Supplement S1: Energy Storage Capability Classification with Response and Energy Memory
Supplement S1 to the conference paper "Energy Storage Capability Classification with Response and Energy Memory" (PE0053). Contents: (S1.1) scope of the supplement; (S1.2) data generator and class-dependent parameter ranges; (S1.3) resource split of 32 training, 8 validation and 10 test resources; (S1.4) the thirteen candidate configurations and their cumulative simplification scheme; (S1.5) acceptance criteria relative to the five-seed O64 validation mean; (S1.6) complete validation results for all candidates; (S1.7) parameter-counting definitions (stored, active, dormant, effective); (S1.8) training recipe including optimizer, schedule, batch size, clipping and seeds 11/22/33/44/55; (S1.9) timing procedure with the measurement environment and warm-up protocol; (S1.10) complete test results; (S1.11) parameter and latency savings. The supplement supports the reproducibility of the selection procedure reported in Tables II-V and Fig. 2 of the paper.
Authors
- Jianbo Nie
- Wenjun Wu
- Chao Zhu (ORCID: https://orcid.org/0000-0003-3537-6414)
- Xiaoze Lin
- Youbing Zhang
Institutions
- State Grid Zhejiang Electric Power Company (China) (CN)
- Zhejiang University of Technology (CN)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-08
- DOI
- https://doi.org/10.5281/zenodo.24047110
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00