Note
Go to the end to download the full example code.
Kazakhstan wind-solar-battery hybrid: reproducible dispatch sensitivity#
News trigger: Ember, 3 November 2026, reports that a Kazakhstan wind-solar-battery hybrid could cover 70-90% of a typical demand profile at a delivered cost below $60/MWh.
This gallery example does not reproduce that cost or coverage claim. It uses explicitly SYNTHETIC 14-day hourly profiles to illustrate how battery energy capacity changes renewable demand coverage. There is no assumption that these profiles represent Kazakhstan measurements.
Original source: https://ember-energy.org/latest-insights/kazakhstan-has-the-wind-to-power-its-demand-growth
Business question: how much does storage increase the share of a demand profile served by co-located renewables? The method is a deterministic hourly energy balance with charge/discharge power limits, storage energy capacity and round-trip efficiency. It excludes network congestion, electricity prices, battery aging, reserve constraints and financing.
import math
import matplotlib.pyplot as plt
from energy_methods.hybrid_dispatch import simulate_hybrid
hours = range(14 * 24)
# SYNTHETIC, deterministic shapes. Not calibrated to Kazakhstan.
wind = [48 + 25 * math.sin(2 * math.pi * h / 79) for h in hours]
solar = [75 * max(0, math.sin(math.pi * ((h % 24) - 6) / 12)) for h in hours]
demand = [90 + 12 * math.sin(2 * math.pi * ((h % 24) - 17) / 24) for h in hours]
capacities = [0, 50, 100, 200, 400, 800]
results = [simulate_hybrid(wind, solar, demand, battery_mwh=e, battery_mw=50) for e in capacities]
baseline = results[0]
for capacity, result in zip(capacities, results, strict=True):
print(
f"Synthetic battery {capacity:3d} MWh: "
f"coverage {result.coverage:.1%}; "
f"curtailment {result.curtailed_mwh:.0f} MWh; "
f"unserved {result.unserved_mwh:.0f} MWh"
)
Synthetic battery 0 MWh: coverage 68.6%; curtailment 3666 MWh; unserved 9495 MWh
Synthetic battery 50 MWh: coverage 70.8%; curtailment 2928 MWh; unserved 8831 MWh
Synthetic battery 100 MWh: coverage 72.7%; curtailment 2274 MWh; unserved 8242 MWh
Synthetic battery 200 MWh: coverage 75.6%; curtailment 1274 MWh; unserved 7370 MWh
Synthetic battery 400 MWh: coverage 78.3%; curtailment 241 MWh; unserved 6560 MWh
Synthetic battery 800 MWh: coverage 78.3%; curtailment 241 MWh; unserved 6560 MWh
fig, ax = plt.subplots()
ax.plot(capacities, [100 * r.coverage for r in results], marker="o")
ax.axhline(100 * baseline.coverage, linestyle="--", label="No storage baseline")
ax.set(
xlabel="Battery energy capacity (MWh)",
ylabel="Demand served (%)",
title="Illustrative hybrid dispatch — SYNTHETIC inputs",
)
ax.set_ylim(0, 100)
ax.legend()
fig.tight_layout()
plt.show()

Interpretation: differences here are scenario sensitivities, not observed Kazakhstan performance. The dispatch is greedy, not globally optimal. Reproduction: uv run python examples/analysis/plot_kazakhstan_hybrid_dispatch.py
Total running time of the script: (0 minutes 0.066 seconds)