|
Article Excerpt
Concerns about the cost of C[O.sub.2] capture and sequestration, and the effectiveness of carbon abatement policies loom large in discussions on climate change mitigation. Several writers address the issue from various perspectives. This paper attempts to add relative realism to discussions on C[O.sub.2] capture costs, and, the deployment of carbon capture technology in the UK by using publicly available company data on the long term capacity expansion and C[O.sub.2] capture investment programmes of selected power plants in the UK. With an estimated 8 [pounds sterling] billion plan to install a generation capacity of 11 GW and capture capability of 44 MtC[O.sub.2]/year, it is imperative to optimise this huge potential investment. A least-cost optimisation model was formulated and solved with the LP algorithm available in GAMS. The model was then applied to address a number of issues, including the choice of an optimal carbon abatement policy within the EU-ETS framework. The major findings of the study include (a) the long term total cost curve of C[O.sub.2] capture has three phases--rising, plateau, rising; (b) alternative capture technologies do not have permanent relative cost advantages or disadvantages; (c) Government incentives encourage carbon capture and the avoidance of emission penalty charges; and (d) the goals of EU-ETS are more effectively realised with deeper cuts in the EUA ratios than merely hiking the emission penalty, as proposed in EU-ETS Phase II.
1. INTRODUCTION
Several policy options have been proposed and/or implemented to meet the emission compliance targets inspired by the internationally-agreed Kyoto Protocol. (1) Thus, in the UK as in the rest of the European Union (EU), the Emissions Trading Scheme (EU-ETS) commenced operations in January 2005. (2) The EUETS is a cap-and-trade scheme in which C[O.sub.2] emission limits are set for qualifying large emitters according to the National Allocation Plans (NAPs) (3) of member countries. Emission allowances (EUAs) (4) equal to the set emission limits are allocated virtually free (5) to each regulated emitter. From April 2006, the allowances have to be "surrendered" or "delivered" in annual returns, in which the actual and traded C[O.sub.2] emission levels are compared with the amount allocated. The amount of the non-delivered emission (NDE) (6) allowance must be zero at the time of filing a return, otherwise a company is liable to pay an emission penalty, currently fixed at 40 [euro] per tonne of C[O.sub.2]. (7)
While it may be too early to make definitive evaluative statements on the performance or effectiveness of the EU-ETS 1, the emerging consensus appears to be that the current NAP Allocations (2005-2007) are overly generous both in terms of their free nature and volume. This has led to (a) the creation of so-called windfall profits from the "intrinsic" rents in the free EUAs (8); (b) considerable carbon price volatility; and, (c) a generally low compliance cost that has, at best, neither significantly curbed carbon emissions nor seriously encouraged carbon capture and sequestration (CCS). Rather, in a worst-case scenario, the Scheme has perversely induced greater C[O.sub.2] emissions (Ellerman, 2006).
Consequently, several writers including Berlin (2007) argue that more stringent emission caps are needed to raise the compliance cost and the carbon price in order to increase the attractiveness of investments in CCS technologies.
Beyond advocacy, the emerging evidence points in the direction of stricter emission limits. Thus, the EU Commission reduced Germany's total allocation of C[O.sub.2] certificates by 8% from 495 mt/C[O.sub.2] per annum in NAP 1 (National Allocation Plan Phase 1) to 453 mt/C[O.sub.2] per annum in NAP 2. Correspondingly, the UK's total allocation was reduced by 4% from 246 to 237 mt/C[O.sub.2], while the country's electricity sector had a much bigger reduction of 20% (DEFRA, 2007).
The present study proposes a methodology for determining the least-cost options of introducing carbon capture technology (9) under the overarching assumptions of (a) increasingly stringent emission caps on fossil-fuelled power plants; and, (b) expanding carbon trading in the EU-ETS market. The approach entails formulating and solving an optimisation model with clearly stated goals, and explicit provisions for the various regulatory, technological and market conditions which offer opportunities and/or restrict corporate decision-making and action-taking.
2. A GENERALISED MODEL OF CARBON CAPTURE
In a carbon abatement regime with sufficiently stringent emission caps, the power plants in a region or country would each face a multi-objective cost function to be minimised subject to technical, market and regulatory constraints.
2.1 The objective Cost Function
According to Pelster et. al (2001), the cost function to be minimised in the circumstance is environomic to the extent that it combines economic, environmental and thermodynamic considerations. Assuming the power plants use different power generation and C[O.sub.2] capture technologies, the aggregated objective function to minimise the present value of the future environomic cost can be written as:
[C.sub.t] = [k.sub.t] ([summation][summation][a.sub.it][x.sub.it] + [summation][summation][b.sub.it][u.sub.it]) + [summation][summation][f.sub.it][y.sub.it] + [summation][summation][e.sub.it][y.sub.it] + ([summation][summation] [m.sub.it][v.sub.it] - [summation][summation][h.sub.it][q.sub.it]) - [summation][summation][g.sub.it]([q.sub.it])/[(1 + r).sup.t] (1)
where:
i = different power generation and C[O.sub.2] capture technologies (i=1, 2, .I)
[k.sub.t] = industry-wide capital recovery factor at time t
[a.sub.it] = unit CAPEX of the core power generating plant type i at time t
[x.sub.it] = effective electricity generating capacity of plant type i at time t
[b.sub.it] = unit CAPEX of the C[O.sub.2] capture equipment of plant type i at time t
[u.sub.it] = installed C[O.sub.2] capture capacity in plant type i at time t
[f.sub.it] = unit fuel OPEX of plant type i at time t
[y.sub.it] = the operating level (or output) of plant type i at time t
[e.sub.it] = unit non-fuel OPEX of plant type i at time t
[h.sub.it] = unit C[O.sub.2] capture OPEX of plant type i at time t
[q.sub.it] = amount of C[O.sub.2] captured in plant type i at time t
[m.sub.it] = unit emission penalty cost to plant type i at time t
[v.sub.it] = excess C[O.sub.2] emission in plant type i at time t
[g.sub.it] = unit Government intervention (tax or subsidy) rate in plant type i at time t
r = discount rate
t = time in years
Equation (1) combines the power stations' new-build or retrofit investments in electricity generation and C[O.sub.2] capture with Government incentives (see below). The various components of the objective function are discussed below in section 2.1.1 through 2.1.4.
2.1.1 The Capital Investment
(a) The core generating plant
The index i in equation (1) refers to the four types of power plants--namely, Pulverized coal (PC), Combined Cycle Gas Turbine (CCGT), oxyfuel-based and Integrated Gas Combined Cycle (IGCC)--and two types of boilers (sub- and super-critical) that are available for the deployment of CCS technology, using either pre- or post-combustion capture technologies (IPCC, 2005).
Given particular electricity generation and carbon capture technologies, the investment cost of the [i.sub.th] plant type at time t consists of the capital expenditures (CAPEX) incurred on generation ([x.sub.it]) and C[O.sub.2] capture ([u.sub.it]) capacities. Owing to the unavoidable barriers to full capacity utilisation, it is customary to distinguish between the nameplate (or notional) and effectively-used capacity of a power plant. Recent IEA GHG studies claim the levelised capacity factor is about 85 per cent (IEA-GHG, 2006). The capacity factor is likely to be further eroded by the anticipated efficiency losses due to the parasitic effect of C[O.sub.2] capture. (10) According to some estimates, the C[O.sub.2] capture process requires between 10 and 40 percent extra energy (or fuel) to produce the same net energy as the equivalent without-capture reference plant. (11) These losses recommend the application of CCS technology only to high efficiency plants in which the efficiency penalties are lower and less costly (see Wall, 2007 and Drax, 2005).
In order to capture the required increase in the nameplate capacity needed to compensate for the efficiency losses, the cost of the effective generation capacity ([a.sub.it] [x.sub.it]) is defined in the present study as:
[a.sub.it][x.sub.it] = [a.sub.it] [x'.sub.it] (1 + [l.sub.it]) (2)
where:
[x'.sub.it] = the nameplate capacity of plant type i at time t (MW)
[l.sub.it] = efficiency degradation due to C[O.sub.2] capture (%)
(b) The C[O.sub.2] capture capacity
A power plant desirous of adding CCS technology to its electricity generation process has a number of options. It can either invest in a new build power plant with integrated or built-in C[O.sub.2] capture facilities, or it can simply add-on a C[O.sub.2] capture system with retrofitted or re-built boilers and turbines which are designed to increase plant efficiency and output (Mitsui Babcock, 2006). There are three leading combustion technologies (oxyfuel, pre- and post- combustion) and one gasification technology often considered in the literature for C[O.sub.2] capture. These technologies are at different stages of development, deployment, and commercialization. Whichever option is chosen will result in capital expenditure (CAPEX ([b.sub.it] [u.sub.it]) which is additional to those incurred in the core electricity generating plant.
(c) The capital recovery factor
The third element in the capital investment component of the objective function is the capital recovery factor (CRF) which is defined as:
[k.sub.t] = r [(1 + r).sup.n]/[(1 + r).sup.n] - 1 (3)
where, in addition to previous definitions:
n = power plant lifetime
2.1.2 The OPEX
The OPEX (operating expenses) of the power plants are the sum of the fuel, non-fuel and C[O.sub.2] capture costs (excluding capital costs) or Operations and Maintenance (O+M) costs. At any given time, the aggregate fuel ([summation][summation] [f.sub.it] [y.sub.it]) and non-fuel ([summation][summation] [e.sub.it][y.sub.it]) costs both depend on the amount of electricity ([y.sub.it]) produced. In turn, power output is a function of the amount of fuel inputs ([f.sub.it]) as well as the load (L) factor, which is determined by the level of plant availability and the power price-marginal cost ratio. That is,
[L.sub.it] = [phi] ([A.sub.it], [P.sub.et]/[MC.sub.et] (4)
with the condition that [P.sub.et] > [MC.sub.et]
where,
[A.sub.it] = plant availability of plant type i at time t
[P.sub.et] = price of electricity at time t
[MC.sub.e] = marginal cost of electricity at time t
Plant availability ([A.sub.it]) is the amount of time the plant is capable of generating electricity, after deducting planned and forced outages. The marginal cost of electricity ([MC.sub.et]) comprises mainly of fuel, non-fuel and C[O.sub.2] capture costs. The profitability criterion ([P.sub.et] > [MC.sub.et]) is that the price of electricity must be greater than the marginal cost of electricity generation and C[O.sub.2] capture. In the medium term, both the price and marginal cost are expected to decline owing to the interplay of several factors including economies-of-scale and improvements in thermal efficiency, resulting from the "cumulative experience" or "learning by doing" (LBD) effect. The potential of declining medium-term marginal cost induced by improvements in the plant thermal efficiency is proxied in the present study by assuming a declining fuel cost per kilowatt-hour of electricity generated. This is consistent with the approach adopted by IEA-GHG (2006).
2.1.3. The Cost of Net Emission Reduction (NER)
A power plant whose carbon emissions at a given output level exceeds...
|
