China Three Gorges Corporation confirmed July 1 that its 1 gigawatt hybrid solar complex in the Gobi Desert has entered commercial trial operation — the first time in history that a single solar facility at this scale can generate electricity for up to eight hours after the sun goes down without a single battery cell.
The milestone matters beyond its record-breaking size. The prevailing assumption in grid planning is that solar power is inherently intermittent — valuable by day, absent at night — and that the only pathway to solar power on demand requires pairing it with lithium-ion battery banks. The China Three Gorges Corporation (CTG) Hami project in Xinjiang’s Gobi Desert is a commercial-scale challenge to that assumption. At 1 gigawatt total capacity, it proves that thermal storage in molten salt can do what batteries cannot yet accomplish economically at grid scale: store daytime solar heat and discharge it steadily for eight hours into the night.
Every solar photovoltaic array in the world suffers the same problem. During the day, the array produces electricity proportional to sunlight intensity. At sunset, output falls to zero. The sharper the solar penetration in a grid, the more acute the problem becomes: grids in high-solar regions like California experience what engineers call the “duck curve” — a demand spike at dusk precisely when solar generation collapses. Grid operators must ramp gas turbines or import power within minutes to cover the gap. The duck curve is not an edge case. It is the defining operational challenge of any grid with large solar capacity.
Batteries can flatten the duck curve — but at a price. Ember’s energy research group calculates that shifting half of daytime solar generation to evening hours via battery storage adds roughly $33 per megawatt-hour to solar’s total cost, yielding dispatchable clean electricity at approximately $76 per MWh. In China, the coal grid tariff benchmark is around $50 per MWh. Battery-backed solar at $76 per MWh does not yet clear that bar.
Molten salt thermal storage, deployed in a hybrid configuration with photovoltaic arrays, offers an alternative path — and the Hami facility is where that path crossed from engineering theory into confirmed commercial practice.
The Hami facility’s key technical component is a 100 megawatt linear Fresnel concentrated solar power (CSP) unit — China’s largest of its kind, backed by eight hours of molten salt thermal storage. The other 900 megawatts comes from conventional solar PV panels covering the same desert site. The two systems feed a centralized control platform that integrates their output and dispatches power to the regional grid.
The linear Fresnel CSP architecture works through a four-stage conversion chain. First, 260,000 high-precision flat tracking mirrors — spanning 800,000 square meters — follow the sun and reflect its rays onto elevated absorber tubes. Second, those absorber tubes sit inside secondary compound parabolic concentrators (CPCs) that concentrate the incoming light further. Third, the concentrated energy heats molten salt to 550 degrees Celsius as it flows through the system. Fourth, when power is needed — day or night — the stored hot salt passes through a steam generation system that produces high-pressure steam, which drives conventional turbine generators to produce electricity.
This is the same thermodynamic cycle that drives coal and gas power plants, just with salt replacing the fossil fuel as the heat source. That similarity is not incidental: it means the CSP unit behaves operationally like a controllable, dispatchable generator rather than a weather-dependent one. Project director Niu Jianle put it directly: “Capable of delivering stable power for up to eight hours straight, it serves as a core technology to guarantee non-stop grid reliability.”
During daylight hours, the PV array generates electricity at full load while simultaneously charging the molten salt storage system. After sunset, or during periods of cloud cover, stored heat drives the steam turbines for stable output. A centralized control platform coordinates both generation streams, achieving frequency regulation accuracy of around 0.02 hertz and response times under one second — precision comparable to conventional power stations.
CSP technology comes in four variants: parabolic troughs (curved reflector channels), power towers (heliostats focusing onto a central receiver), dish-Stirling systems, and linear Fresnel reflectors. The first two dominate global CSP capacity. The Hami plant uses linear Fresnel — a technology regarded within the field as less optically efficient than the alternatives but significantly cheaper to build and maintain.
Linear Fresnel reflectors use flat (rather than curved) mirror strips. Flat mirrors are cheaper to manufacture, require less structural steel, and because the absorber tubes are fixed rather than moving, the system demands less maintenance than a parabolic trough. The tradeoff is lower concentration factors and operating temperatures that max out around 550°C compared to 565°C or more for power towers. Linear Fresnel’s thermal efficiency typically falls in the 8 to 12 percent range, compared to power tower systems that can achieve 23 to 35 percent thermal-to-electrical conversion.
The Hami plant is betting that the cost advantage outweighs the efficiency penalty at 100 megawatt commercial scale. Niu described it as a “landmark leap, bringing the technology out of laboratory research and into large-scale commercial rollout,” according to PV Tech. Whether that claim holds up over a 20-year operational life will determine whether linear Fresnel CSP becomes a template for China’s second-generation solar base expansion — or remains a one-off demonstration.
It is worth noting that independent benchmarking of the Hami plant’s actual levelized cost of electricity (LCOE) has not yet been published. The facility entered commercial trial operation on July 1, 2026, and has produced a confirmed 6.54 million kilowatt-hours of clean power from its CSP unit since first grid connection in September 2025 — a data set too limited for long-run economic validation. Developer projections estimate the plant at the Chinese coal tariff benchmark of around five US cents per kilowatt-hour, but that remains to be confirmed by independent auditors over multiple years of operation.
The Hami plant displaces the Noor Energy 1 facility in Dubai, United Arab Emirates — a 950 megawatt hybrid combining 700 megawatts of CSP with 250 megawatts of PV — as the world’s largest integrated CSP-PV installation. With 1,000 megawatts total (100 MW CSP + 900 MW PV), the CTG facility exceeds Noor Energy 1 by 50 megawatts.
The record will not stand for long. China Energy Engineering Corporation (CEEC) broke ground on a 1.5 gigawatt hybrid project — consisting of 1.3 gigawatts of PV and 150 megawatts of CSP — also in Hami, approximately 30 kilometers from the CTG site. CEEC’s Hami project, with a total investment of CNY 6.5 billion (about US$952 million), was scheduled to reach grid-connection conditions by June 2026, with full-capacity commissioning targeted for October. Whether it met its June target has not been confirmed at time of writing. Once operational, the CEEC project will become the world’s largest CSP-PV hybrid.
The two facilities together represent a deliberate Chinese policy cluster. Beijing explicitly designated hybrid CSP-PV systems with thermal storage as a strategic pathway for managing solar intermittency at grid scale in China’s 14th Five-Year Plan. A dedicated policy document — “Some Opinions on Promoting the Large-scale Development of CSP” — was jointly issued by the NDRC and National Energy Administration on December 23, 2025, establishing the legal and economic framework for the next wave of CSP deployment.
Hami sits in one of China’s highest-irradiance zones, with more than 3,000 annual sunshine hours — a solar resource comparable to the most productive sites in the American Southwest or the Atacama Desert. The facility occupies 1,817 hectares of desert at the southern foot of the Tianshan mountains, terrain that offers abundant solar irradiance and is otherwise unsuitable for agriculture.
Strategically, Hami is positioned on the Zhundong-Wannan ultra-high-voltage direct-current (UHV DC) transmission corridor, which connects remote renewable generation in Xinjiang to electricity demand centers in eastern China. The project’s integration into this corridor is expected to lift Xinjiang’s renewable utilization rate above 95 percent — addressing the chronic curtailment problem that plagued the region as recently as 2016–2018, when surplus generation exceeded transmission capacity and up to 30 percent of renewable output was simply discarded unused.
At full operation, the facility is projected to generate 2.07 terawatt-hours annually — enough electricity to supply roughly 830,000 households — while avoiding an estimated 1.63 million metric tons of carbon dioxide emissions per year.
The commercial validation of 8-hour solar thermal storage at 1 gigawatt scale arrives at a moment when the energy storage industry is debating whether CSP-with-thermal-storage or PV-with-lithium-batteries will be the dominant solution to solar intermittency at long durations.
NREL researchers have projected that by 2030, PV paired with 10-hour lithium battery storage will cost roughly what PV with 4-hour battery storage costs today — potentially eliminating CSP’s cost advantage at longer discharge durations. Other researchers project that CSP-plus-TES will be cheaper than PV-plus-lithium for storage durations above four hours per day.
The Hami plant provides the first commercial-scale operational data point for the CSP-plus-thermal-storage side of that argument. It does not resolve the debate — one plant in its first months of commercial trial is insufficient evidence for a global technology conclusion — but it does confirm that the architecture works at 1 gigawatt scale, and that China’s state-owned energy sector has committed CNY 3.53 billion (approximately US$480 million) to proving out the economics in real-world conditions, as reported by World-Energy.
It is also worth noting that CSP plants have historically underperformed their projected output: real-world experience across multiple global installations shows production shortfalls of 25 to 60 percent against developer projections. The Hami plant’s projected 2.07 TWh annual output carries that caveat.
Any account of this project must note what it does not include: an independent audit of the supply chain that produced its 900 megawatts of solar PV capacity. The Hami facility is located in Xinjiang, and the Zhundong Economic and Technological Development Zone — adjacent to Hami — hosts major polysilicon manufacturers whose labor practices have been the subject of extensive international scrutiny. The US Uyghur Forced Labor Prevention Act (UFLPA), enacted in 2021, presumes that goods originating from Xinjiang involve forced labor unless proved otherwise. As of 2025, Adrian Zenz documented more than 3 million labor transfers of Uyghurs and other ethnic minorities in the region.
No evidence links the CTG Hami power plant itself to forced labor practices; the concern applies to the upstream polysilicon and panel supply chain, not to the facility’s operation. CTG declined to address supply chain provenance. The absence of an independent supply chain audit means this question remains open, and readers evaluating the project’s replicability outside China should factor in the labor and trade compliance implications of sourcing from Xinjiang.
Three Gorges Group has indicated plans to expand the Hami renewable energy base to 3 gigawatts in a second phase and to apply the integrated CSP-PV model in other resource-rich regions including Inner Mongolia and Gansu.
Construction began in 2023. The PV section was completed and grid-connected by December 2024. The thermal plant entered full operation in September 2025 — 42 days ahead of schedule. Total investment reached CNY 3.53 billion (approximately US$480 million). Core engineering, procurement, and construction services were provided by China Energy Engineering’s Northwest Institute and other domestic suppliers, along with project developer Dunhuang Dacheng Shengneng New Energy Technology. Ecological mitigation measures are underway, including drought-resistant vegetation across the site to counter desertification.
The facility is listed on China’s national registry of advanced low-carbon demonstration projects. Its commercial trial period — the phase that began July 1 — will determine whether the financial and performance claims are sustained by actual operating data. If they are, the Hami plant will represent more than a size record. It will represent the first commercially verified proof that solar power can be stored in salt and dispatched at grid scale through the night — without a battery in sight.
Yes — but only with thermal storage, not from standard solar PV panels. At Hami, a 100 megawatt linear Fresnel CSP unit heats molten salt to 550°C during daylight hours. After sunset, that stored heat drives conventional steam turbines for up to eight hours of continuous output. The 900 megawatt PV component produces electricity only during daylight. The CSP-plus-storage unit is what makes the full facility dispatchable around the clock.
The Hami plant’s developer, China Three Gorges Corporation, claims the facility is economically viable at China’s coal grid tariff benchmark of approximately five US cents per kilowatt-hour — but that claim has not yet been independently verified. Battery-backed solar currently yields dispatchable electricity at approximately $76 per MWh when the storage cost is included, according to Ember’s December 2025 analysis. Whether CSP-plus-thermal-storage can undercut that price at commercial scale is precisely what the Hami facility’s operational record will demonstrate over the coming years.
At the Hami plant, the molten salt system provides eight hours of continuous discharge at full output — enough to serve the grid from sunset through roughly two in the morning. More advanced CSP projects in China use tower designs with 12 or even 16 hours of thermal storage. Molten salt thermal energy storage is not subject to the cycle-life degradation that limits lithium-ion batteries; the salt does not chemically change during charge-discharge cycles, and a well-maintained system can sustain its rated storage capacity for the full 20-year design life of the plant.
Linear Fresnel reflectors are cheaper to build than parabolic trough or power tower CSP systems, but they are also less efficient. Their thermal-to-electrical conversion efficiency typically falls in the 8 to 12 percent range, compared to 23 to 35 percent for power tower designs. The technology is sensitive to dust accumulation on mirror surfaces — a significant operational challenge in the Gobi Desert. And as with all CSP plants, real-world output has historically fallen 25 to 60 percent short of developer projections. The Hami plant is the first 100 megawatt linear Fresnel facility to reach commercial scale globally, which means its multi-year performance data will be the primary evidence base for whether the technology can sustain its projected economics.
ⓒ 2026 TECHTIMES.com All rights reserved. Do not reproduce without permission.

source