mrvictoru/AEMO_simulated_trade_sdp

AEMO SDP-Teacher Trajectories

Offline trajectories for Decision Transformer training, generated by replaying the honest SDP/MPC executor on historical Australian NEM (AEMO) market data. These are the teacher trajectories from the energydecision research codebase.

Each row is a single 5-minute market interval with a self-consistent (normalized observation, 9-dim action, reward) triple: the action is what the honest optimal planner dispatched, the reward is what it earned, and the observation is the environment state under that action.

2026-09 update (physics-v2): the shipped corpus was regenerated with the corrected degradation/efficiency physics and a wear-calibrated teacher (planner per-step wear matched to the environment's realized per-cycle wear; --deg-calibration 0.12). It supersedes dt_trajectories_jtsoc_combined.parquet.

Files

(Removed 2026-09: the Stage B discounted-return corpora dt_trajectories_full.parquet and dt_trajectories_aggressive.parquet, which paired with the deleted Stage B model weights.)

What is different about the physics-v2 corpus?

1. 1.Wear-calibrated teacher. The planner's per-step wear estimate previously over-counted the environment's realized (per-closed-cycle) wear by ~8.4× on identical actions, making the distilled policy over-conservative. A deg_calibration factor was added and the corpus regenerated with --deg-calibration 0.12, so the teacher's planned wear matches the environment's accounting. Net effect vs the pre-fix corpus: energy dispatch ≈ −27%, FCAS bidding ≈ +14%, and the retrained standalone DT beats the previous shipped checkpoint on 3/4 identity surfaces.
2. 2.Corrected shared physics. Round-trip efficiency (0.80), calendar aging, and corrected rainflow C-rate units are active in both the teacher and the environment.

The shipped model trained on this corpus is `energydecision-dt-v2-sdp` (aemo_dt_sdp_jtsoc_v2cal.pt). Train on it with --rtg-source j_t_soc (consumed via the rtg_value column).

Schema

Dataset composition

• —Shipped physics-v2 corpus: 320 episodes / 3.13M rows
• —5 regions × 2 horizons (short ≈ 14-day, medium ≈ 8-week) × 4 battery sizes × 8 episodes each — balanced
• —Time range: 2021–2023 (training-era, pre-eval)

How it was generated

Each episode is a random window sliced from the region's cached preprocessed 5-minute AEMO data (processed_{REGION}_2021-2023.parquet), replayed through the honest SDP/MPC executor:

1. 1.A seasonal time-of-day RRP forecast built from training-era data only (the predictable diurnal/monthly structure; spikes excluded).
2. 2.A stochastic dynamic program (AEMOSDPSolver) plans energy dispatch (backward induction over the forecast, degradation-aware via a corrected per-step wear estimate scaled by deg_calibration).
3. 3.A greedy current-price FCAS bidder allocates residual headroom to the highest-priced raise/lower service each step.
4. 4.The free-terminal recursion yields the Jt(soc) cost-to-go table, sampled per step into `rtgvalue`.

No future (realized) prices are used — the teacher is honest (non-clairvoyant). See src/aemo_sdp_executor.py and scripts/generate_sdp_dt_trajectories.py in the energydecision repo.

Usage

import polars as pl

df = pl.read_parquet("hf://datasets/mrvictoru/AEMO_simulated_trade_sdp/dt_trajectories_jtsoc_v2cal_conservative.parquet")
print(df.height)  # 3,133,440

Decision Transformer training

The schema matches the TrajectoryDataset used by scripts/pretrain_decision_transformer.py. Train a 9-dim full_fcas Decision Transformer (8×768, mixed Tanh/Sigmoid action head) on the physics-v2 corpus with --rtg-source j_t_soc → the shipped configuration.

Related

• —Shipped model: energydecision-dt-v2-sdp
• —Source repo: mrvictoru/energydecision
• —Older AEMO corpus (behaviour-cloning era): AEMO_simulated_trade