price scaling

HSEM Price Interval Semantics and EDS Scaling

This document explains how HSEM handles the interaction between Energi Data Service (EDS) price update intervals and planning slot widths.

---

The problem

HSEM supports two independent interval settings:

Setting	Values	What it controls
energi_data_service_update_interval	15 or 60 minutes	How often EDS publishes price records
recommendation_interval_minutes	15 or 60 minutes	The width of each planning slot

When these differ (most commonly: EDS 60 min, slots 15 min), the price rate must be correctly scaled so the planner always sees the full currency/kWh rate.

---

The eds_share conversion factor

$$ eds\_share = \frac{\mathrm{EDS interval}}{\mathrm{Slot width}} $$

EDS interval	Slot width	eds_share	Effect
60 min	15 min	4.0	Price ÷ 4 stored; planner gets price × 4 back
15 min	15 min	1.0	No scaling
60 min	60 min	1.0	No scaling

---

Scaling pipeline

flowchart TD
    A[EDS raw price P\nfull hourly currency per kWh]
    B[HourlyDataPopulator._async_update_hourly_field]
    C[Recommendation slot storage\nHourlyRecommendation objects]
    D[coordinator._build_planner_input]
    E[Planner engine PricePoint\nimport_price = P]

    A --> B
    B -->|Per-slot stored value = P / eds_share| C
    C --> D
    D -->|Planner input value = stored value × eds_share = P| E

Loading

What this is NOT

• eds_share is not a VAT multiplier
• eds_share is not a currency conversion
• eds_share is not an energy-splitting factor (prices are rates, not energy)

---

Invariants

For any configuration:

1. A 60-min EDS price of P must reach the planner as P (not P/4 or P*4)
2. A 15-min EDS price of P must reach the planner as P
3. Intermediate per-slot stored values must equal P / eds_share
4. Changing energi_data_service_update_interval from 60 to 15 with the same price input must not change the price seen by the planner engine
5. Negative prices must survive the full pipeline unchanged (no absolute-value clipping, no zero-flooring)

---

Multi-day price data

For horizons beyond 24 hours, prices and PV data are projected onto the shared time-series index per calendar day:

Field	Source	Day offset
Today's prices	Live EDS sensor	day_offset = 0
Tomorrow's prices	EDS tomorrow sensor	day_offset = 1
Day+2 prices	EDS day+2 sensor	day_offset = 2

Missing future-day data is surfaced in DataQuality as:

• tomorrow_price_missing_hours
• day2_price_missing_hours
• tomorrow_pv_missing_hours
• day2_pv_missing_hours

Non-critical missing data triggers Degraded mode (writes allowed).