Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Ica Modeling Telemetry topic
No spam. Unsubscribe anytime.
Utilities point to data and modeling limits — AMI telemetry, LTCs and a new LTPT tool — as causes of ICA inaccuracies
Summary
Utility staff at the CPUC ICA workshop said many zero‑capacity results stem from conservative study assumptions, limited telemetry (AMI/SCADA) and model limitations; they described long‑term tool builds, telemetry upgrades and interim automation to improve results.
Utility engineers at the CPUC’s quarterly ICA workshop described technical drivers behind a large share of zero hosting‑capacity results and said the next major improvements depend on richer telemetry and a multi‑year tool build.
SCE, PG&E and SDG&E told stakeholders that common technical contributors to 0‑capacity outputs include low source voltage episodes captured in time‑series profiles, conservative study thresholds, missing transformer or substation telemetry, and model backfilling where historical AMI or SCADA data are unavailable. SCE engineers explained that their steady‑state voltage (SSV) studies use a ±3% tolerance around a 120‑volt baseline (resulting in a 118–126 volt threshold) and that adjusting lower‑limit assumptions yielded only modest improvements in zero counts.
Key technical fixes discussed: - Telemetry and AMI: Utilities said Advanced Metering Infrastructure (AMI2) rollout and improved substation telemetry would materially improve input quality. PG&E and SCE said AMI2 is expected to arrive in stages over several years (SCE cited 2028 in testimony for improved AMI inputs) and that better voltage reads will reduce the need to backfill missing data. - Long‑term planning tool (LTPT / LPTTSMT): SCE and PG&E are building centralized long‑term planning and system modeling platforms intended to unify planning assumptions, allow unbalanced flow analysis, and incorporate forecasted loads and planned projects. SCE said an engineering dashboard and LTPT milestones are targeted for delivery through 2025–2026 with additional capabilities into 2028. - Modeling choices and thresholds: Utilities said hourly time‑series methods, selection of a ‘‘most‑limiting hour’’ vs. node‑hour averaging, and how volt‑VAR support is simulated materially change outcomes. SCE noted volt‑VAR support is simulated only when network voltage drops below approximately 97% of nominal, which can leave some SSV violations unmitigated in the model.
Interim operational steps: Utilities described automation to reduce manual model fixes and a new batch refresh trigger for monthly ICA publication that prioritizes circuits with critical system changes, Rule 21 fast‑track applications, or long refresh gaps. They said automation and additional compute capacity would enable more predictable refresh cadence.
Stakeholder reaction: Engineers and advocates pressed utilities to identify near‑term practical measures for improving the accuracy that do not require waiting for AMI2 or a full tool rewrite. PG&E and SCE said they are testing incremental fixes (capacitor setting changes, backfill improvements) but that many solutions depend on telemetry or cross‑team coordination.
Bottom line: Utilities argued most zeros reflect real limiting conditions under current inputs; stakeholders replied that better telemetry and clear remediation timelines are necessary for ICA to function as an actionable developer tool.
