再エネ出力制御　GWを終えて

What Golden Week 2021 Told Us About Curtailment

For solar plant owners in Japan, the Golden Week period has become an important checkpoint for understanding how grid operators are handling renewable curtailment (再エネ出力制御). In 2021, there was widespread speculation that curtailment might extend to Hokkaido and Tohoku alongside Kyushu. In the end, only Kyushu Electric Power Transmission & Distribution actually implemented output controls during the holiday period.

That outcome is worth pausing on. It tells us that curtailment risk is still heavily concentrated in specific regions, but it also shows how quickly that can change as more solar capacity connects to weaker grid areas. For owners under the sotsu-FIT scheme, this is not an abstract policy question — it directly affects revenue assumptions and the timing of the 10-year inspection cycle.

The data released by Kyushu Electric after Golden Week gives us a rare chance to compare 2021 against 2020 on a like-for-like basis. What emerges is a picture of rising control frequency, shifting operational rules, and a clear signal that owners need to track their own exposure rather than rely on general market commentary.

Control Frequency Under the Old Rules

One of the most useful comparisons from the Golden Week data concerns the so-called "old rules" (旧ルール) for curtailment. Because the operational method changed in April 2021, the only fair comparison is to count old-rule events separately.

From April through May 6, the number of old-rule control events was roughly 8 to 9 in 2020 and 9 to 10 in 2021 — essentially the same. That stability is somewhat misleading, however. The reason the operational method was changed at all was that the previous approach was projected to push old-rule control events beyond 30 in a single year.

For context, the final old-rule control count for fiscal 2020 was 22. If the old method had continued, fiscal 2021 would almost certainly have exceeded that. This means owners who assumed last year's curtailment levels would repeat should revisit their assumptions for autumn and the following March, when surplus conditions historically peak.

Another surprising detail: more than 20 new projects were still being built and connected under the old rules between 2020 and 2021. That suggests some developers were either unaware of the changing curtailment environment or were willing to accept the risk. Either way, it adds to the total volume of capacity exposed to controls.

Comparing Control Volumes: 2020 vs 2021

Beyond the number of events, the volume of curtailed output tells a more nuanced story. When you line up control volumes by date and sort them from smallest to largest, you get a clear sense of how the distribution has shifted.

The designated-rule (指定ルール) category has seen a notable increase in control frequency following the April 2021 operational change. Owners under this category have reported being caught off guard by controls occurring on consecutive days — something that did not happen the previous year. There was also an expectation that moving from rotating (輪番) to uniform (一律) curtailment would mean more events but a lower curtailment rate per event. That expectation has not held up in practice.

This matters for financial modeling. If you are projecting revenue for a plant in Kyushu, assuming a fixed curtailment rate per event is no longer safe. The pattern now appears to be more frequent events with rates that are not necessarily lower. Owners should be stress-testing their assumptions against both higher frequency and sustained curtailment rates.

Why the Operational Change Happened

To understand why the rules changed, it helps to look at the underlying supply-demand structure in Kyushu. The concept of a duration curve — sorting demand data from smallest to largest regardless of time — is a useful tool here.

When you plot Kyushu's 2020 electricity demand as a duration curve, you can see how often the system operates near its minimum. During those low-demand hours, especially in spring and autumn when solar output is high, the grid has limited room to absorb additional renewable generation. The old rotating curtailment method was designed to spread the burden, but it became clear that this approach would not keep annual control events under the threshold.

The shift to designated rules was essentially a response to that math. It concentrates curtailment on a defined group of generators rather than rotating across all of them. For owners in that group, the practical effect is more frequent controls. For owners outside it, the effect may be less immediate — but as capacity grows, the boundary of who is included can shift.

The Role of Interconnection Lines and Nuclear Restarts

Curtailment in Kyushu is not solely a solar problem. When the first output control was implemented on October 13, 2018, it followed the restart of the Genkai nuclear units — Unit 3 in May and Unit 4 in July of that year. Comparing April and October supply-demand data at noon, demand rose about 102% while supply rose 108%. Nuclear output increased 23%, while solar and thermal supply actually decreased.

This is a critical point for solar owners: curtailment risk is driven by the total generation mix, not just by how much solar is installed. Nuclear restarts, thermal minimum generation, and interconnection line capacity all shape how much room is left for renewables.

Interconnection line reinforcement (連系線増強) is often proposed as a solution. It can help, but it is slow, expensive, and does not address local surplus conditions within a region. For plant owners, the more actionable takeaway is that curtailment exposure depends on a combination of factors — location, rule category, and the broader generation mix — that can change over time.

What This Means for Your Plant

The Golden Week 2021 data confirms that curtailment is no longer a hypothetical risk for solar owners in Japan. It is an operational reality in Kyushu, and the conditions that drive it are evolving. Old-rule control counts are stable for now, but the underlying projections suggest upward pressure. Designated-rule owners are already seeing more frequent events than expected.

For owners preparing for the 10-year inspection or reassessing revenue forecasts, the priority is not to rely on regional averages or last year's numbers. Curtailment exposure varies significantly by rule category, location, and timing. The most reliable starting point is to accurately understand your own plant's condition — its rule classification, its historical curtailment record, and its specific grid connection context. From there, you can build realistic assumptions and make informed decisions about maintenance, investment, and long-term planning.