(431e) Evaluation of Thermophotovoltaics As a Decarbonization and Clean Energy Storage Technology: Techno-Economic Analysis and Uncertainty Assessment

Thermophotovoltaics (TPV) is an emerging decarbonization and renewable energy technology that converts heat to electricity. Figure 1 shows the TPV process schematic. It requires a heat source that can be any of these four- solar, chemical, nuclear, or electrical. The heat from these sources (in the range of 1300-2500 K) or waste heat sources reaches the thermal emitter, which converts it into radiation. The presence of a thermal emitter is the difference between a solar PV and a TPV system. The emitter emits a more suitable radiation spectrum (compared to the Sun in the PV system) that reaches the PV cell, which converts the part of incoming photons having higher energy than the PV cell bandgap into electricity. The photons having less energy than the bandgap of the PV cell (sub-bandgap photons) are not useful and do not convert to electricity but overheat the TPV cell, decrease the emitter temperature and heat-to-radiation efficiency, and thus reduce the overall efficiency.

TPV is increasingly becoming popular as an energy conversion pathway due to its numerous applications, some of which include energy storage, combined heat and power (CHP), waste energy recovery, and in-space programs. It can be used for CHP applications as it provides electricity and can be used to store heat due to its compatibility with storage applications. Due to its higher power density (>2.5W/cm²) and reduced cost, there has been an increasing focus on energy storage using TPV, called thermal energy electrical storage (TEES) or thermal energy grid storage (TEGS), which is competitive with electrochemical batteries, and is suitable for long-duration energy storage applications.

In this study, we consider an energy storage model by Datas et al. (2019)[1] (Figure 2). It includes three energy sources: solar PV, grid, and natural gas; a boiler (running on natural gas); two heat storages: low-temperature energy storage (LTES) and high-temperature energy storage (HTES); a power generation unit (PGU, here, it is TPV); and a heat pump (either electric or thermal). It computes the energy going into each storage based on the interplay of electric and heat demand and generation.

We calculate and optimize the levelized cost of consumed energy (LCOE) and electricity (LCOE_el) for a TPV energy storage system for a building in Boone (Iowa) using an energy management algorithm (along with our modifications) in Python. For optimization, we vary the system size- nominal PV capacity, HTES maximum energy capacity and PGU maximum energy capacity. LCOE represents the cost of supplying heat and electricity; however, LCOE_el only considers electricity costs. We perform a sensitivity analysis and Monte Carlo uncertainty assessment to better model the uncertainty in these input parameters- lifetime, nominal weighted average cost of capital (WACC), PGU/TPV efficiency, electricity price, PGU CAPEX (capital cost factor), PV CAPEX, HTES CAPEX, inflation rate, and natural gas price. The Monte Carlo method is an analysis tool that uses random numbers (generated using a fitted probability distribution for each parameter from literature data) to calculate the output for a modeled stochastic process.

The results indicate that after optimization (and using the energy management algorithm), LCOE reduces, and the self-consumption ratio (SCR, representing the amount of solar PV energy being utilized for energy storage and electricity supply) of solar PV increases from 57.2% to 61.7%. Less PV energy is lost, and the consumer utilizes more. For both LCOE and LCOE_el, the optimum system converges to the smallest system bounds: nominal PV capacity = 5 kW_el, HTES max. capacity = 15 kWh_th, PGU max. generation capacity = 0.5 kW_el. LCOE has a base-case value of $0.08/kWh and optimizes to $0.06/kWh. For our study, the base-case LCOE_el is $0.34/kWh, which is reduced by 35% to $0.22/kWh on optimization.

The Monte Carlo uncertainty assessment (results shown in Figure 3) on an optimized system also produces the same mean LCOE of $0.06/kWh (general extreme value distribution with a standard deviation of $0.015/kWh) and LCOE_el (general extreme value distribution with a standard deviation of $0.048/kWh) of $0.23/kWh, respectively. These values are still higher than the average price of electricity = $0.124/kWh. However, the box plot (Figure 4 shown for LCOE; LCOE_el has a similar trend) shows that it is possible to reduce the cost further by improving these parameters- lifetime, PV CAPEX, inflation rate, natural gas price, and electricity price. These parameters have dominant effects on the cost, and thus, future research should be more focused on improving them to make the system economically feasible.

We thus observe that the LCOE and LCOE_el reduce on optimization, and performing Monte Carlo uncertainty assessment reveals an enormous potential for cost reduction. Not many large-scale TPV prototypes exist, and our study provides a future research direction toward economic feasibility.

Our future work would involve performing a preliminary life cycle assessment (LCA) as the first step in quantifying potential decarbonization enabled by TPV.

Acknowledgements

We acknowledge the funding for the work by the Decarb 2040 Community Feasibility Grant Program.

References

(1) Datas, A.; Ramos, A.; del Cañizo, C. Techno-Economic Analysis of Solar PV Power-to-Heat-to-Power Storage and Trigeneration in the Residential Sector. Applied Energy 2019, 256 (October), 113935. https://doi.org/10.1016/j.apenergy.2019.113935.