Fleet load balancing is the real-time distribution of available electrical power across multiple EV chargers (and sometimes across multiple depots) so a fleet can charge many vehicles without exceeding site limits, triggering demand peaks, or causing outages—while still meeting vehicle readiness targets.
What is fleet load balancing?
In a fleet depot, load balancing typically means:
– Setting a site power cap (e.g., 120 kW total for all chargers)
– Automatically sharing that power across active charge sessions
– Adjusting charger setpoints continuously as vehicles plug in/out
– Applying priority rules so critical vehicles get power first
Load balancing can be:
– Static (fixed split, simple equal sharing)
– Dynamic (real-time allocation based on constraints, priorities, and deadlines)
Why fleet load balancing matters
– Prevents overloading the transformer/main breaker and nuisance trips
– Reduces costly peak demand and demand charges (where applicable)
– Lets you support more vehicles per depot with limited grid capacity
– Improves operational reliability and readiness by avoiding congestion
– Enables phased expansion: add bays now, add power later
Common load balancing types in fleet depots
1) Charger-to-charger balancing (local)
– A group of chargers shares a fixed maximum power
– Example: 10 chargers share 70 kW total, dynamically allocated
2) Site-level balancing (main feeder)
– All chargers stay within a site cap that accounts for other building loads
– Requires a meter or signal from the main incomer / energy meter
3) Phase balancing for AC depots
– Ensures L1/L2/L3 phases are used evenly to avoid phase overload
– Important when many single-phase vehicles or uneven phase assignments exist
4) Priority-based balancing (fleet-aware)
– Allocates power based on business rules:
– earliest departure first
– lowest SoC first
– critical route vehicles first
– “must-charge” exceptions
5) Multi-site balancing (portfolio orchestration)
– Shifts charging load across depots based on capacity, price, or DR events
– Usually requires a centralized EMS and consistent data/control rights
How load balancing is implemented
Control layer (CPMS/EMS)
– Uses OCPP or vendor APIs to set charger power limits (kW)
– May control per connector, per charger, or per charger group
Measurement layer (the key to accuracy)
– Main meter / CTs measure total site load in real time
– The system subtracts non-EV building load to compute “available for EVs”
– Without measurement, balancing becomes conservative or inaccurate
Rule engine
– Defines minimum power per active session (avoid too-low trickle charging)
– Defines max power per charger (hardware limit)
– Defines how to allocate available power (equal share vs priority score)
Inputs you should define (minimum)
– Site limit (kW) and whether it is hard or soft
– Charger groupings (which chargers share a cap)
– Minimum and maximum per-charger power
– Priority logic (deadline-based readiness vs equal sharing)
– How other building loads are measured and included
– Fallback behavior if metering or connectivity fails
KPIs that show load balancing is working
– Peak kW reduction (before vs after)
– Readiness rate (vehicles charged by departure)
– Charger utilization and queue time
– Number of breaker trips / overload events (should approach zero)
– Public charging fallback rate (should reduce)
– Cost per kWh and demand charge impact (if applicable)
Common mistakes
– Setting a site cap without measuring other site loads (cap becomes wrong)
– Balancing only by equal sharing and ignoring departure deadlines
– No phase planning for AC depots → one phase overloads early
– Too-low minimum power → vehicles never reach targets
– No exception/override workflow for urgent vehicles
– Unclear ownership of connectivity/firewall issues → balancing fails at the worst time
Related glossary terms
Dynamic load management
Active power throttling
Fleet charging scheduling
Depot power management
Demand charges
Charger utilization rate