Day-Ahead Bidding: From Schedule to Bids

So far, your optimizer takes prices as given and produces a dispatch schedule. But in reality, you don't know tomorrow's prices when you bid. The day-ahead auction clears all bids simultaneously — your bid influences the price. This lesson teaches you how the day-ahead market actually works and how to translate your optimizer output into valid exchange bids.

How the day-ahead auction works

The EPEX SPOT day-ahead auction (used across Central/Western Europe) operates like this:

1. Gate opens: Market participants submit bids for each hour of tomorrow (products: hourly, block).
2. Gate closes: Typically 12:00 CET on day D-1.
3. Clearing: The exchange aggregates all buy and sell bids into supply/demand curves per hour. The intersection determines the Market Clearing Price (MCP).
4. Results: Published ~12:42 CET. If your bid price ≤ MCP (for sells) or ≥ MCP (for buys), you're accepted.

Bid types for batteries

Hourly bids (stepwise or linear)

You submit a price-quantity pair for each hour. Example: "I will buy 10 MW in hour 3 if price ≤ €25/MWh" or "I will sell 10 MW in hour 17 if price ≥ €60/MWh."

For a battery, you submit two bids per hour:

• Buy bid (charge): "Buy up to P_max MW if price ≤ my willingness-to-pay"
• Sell bid (discharge): "Sell up to P_max MW if price ≥ my minimum acceptable price"

Block bids (linked hours)

A block bid is an all-or-nothing bid spanning multiple consecutive hours. Example: "Buy 10 MW for hours 1–4 at average price ≤ €22/MWh — accept all hours or none." Useful for batteries because charging needs a sequence of hours, not just one.

Linked block bids

Parent-child links: "Accept my discharge block (hours 17–19) only if my charge block (hours 2–5) was also accepted." This ensures the battery has energy before it tries to sell.

From optimizer to bids: the strategy

The simplest approach — and the one used by most BESS operators in production:

1. Forecast prices — use a price forecast for tomorrow's 24 hours
2. Run your optimizer from Lesson 3 using the forecasted prices
3. Determine bid prices — for each hour, calculate your breakeven price:
   • Charge hours: bid at a price slightly above your forecast (ensures acceptance if price is low)
   • Discharge hours: bid at a price slightly below your forecast (ensures acceptance if price is high)
4. Submit bids — quantities from optimizer, prices from step 3

"""
day_ahead_bidding.py — Generate exchange bids from optimizer output
"""
import pyomo.environ as pyo
import numpy as np
import pandas as pd

# ... (model from Lesson 3, solved) ...
# Assume: model is solved, schedule DataFrame exists

# --- Bid generation ---
price_forecast = prices  # In production: from your forecasting model
margin = 2.0  # €/MWh — bid margin (tolerance for forecast error)

bids = []
for t in range(T):
    charge_mw = round(pyo.value(model.p_charge[t]), 2)
    discharge_mw = round(pyo.value(model.p_discharge[t]), 2)

    if charge_mw > 0.1:  # Charging hour
        # Buy bid: willing to pay up to forecast + margin
        bids.append({
            'hour': t,
            'direction': 'BUY',
            'quantity_mw': charge_mw,
            'price_eur_mwh': round(price_forecast[t] + margin, 2),
            'type': 'hourly'
        })
    elif discharge_mw > 0.1:  # Discharging hour
        # Sell bid: willing to sell at forecast - margin (minimum)
        bids.append({
            'hour': t,
            'direction': 'SELL',
            'quantity_mw': discharge_mw,
            'price_eur_mwh': round(price_forecast[t] - margin, 2),
            'type': 'hourly'
        })

bid_df = pd.DataFrame(bids)
print(bid_df.to_string(index=False))

What if bids are partially accepted?

If the clearing price falls between your charge and discharge bid prices, some hours may not clear. Your schedule is then partially filled. This is a problem: maybe only the charge hours cleared but not the discharge hours. You're stuck with a full battery and no sell position.

Solutions:

• Linked block bids — ensure charge and discharge are accepted together
• Intraday correction — trade out of unwanted positions in the intraday market (Lesson 5)
• Conservative bidding — bid at extreme prices (€-500 for buys, €3000 for sells) to guarantee acceptance, accepting whatever the clearing price turns out to be. This is called a "price-taker bid."

Price-taker bidding (the simple approach)

Many battery operators skip the price-guessing entirely and submit price-taker bids:

# Price-taker: guarantee acceptance, take whatever price clears
price_taker_bids = []
for t in range(T):
    charge_mw = round(pyo.value(model.p_charge[t]), 2)
    discharge_mw = round(pyo.value(model.p_discharge[t]), 2)

    if charge_mw > 0.1:
        price_taker_bids.append({
            'hour': t,
            'direction': 'BUY',
            'quantity_mw': charge_mw,
            'price_eur_mwh': 3000.00,  # Max allowed — always accepted
            'type': 'hourly'
        })
    elif discharge_mw > 0.1:
        price_taker_bids.append({
            'hour': t,
            'direction': 'SELL',
            'quantity_mw': discharge_mw,
            'price_eur_mwh': -500.00,  # Min allowed — always accepted
            'type': 'hourly'
        })

The full bidding workflow

┌─────────────────────────────────────────────┐
│  D-1, morning: receive price forecast       │
│  ↓                                          │
│  Run optimizer → get schedule               │
│  ↓                                          │
│  Generate bids (hourly or block)            │
│  ↓                                          │
│  Submit to exchange before 12:00 CET        │
│  ↓                                          │
│  ~12:42: receive clearing results           │
│  ↓                                          │
│  Check accepted vs. rejected bids           │
│  ↓                                          │
│  If partial fill → correct in intraday      │
└─────────────────────────────────────────────┘

Check your understanding

Rechtliches: Impressum