A low solar LCOE does not guarantee strong returns when grid constraints, midday prices and curtailment reduce project revenues
Storage can add energy-shifting, capacity and ancillary-service value, but the available revenue stack depends on the market
JA ESS’s 5.015 MWh JAGalaxy BESS uses 314 Ah LFP cells and is rated at 2.508 MW and 0.5P
The cheapest solar electricity is not always the most valuable. With increasing PV deployment, especially the utility, major chunk of the solar power reaches the grid at the same time. This has led to grid constraints, falling capture prices and curtailment, of of which can reduce the amount of electricity sold and the revenue earned by a plant, even when its levelized cost of electricity (LCOE) is low.
Gloria Gao, Marketing Director at JA ESS, described this as the “solar paradox” during the TaiyangNews Solar Power Plants & Storage Conference 2026. She said LCOE remains an important measure of generation cost, but it does not account for it’s value that in deed varies with when electricity is produced, whether the grid can accept it and finally how much revenue the project earns.
Gao identified 3 connected factors making low-cost solar harder to monetize. Solar capacity is expanding faster than grid infrastructure in some markets, creating interconnection delays and transmission constraints. At the same time, high solar generation during midday increases electricity supply and lowers market prices, reducing the capture price earned by PV plants. Grid limitations can also force projects to curtail generation, leaving some available electricity unsold. Together, these factors are shifting the main bottleneck from generation capacity to flexibility, she said. In her view, storage is therefore moving from an optional addition toward a core part of power plant design.
However, the cost gap between solar and storage remains significant. Citing BloombergNEF data, Gao listed a global benchmark LCOE of $39/MWh for fixed-tilt utility-scale solar. Her presentation also showed a cost of $78/MWh for 4-hour battery storage. Still, developers added 87 GW of combined solar and storage in 2025, with an average delivered electricity cost of $57/MWh.
Moving “beyond LCOE,” Gao said projects should also be assessed according to the value stack they can access. LCOE and the levelized cost of storage (LCOS) provide the cost base. Whereas storage can add value through energy shifting, capacity, ancillary services, and grid support.
Gao described the main value of storage as “optionality.” A battery gives operators more control over when electricity is sold, where flexibility is provided, which services the system performs and how quickly it responds. For example, it can charge when solar generation is high and prices are low, and discharge when demand and prices rise. Part of its capacity can also be reserved for frequency response, ramp control, voltage support and other grid services.
According to Gao, storage changes a conventional solar plant at physical, operational and commercial levels. Physically, a generation-only plant becomes a system that can generate, store and supply electricity. The PV and battery may use a DC-coupled or AC-coupled architecture and share a grid connection or substation. Operationally, the plant can shift electricity to higher-value periods, recover energy that might otherwise be curtailed and respond to grid requirements. Commercially, it can move beyond a single power purchase agreement (PPA). Potential revenue sources include energy sales, price arbitrage, capacity payments, ancillary services and other grid services. Gao described the resulting system as a “flexible energy asset.”
The value available from storage differs between markets. In the US, projects can combine energy trading, capacity payments and ancillary services through markets such as ERCOT, CAISO and PJM. European markets place greater emphasis on balancing and flexibility products. Increasing periods of negative electricity prices also create opportunities to store surplus renewable generation.
In China, energy, capacity and flexibility mechanisms vary between provinces, with spot-market participation and both independent and shared-storage models developing. In the Middle East and North Africa (MENA), storage is more closely linked to large-scale solar, government-backed procurement.
Gao said there is no universal storage business model. The same battery can produce different financial results depending on market rules, grid conditions and the operating strategy. Storage systems must therefore be designed around the sources of value available in each market.
She presented an illustrative MENA configuration combining 500 MW of PV with a 200 MW/800 MWh lithium iron phosphate (LFP) battery. The battery would provide 4 hours of storage. The proposed DC-coupled system would share a point of interconnection and include an energy management system (EMS), supervisory control and data acquisition (SCADA), and a grid-forming-capable inverter.
The battery could store solar electricity that might otherwise be curtailed and supply it during higher-price periods. It could also provide grid services where market rules permit and support firm or time-matched clean-power contracts. Gao clarified that the configuration was an illustrative business case rather than a disclosed project. The results would depend on the site, market design and operating strategy.
JA ESS, JA Solar’s energy storage business, offers the JAGalaxy utility-scale battery energy storage system (BESS). The liquid-cooled system has 5.015 MWh of rated energy capacity and 2.5 MW of rated power. It uses 314 Ah LFP cells and has a charge and discharge rate of 0.5P. The system has a rated DC voltage of 1,331.2 V and a voltage range of 1,164.8 V to 1,497.6 V. Its battery compartment has an IP55 protection rating, while the specified operating temperature ranges from –30°C to 55°C.
The JAGalaxy includes real-time monitoring, pack-level fire detection, fire suppression, thermal monitoring, pressure relief and explosion protection. Its liquid-cooling system maintains the battery temperature at no more than 35°C, with a stated temperature difference of no more than 3°C across the system.
JA also lists AI-based predictive maintenance, remote diagnostics and remote software updates. According to the company, predictive maintenance can reduce unplanned downtime by up to 90%, while more than 90% of system issues can be addressed remotely.
Looking ahead, Gao identified reliable clean power supply, grid-forming storage and software-supported operation as 3 trends shaping flexible solar. Grid-forming batteries can contribute to voltage and frequency stability, particularly in weak grids and power systems with high shares of inverter-based generation. Software can improve generation forecasting, battery dispatch and revenue management.
Gao also referred to a 60 MWh solar-plus-storage project serving a data center in northwestern China. She presented it as an example of an integrated solar and storage system.
She described the development of solar plants in 3 stages. Solar 1.0 focused on achieving the lowest LCOE and generating as much electricity as possible. Solar 2.0 added storage, allowing projects to generate, store and shift electricity. The next stage combines PV, storage and digital controls to manage when, where and how electricity is supplied.
Gao said grid constraints, curtailment and pressure on project returns are moving the industry from cost optimization toward value optimization. Flexibility is becoming the next competitive advantage for utility-scale solar.
