Balancing the Norwegian regulated power market anno 2016 to 2022

TL;DR

This study analyzes public Nordic balancing-market data for Norway's NO1–NO5 zones from 2016 to 2022 to understand how renewables and electrification affect balancing needs. It combines time-series analyses (ACF/DFT/ADF/Spearman) with an XGBoost-based predictor to examine volumes and prices, revealing that most hours have zero balancing, but non-zero balancing volumes rise over the period, while prices mainly track day-ahead levels with increasing non-linearities. Wind power growth raises balancing volumes but does not increase balancing-frequency, suggesting a shift in the distribution rather than in occurrence. The results show limited predictability of balancing volumes from public market data alone (per-zone $R^2$ circa $0.47$–$0.57$), with auto-correlation (lag-1) dominating temporal structure, underscoring the need for advanced models and continuous data updates for accurate forecasting.

Abstract

The balancing market for power is designed to account for the difference between predicted supply/demand of electricity and the realised supply/demand. However, increased electrification of society changes the consumption patterns, and increased production from renewable sources leads to larger un-predicted fluctuations in production, both effects potentially leading to increased balancing. We analyse public market data for the balancing market (manual Frequency Restoration Reserve) for Norway from 2016 to 2022 to investigate and document these effects. The data is newer than for similar analyses and the eight years of data is more than double the time span previously covered. The main findings are: a) The balancing volumes are dominated by hours of zero regulation but for non-zero hours, the balancing volumes are increasing during the eight-year period. b) The balancing prices are primarily correlated with day-ahead prices and secondary with balancing volumes. The latter correlation is found to be increasingly non-linear with time. c) The balancing volumes and the price difference between balancing price and day-ahead price are strongly correlated with the previous hour. d) The increasing share of wind power has not impacted the frequency of balancing, which has remained stable during the 8 years studied. However, the volumes and share of balancing power compared to overall production have increased, suggesting that the hours which are inherently difficult to predict remain the same. e) Market data alone cannot predict balancing volumes. If attempting, the auto-correlation becomes the main source of information.

Figures (16)

• Figure 1: The Norwegian power market and neighbouring zones throughout the period 2016-2022. The Norwegian zones NO1-5 are in part defined by bottlenecks in the transmission grid. The Swedish (SE1--4), Danish (DK1--2), Finnish (FI), and Baltic (EE, LV, LT) zones are regulated by the respective TSOs. Figure from FigElectricityPriceArea.
• Figure 2: The power production composition in the Norwegian market zones (columns) shown as monthly averages (red lines, left axis) and fractions of total for the same month and zone scaled between 0 and 1 (green lines, right axis). The bottom row shows the total production in each zone and the fraction of electricity production used for balancing within the same month (orange, right axis). Historically, hydropower has dominated, but for all zones there is a relative decrease in hydropower since 2016 (top row, green lines), and an increase in volatile sources e.g. onshore wind (third row, green lines). On a monthly basis, the balancing only makes up a few percent of the total production volume (bottom row, orange line). NO2 is the zone with larges absolute production and largest seasonal variations (bottom row, second column, red line). NO3 has the largest relative fraction of directly weather based production from from run-of-river and onshore wind.
• Figure 3: Hourly actual production (red, left axis), consumption (yellow, left axis) and balancing volumes (blue, right axis) for 2019 for each of the Norwegian zones from NO1 (top panel) to NO5 (bottom panel). In all zones, we observe that consumption is higher in summer than in winter (yellow). Some zones are roughly balanced between consumption and production (e.g. NO2) while others are dominated by consumption (e.g. NO1) or production (e.g. NO5). This will naturally affect the regulation volumes as well (blue). Since regulation is a correction to the day-ahead plan, it can take on both positive and negative values, and does not follow any obvious pattern.
• Figure 4: Upper two panels: Hourly averages and standard deviations (shaded) of 2019 data for production/consumption (red/yellow) volumes and corresponding balancing (blue) for NO1 and NO5. The characteristic daily pattern is visible in consumption and production, while there is no clear pattern in the balancing. Lower two panels: Monthly averages and standard deviations of 2016--2022 data for NO1 and NO5. The consumption and production show a characteristic seasonal pattern, while the balancing fluctuates over the entire year. In both cases, we show NO1 which is consumption-dominated and NO5 which is production-dominated, but similar patters occur for the other zones as well.
• Figure 5: Upper panel: Balancing volume histograms for the period 2016--2022: Actual balancing volumes for frequency of hours with zero regulation (orange, first bin), non-zero up-regulation (green) and absolute values of non-zero down-regulation (blue). The vertical dashed lines shows the standard deviation of the down- and up-regulations. Similar behaviours are present if the balancing volumes are normalised to actual production. The evolution of regulation volumes and potential asymmetry over time is shown in Fig. \ref{['fig:regulation_volume_moving_avg']}. Lower panel: Histograms of day-ahead and balancing prices for 2016--2022 for each of the Norwegian zones. We observe that the distributions seem to consist of two overlapping distributions (note the logarithmic scale): A narrower Gaussian centred around the median price (40 €/MWh) and a wider distribution of higher prices. A few counts with prices higher than 700 €/MWh are outside the plotting range.