finland, Germany, Greece, Irland, Italy, Japan, Korea, The Netherlands, Norway, Spain, Sweden, Switzerland, Turkey, United Kingdom, United States • Main goals: - to develop modelling tools that represent different systems of energy flows and technologies - to improve the knowledge of global, regional, national and local energy and environment systems - to contribute to the energy environment debate with quantitative and methodologically sound analyses

41. 7 - Generatore di modelli MARKAL (1979) (sometimes called marginal price generating model) • fixed multi-time periods Pareto Optimal models minimising the discounted total system cost • energy flows and technologies (energy system only) • single region • with price sensitive supply curves, non price sensitive demand curves expressed in final/useful energy terms • time/Reference Energy System (RES) perfect foresight • calculated trade-off curves among cost, energy security and emissions • 2 versions of the code: FORTRAN and OMNI

42. 7 - Versioni del MARKAL (1980-2002) • MARKAL Materials, can represent tens of pollutants and in principle the whole economy • MACRO, NLP general equilibrium version, with a single production function • MICRO, NLP partial equilibrium version, with demand elastic to prices (own/cross) • Stochastic to calculate optimal hedging strategies • Elastic Demand, linearised partial equilibrium version • Multi-regional, with endogenous trade • Endogenous Technology Learning (MIP)

43. 7 - Evoluzione del Software • automatic creation of a reference energy system from the IEA energy balances (TEMPLATE spreadsheets) • projections of the demand for energy services from common drivers and own price elasticities • common technologies repository, with efficiencies, emissions, materials use, costs and learning of existing and new technologies • multiple model shells (MUSS, Answer, VEDA FE) • multiple model generator programs (GAMS) • multiple LP, NLP, MIP solvers • multiple / flexible reporting tools (VEDA BE, Answer)

44. 7 - MARKAL/TIMES oggi • variable time periods length (TIMES) competitive partial equilibrium models maximising the discounted sum of the consumer and producer surplus • models flow and technologies of energy systems + materials + wastes + pollutants + other sectors • multi-grids, multi-regional with endogenous trade • price sensitive supply and demand curves (free units) • clairvoyant, or stochastic, or time-stepped • In each market technologies with the best marginal benefit / cost ratio are chosen, including externalities • coded in GAMS

45. 7 - Equilibrio statico • hundreds of energy good and commodity markets are represented (from coal to passenger*km) • the stepwise supply and demand curves of each market are calculated including independent investment, fixed, variable, fuel, environment, material costs (based upon separate variables) • the equilibrium point in each market of the RES calculates Quantities, Prices and indicates both supply and demand marginal technologies • the distance from competitiveness of each technology and the technology gaps for reaching the desired equilibrium points are calculated by the model

46. 7 - Equilibrio dinamico The reaction of the energy system to exogenous dynamic changes is represented in the model through • starting point is the present stock of technologies and the possible future availability of well defined technologies (not upon past behaviour) • substitution among competing and time improving processes commodities, similar to the mechanism of optimal Von Neumann multiple producers/multiple commodities I/O models (not through price dependent technical coefficients of a Leontief I/O square matrix) • variable depreciation plans for investments

47. 7 - Visita i siti web del gruppo … • www.iea.org/ • www.ecn.nl/unit_bs/etsap/main.html • www.crt.umontreal.ca/~amit/THEMODEL/ • www.ier.uni-stuttgart.de/ • www.kier.re.kr/ • www. abare.gov.au/ • www.tokai.jaeri.go.jp/ • etc.

48. .. o contatta i ricercatori del gruppo • Fridtjof Unander <fridtjof.unander@iea.org>, IEA desk officer • Prof. <Clas-Otto.WENE@iea.org> • <Dolf.GIELEN@iea.org> • Prof. Enzo Cuomo <cuomo@unibas.it> • <Emmanuel.Fragniere@hec.unil.ch> • Gary Goldstein <ggoldstein@irgltd.com>, SW coordinator • Prof Alain B. Haurie <haurie@ibm.unige.ch> • Amit Kanudia <amit@crt.umontreal.ca> • Barry Kapilow-Cohen <b.cohen@eia.doe.gov' • Anna Krook <annak@entek.chalmers.se> • Socrates Kypreos <socrates.kypreos@psi.ch> • <Denise.VanRegemorter@econ.kuleuven.ac.be>

49. .. altri ricercatori del gruppo • Prof. Evasio Lavagno <lame@polito.it> • John C. Lee <jcl@bnl.gov> • Prof. Richard <loulou@management.mcgill.ca> • Prof. Alan S. Manne <asmanne@attglobal.net> • Ken Noble <knoble@abare.gov.au> • Osamu Sato <sato@ruby.tokai.jaeri.go.jp> • Chris Schlenzig <cs@ier.uni-stuttgart.de> • Heesung Shin <hsshin@kier.re.kr> • Prof. Shukla <shukla@iimahd.ernet.in> • Peter Taylor <peter.g.taylor@aeat.co.uk> • GianCarlo Tosato <gctosato@tiscalinet.it>, project head • Phillip Tseng <phil.tseng@hq.doe.gov>, chairman

50. TITOLO 1 • TESTO 1