Energy Dispatch Optimizer
Multi-Period MPC Dispatch with Lunar Diurnal Forecast
Model-predictive dispatch optimizer that schedules charge/discharge across all storage technologies over multi-day horizons, weighing technology round-trip efficiencies, SOH-derived life-cycle costs, lunar day/night forecasts, and load priorities.
Purpose
Decide which storage discharges and when, given a portfolio of fast (supercap), short (Li-ion), and long-duration (RFC, thermal) options — minimizing waste, maximizing lifecycle of valuable assets, and ensuring critical loads are never starved.
Context
Top-level optimizer in L2-ESS-MGMT, drawing on L3-ESS-MGMT-SOC and -SOH; commanding L3-ESS-MGMT-CHG for charging actions and the individual L3-ESS-*-CTRL nodes for discharge. Constraints from L3-ESS-MGMT-SAFE override optimization for safety.
Principles
- ▸Model-Predictive Control (MPC) solves a constrained optimization over a rolling N-step horizon, applying only the first step before re-solving
- ▸Mixed-integer programming captures discrete decisions (which technology to dispatch) alongside continuous power flows
- ▸Cost function balances round-trip efficiency (energy wasted), wear (cycles consumed), and reliability (critical-load risk)
- ▸Lunar diurnal forecast (day/night, libration) provides predictable boundary conditions over multi-week horizons
- ▸Load classification (critical, essential, optional) drives priority during shortfalls
- ▸Reserve margin (e.g., 20% SOC always held back) protects against forecast errors
Typical implementations
- ▸Terrestrial microgrid energy management systems (Tesla Megapack, Generac PWRcell)
- ▸ESA mission planning autonomy frameworks
- ▸NASA cFS scheduler applications
- ▸Academic MPC publications for hybrid energy storage systems
- ▸ISS power management software (heritage centralized power planner)
Lunar considerations
- ▸Lunar night length is deterministic (~336 hr at equator, indefinite at PSR) → predictable discharge horizon
- ▸Eclipse of sun by Earth (lunar eclipse) every ~6 months adds short bonus discharge events
- ▸Long mission means dispatch optimizer accumulates statistics on load patterns and storage performance for improvement
- ▸Robotic maintenance windows must avoid power-critical periods → coordinated with L1-MNT scheduler
- ▸Solar particle events trigger derated power → dispatch must reserve battery for transient
- ▸Year-over-year drift in solar array output (dust accumulation) accounted for in long-horizon plans
Specifications
Functional
| primary function | Plan optimal multi-period charge/discharge schedule |
| inputs | SOC from L3-ESS-MGMT-SOC, SOH from L3-ESS-MGMT-SOH, Load forecast from L3-CDH-AUTO, Source power forecast from L1-PWR, Lunar diurnal/eclipse calendar from L3-COM-TIME-USOC, Safety constraints from L3-ESS-MGMT-SAFE |
| outputs | Charge plan to L3-ESS-MGMT-CHG, Discharge profiles to per-technology controllers, Reserve recommendations and risk telemetry, Long-horizon power forecast to L2-CDH-AUTO |
| planning horizon days | 14 |
| planning resolution minutes | 30 |
| optimization algorithm | Mixed-integer MPC |
| re plan interval min | 60 |
| supported load priorities | critical, essential, optional |
| reserve margin percent | 20 |
Physical
| mass kg | 0.0 |
| dimensions | Software on L1-CDH host |
| materials | Code on rad-hard NOR flash |
| operating temperature c | -40, 70 |
Operational
| power consumption w | 2 |
| thermal range c | -40, 70 |
| lifetime years | 100 |
| mtbf hours | 500000 |
Interfaces
Provides
- Charge schedule with targets per unit
- RFC dispatch commands
- Battery dispatch commands
- Power forecasts for autonomous goal planning
Requires
- Current and forecast SOC
- Health/RUL for cost weighting
- Source power forecast (reactor + solar)
- Lunar calendar for diurnal forecasting
Cite this entry
Lunar Ark Codex. "Energy Dispatch Optimizer" (L3-ESS-MGMT-DISP). Retrieved 10 September 2026, from https://lunarark.com/entry/L3-ESS-MGMT-DISP
Licensed CC-BY-SA 4.0. You may reuse and adapt this entry with attribution, under the same licence.