Dynamic Load Balancing Explained

Dynamic load balancing optimizes the charging speed of EVs while preventing overloading of the building's electrical system.

Introduction

Electric vehicles (EVs) are becoming increasingly popular as a more sustainable and environmentally friendly alternative to gasoline-powered cars. The widespread adoption of EVs poses a challenge to the electrical infrastructure of the buildings and parking lots in which they are deployed. Most buildings do not have a dedicated electrical connection for EVs and must use the existing connection, which is shared with other loads such as lighting and ventilation. Dynamic load balancing is a technology that can help identify the available power and optimize its use by electric vehicles.

What is load balancing?

Load balancing, often known as load management, is a technology that monitors and controls the charging of EVs in real-time. It works by dynamically adjusting the amount of power that is delivered to each EV charger. This adjustment is based on several factors, such as the availability of grid power, the current demand for EV charging, and the charging needs of individual EVs.

Dynamic vs. static load balancing

Dynamic and static load balancing are two different approaches to EV charging management. Both methods aim to optimize EV charging to reduce strain on the electricity infrastructure and improve efficiency. However, they differ in their approach and level of complexity.

  • Static load balancing – assumes a fixed (static) amount of electricity is available to all EV chargers on the site. This amount is then distributed to all the chargers based on the load management algorithms. For example, a building with a 200A connection, where 50A is used for LED lighting, can allocate 150A to EV charging, given that the LED lighting load is quite low and predictable.
  • Dynamic load balancing – involves the constant measurement of non-EV loads on the panel to calculate the load available for EV usage. Let’s look, for example, at a building with a 200A connection shared between EVs and the building’s water pump, rated at 100A.  The water is pumped to the rooftop tank only a few times daily (typically when people use water for showers, etc.). Most of the day, however, the pump is off.
    If we had to use static load balancing, we would have to allocate only 50A for the EV chargers. By placing a meter on the main panel, we can detect when the pump isn’t active and allocate the full power for EV charging. When the pump activates, the system will reduce EV charging to accommodate the needed capacity.

The table below compares the two systems:

Static load balancing Dynamic load balancing
Main usage Simple sites with a dedicated EV feed or relatively flat non-ev consumption Energy-constrained sites or sites with fluctuating non-ev loads.  
Need hardware No hardware is needed other than OCPP smart chargers. A meter on the main panel.
Simplicity Very simple to implement Simple to implement

Wevo’s support for static load balancing

With Wevo, administrators can define the building’s full hierarchical electrical structure, including panels, sub-panels, and chargers. The system automatically calculates the load and capacity on each node and implements advanced AI algorithms to ensure each vehicle is charged when needed. Implementing static load balancing is simple. Just denote the rated capacity of the panel and the statically allocated capacity to EV loads:

Static load balancing settings
Static load balancing settings

With static load balancing, the panel will always allocate static capacity for other (non-EV loads), as can be seen in the image below. In the example below:

  • A three-phase panel of 40A is used
  • 20A are statically allocated to non-EV (denoted in light red in the image below)
  • The remaining 20A are allocated to four chargers on the panel.
  • In the image below, Charger 1 is active using 16A per phase.
  • When the other chargers connect, they will share the 20A between them, always leaving the statically allocated 20A to other loads.

 

 

Wevo’s support for dynamic load balancing

To support dynamic load balancing, simply add a meter to the panel and enable the meter in the user interface:

Dynamic load balancing settings
Dynamic load balancing settings

When the meter is active and online, its actual readings will be dynamically used to calculate the available power for EV charging. In the example below:

  • A three-phase panel of 40A is used
  • The connected meter constantly measures the non-EV loads (denoted in dark red in the image below).
  • The remaining panel capacity is allocated to the four chargers on the panel.

 

Dynamic load management

 

What happens when the meter is offline or faulty?

When the meter is offline or faulty, the system automatically returns to static load balancing using the static EV limit set at the panel. This may be less optimal, but it ensures the system maintains its integrity and never exceeds the available capacity.

Which hardware does Wevo support for load balancing?  

The Wevo system is hardware agnostic.  The algorithms are implemented in the cloud using OCPP-compliant chargers.  Static load balancing requires no additional hardware and is implemented out of the box.

Dynamic load balancing uses connected meters. Wevo supports many leading meter providers, including:

All these meters are connected to the Internet. Some require an extra LTE modem or a stable Wifi connection to operate. Most are quite inexpensive and cost less than $150.

Shelly meter
Shelly meter

Summary

The Wevo platform has a powerful built-in energy management system. By default, and without any hardware added, it can perform AI-based static load balancing using a fixed panel limit. For more congested sites or sites with variable loads, adding a hardware meter can enable dynamic load balancing to maximize the capacity of the building even further.

 

 

 

SHARE

We thought that you would be interested in reading these articles:

Talk to our Specialist

Fill in the form below to book a 30 minute no-obligation consulting session.

Your information will be processed as detailed in our privacy policy.

Skip to content