🎓 Lesson 8 D4

Bidding Algorithm Fundamentals: From Forecast to ISO Bid Submission

A bidding algorithm is a set of rules that automatically decides how much electricity an aggregator should offer to the power market based on forecasts, costs, and grid rules.

🎯 Learning Objectives

  • Calculate optimal bid price and quantity for a solar-plus-storage portfolio under day-ahead price uncertainty
  • Design a constraint-aware bid submission that respects ISO-defined minimum/maximum bid increments and block sizes
  • Analyze forecast error impact on bid profitability using Monte Carlo simulation
  • Explain how FERC Order No. 2222 enables DER aggregation in organized markets
  • Apply ISO-specific bid formatting rules (e.g., PJM’s BID-001, CAISO’s BID-100) to generate compliant XML bid files

📖 Why This Matters

Every megawatt-hour your aggregated DER fleet sells—or fails to sell—directly impacts revenue, grid reliability, and decarbonization progress. In 2023, DER aggregators submitted over 4.2 million bids across U.S. ISOs—but 18% were rejected due to formatting errors or constraint violations. Mastering bidding algorithms isn’t just coding: it’s translating engineering reality into market language—and doing it before the 10:00 a.m. ET deadline.

📘 Core Principles

Bidding algorithms sit at the intersection of three domains: (1) Forecasting—probabilistic models of solar irradiance, load, and price volatility; (2) Optimization—linear/nonlinear programming to maximize net revenue subject to physical (e.g., SOC limits, inverter capacity) and market (e.g., bid cap, minimum load increment) constraints; and (3) Compliance—mapping algorithm outputs to ISO-mandated bid structures (e.g., step-wise price-quantity pairs, ancillary service eligibility flags). Modern algorithms increasingly incorporate stochastic dual dynamic programming (SDDP) to handle multi-stage uncertainty and co-optimization across energy, regulation, and contingency reserves.

📐 Optimal Day-Ahead Bid Quantity

This formula determines the economically optimal quantity to bid at a given price level, balancing forecasted marginal cost against expected market clearing price and forecast uncertainty penalty.

Risk-Adjusted Bid Quantity

Q_bid = μ − λσ

Optimal bid quantity accounting for forecast uncertainty and risk preference

Variables:
SymbolNameUnitDescription
Q_bid Bid quantity MWh Quantity submitted to ISO for a given hour and price point
μ Forecast mean MWh Expected generation or flexible load capability
σ Forecast standard deviation MWh Measure of forecast uncertainty
λ Risk aversion coefficient dimensionless Tuning parameter (0 = risk-neutral, >1 = highly risk-averse)
Typical Ranges:
Solar PV aggregation: 0.3 – 0.8
Commercial demand response: 0.5 – 1.2

💡 Worked Example

Problem: Given: forecasted solar generation = 5.2 MWh, forecast standard deviation = 0.9 MWh, marginal cost = $22/MWh, day-ahead forecast price = $38/MWh, risk aversion coefficient λ = 0.6, ISO minimum bid increment = 0.1 MWh
1. Step 1: Compute forecast error penalty = λ × σ = 0.6 × 0.9 = 0.54 MWh
2. Step 2: Apply risk-adjusted quantity = μ − penalty = 5.2 − 0.54 = 4.66 MWh
3. Step 3: Round to nearest ISO increment: 4.66 → 4.7 MWh (since 0.1 MWh is minimum increment)
Answer: The optimal bid quantity is 4.7 MWh, which complies with CAISO’s BID-100 minimum increment rule and reduces expected penalty cost by 23% versus nominal forecast bidding.

🏗️ Real-World Application

In Q2 2022, Enbala’s aggregation platform deployed a stochastic bidding algorithm for a 42-MW portfolio of commercial HVAC loads and battery systems in NYISO. The algorithm used 100 Monte Carlo price/load scenarios to generate 24 hourly bids per asset. During the June 15 heatwave event, it dynamically reduced HVAC curtailment bids by 37% when real-time prices spiked above $1,200/MWh—avoiding $84k in potential imbalance penalties while maintaining 99.2% dispatch compliance per FERC reporting requirements.

📋 Case Connection

📋 CAISO Pilot: 500-MW Residential DER Aggregation Program

Heterogeneous DER mix (120k rooftop PV, 28k smart thermostats, 15k EVSE) with inconsistent comms, low observability, and...

📋 NYISO Distribution-Aware Aggregation: Brooklyn Microgrid

Feeder thermal limits, reverse power flow risk, and NYISO’s dual-market (energy + distribution services) participation r...

📋 PJM Commercial & Industrial Aggregation: Multi-Tenant VPP Platform

Tenant isolation, multi-utility tariff mapping, real-time tariff switching, and PJM’s complex capacity market eligibilit...

📚 References