M3: Energy Transition

The Energy Transition model captures the question “How fast can we substitute from one source of primary energy to another?”. For each transition, our “belief” is a roadmap, a function that tells for each year which share of energy consumption may be transferred to another source. Since there are four kinds of primary energy in the CCEM model, and since we assume transitions to be oriented (a simplifying assumption), there are six transitions to consider: Coal to Oil (using CTL techniques), Coal to Gas (which we have seen a fair amount in the US during the last decade), Coal to Clean, Oil to Gas, Oil to Clean, and Gas to Clean.

There are many constraints that make this energy transition difficult and costly. Energy sources have different uses with different constraints (such as mobility, intermittence, etc.) which yields the use of secondary sources of energy, also called “vectors” (electricity, hydrogen, …).  The next figure is a very simplified illustration which illustrates why some substitutions are easier than other. Substitutions require time and investment. Therefore, they are represented in M3 as a “belief”, a transition roadmap for each zone that says, for each of the fix transition (A →B), which share of A’s consumption may be transformed into B. The model will compute the actual level of substitution achieved for a given year and will also generate the requested “energy investments”. As we shall discuss in Section 4.1, Energy Transition is a critical belief and one where there is a huge difference between the techno-optimists who believe that electrification of energy can be pushed forward very fast, and the “realists” who see a lot of viscosity in the transfers represented in the figure displayed in the previous slide illustration.

M3 uses the following state variables to further describe the energy system:

•   Pe(y): price in $ for 1 toe for energy e, at year y

•   Uz(e,y): usage (constrained consumption) for zone z of energy e

•              Sz(y): percentage of savings reached at year y

•   CNz(y): percentage of consumption canceled in zone z at year y, because the price is too high

•   IEz(y) : investments for new energy capacity for energy source z at year y

•   SP(y): steel price for year y

The input of M3 are the demand and supply price-vectors computed by M1 and M2, the transition matrix (transitionRate(z,e1,e2, y), which is the core “belief” of M3), and a parameter that describes the decline of energy transformation investments in time, as technology improves. The last table that we use as an input in M3 is the CO2 tax table, for each region, that sets the CO2 tax level as a function of the CO2 concentration that has been reached. M3 may be described with the following numbered equations.

(1)    The first step is to compute the constrained energy consumption Uz(e,y) for every energy source e and every zone z. We apply the cancellation factor associated to the sell price. By construction the sum of the demands matches the sum of the consumption for each energy source.

(2)    We compute the part of the dematerialization (M2) that is linked to voluntary efficiency savings. The saving ratio Sz(y) is computed from the desired level (cf M5) modulo the constraint on max yearly growth

(3)    We then compute the new transfer levels Tr(e1,e2,y), for each 6 transition from one source e1 to e2.

(4)    The sum of CO2 taxes is derived for each zone through the sum of multiplying the consumption of energy e by the CO2 ratio (g/KWh) for each energy source.

(5)    M3 records all necessary investments IEz(y) for energy capacity growth, energy savings and energy transfers. Notice that the price of steel, which is produced in M4, is used to evaluate the costs of green energy growth.

(6) CCEM computes an approximation of the electrification factor, through a heuristic estimate of how much of energy source e usage is used through electricity as a vector (100% for Clean). We see that if heat%(e1,e2) = 1 (100% heat to heat), there is no gain in electrification, but if heat%(e1,e2) = 0, all energy transfer is electricity since the net gain ratio is (1 – elec%(e1)), since we make the simplifiying assumption of homogeneous transfer.

(7)    Last, we compute the CO2 emissions for zone e, using an equation similar to (4). Because we track all fossil energy use, this is a simple formula that covers use of fossil fuels in industry (cement, steel, etc.).

Producing the energy transition matrix is a big task (even with only 6 transitions), fortunately historical data may help for the calibration. These equations used additional parametric functions that represents the “known unknown” associated to M3:

•   transitionRate(z,e1,e2, y): maximum transfer of energy needs from primary source e1 to e2 at year y, expressed as a percent

•   techEfficiency(z): yearly growth of tech efficiency (cost reduction in investment to build a production capacity)

•   invest Price(e): investment that is necessary to build a capacity of 1Gtoe/y at year 1

•   ftech(z): expected yearly decline of investPrice in zone z (technology progress)

•   steelFactor(e): part of steel cost in total cost of investment for e

•   eRatio(e,s) : fraction of energy e consumption for zone z (year 1) that is used for electricity

•   elec%(e) : fraction of energy source e that is used to produce electricity at year 1

•   heat%(e1,e2): when we transition energy consumption from source e1 to e2, fraction of that energy that was used without  electricity (such as heat) that is converted to another non-electric usage (heat to heat is a good example).