Improvement
- Added
set_deferrable_max_startupsparameter to enforce a hard physical limit on the maximum number of times a deferrable load can be turned on during an optimization horizon, preventing unwanted chattering - Added new Websocket section to documentation
- Improving coverage: MLForecaster, web server and command line utils
- Improved automatic testing: Add tests for converting nested parameters to config, deferrable load padding, export time range parsing, NaN handling, and resampling/filtering of time series data. Add an MPC cache behavior test to verify cache hits for non-structural plant changes and cache misses for structural plant changes
- Deactivate binary variables for non-thermal loads with 0 operating timesteps to reduce MIP solve time
Fix
- Ensure cached MPC optimizers are invalidated when structural plant configuration changes while still updating runtime-dependent parameters on cache hits
- Prevent stale internal optimization configuration by refreshing cached optimizer plant and optimization configs with latest runtime values
- Fixed logging initialization issue
- Fixed an issue where the REST API sends 'Bearer empty' when the token is not configured in the add-on
- Gracefully skip days with no history data during REST API retrieval
- Fixed Solcast API Accept header and sub-30min resampling issues
- Fix to REST API authentication fails when token is set to "empty" in addon mode issue
What's Changed
- New emhass release v0.17.0 by @davidusb-geek in #732
- Multiple improvements and fixes after v0.17.0 by @davidusb-geek in #735
- Second round of improvements and fixes after v0.17.0 by @davidusb-geek in #737
- fix: Solcast Accept header and sub-30min resampling by @kallegrens in #739
- fix: gracefully skip days with no history data in REST API retrieval by @kallegrens in #740
- Fix: REST API sends 'Bearer empty' when token not configured in addon… by @who2us in #742
- Fix include all parameter into cache hit/miss by @sokorn in #743
- Deactivate binary variables for non-thermal loads with 0 operating timesteps to reduce MIP solve time by @sokorn in #744
- Fixing dynamic lenghts affecting thermal loads and preparing for v0.17.1 by @davidusb-geek in #748
New Contributors
- @kallegrens made their first contribution in #739
- @who2us made their first contribution in #742
Full Changelog: v0.17.0...v0.17.1