AI Battery Pack Controls for Superior Performance
EVE-Ai™ creates Range Extension, Fast Charging Strategy, and Lifetime Reassurance strategies for every battery pack. AI Algorithms decide the best method to get there for each unique pack.
Dynamic State of Charge
+12% range extension
+11% lifetime extension
Adjust the min/max SOC based on CAN & BMS data on temperature, humidity, and cell SOH.
EVE-Ai™ uses battery pack-specific AI to increase usable energy and decrease energy consumption, extending the vehicle's range capabilities.
Velocity Recommendations provided throughout drive cycle operate in conjunction with existing route planning software
Dynamic State of Charge (SOC) Limits expand the minimum and maximum SOC during safe conditions to extend range without compromising BMS safety functions
Eco routing is unlocked with ADAS and enhanced in Autonomous Vehicle installations
EVE-Ai™ reduces the strain of charging with Dynamic Thermal Charging Strategy, a Predictive Charging Model, and Overnight Charging Optimization.
Dynamic Thermal Charging Strategy adapts charging strategy to diverse temperatures across the battery pack for fast charging with limited damage to cell lifetime
Predictive Charging Model created for each individual based on historical and fleet data for unique user habits
Overnight Charging Optimization protects battery life by timing overnight charging with predicted morning drive time, allowing for fast charging the next day
EVE-Ai™ ensures battery packs meet service life and warranties by continuously updating cell modeling and usable lifetime predictions based on battery cell wear.
Accurate State of Health (SOH) and State of Charge (SOC) modeling is made possible with our Adaptive Cell Model trained on laboratory cell data and improved with battery pack utilization
Battery Fault Predictability uses AI to classify battery cell usable lifetime and alert before cell conditions become unsafe
Cloud Connectivity allows the AI model to continuously learn from worldwide fleet data for faster and more precise adaptation to specific driver profiles
EVE-Ai™ assesses fleet-wide battery pack SOH and SOC data to minimize environmental impact, maximize fiscal savings, and enhance driver safety.
AI Learning Using Fleet Data
Remaining Useful Life (RUL) Estimation
Embedded software installation in ECU, BMS, or Hybrid
Learns from historical and live data
Cloud connectivity for Over-the-Air (OTA) updates and V2X data sharing
Integration with BMS strategies
Integration with Autonomous and ADAS Systems
Battery Cell Chemistry Agnostic
CAN Bus communication
Neural Network Algorithms
Proprietary strategy based on 360 degrees of data coming from the battery pack, the vehicle, the driver, and the environment, adapts to each model to create an custom controls approach. These models include:
Energy Storage Models
Vehicle Characteristic Models
Individual Driver Profile Models
EVE-Ai™ Onboard AI Algorithm is Hardware Agnostic
Install in eMobility vehicles, no matter the application, installation location, or cell chemistry.
EVE-Ai™ is hardware agnostic and functions within standard BMS safety parameters
Deployed in current or future ECU, BMS, or Hybrid installation
Electra's proprietary neural network algorithms work on the back-end to inform AI learning
Data collection, cleaning, and labeling occurs in the Cloud, saving memory storage on each hardware installation
Have software integration questions?
V2X Communication for Energy Storage Precision
EVE-Ai unlocks V2V, V2I, and V2G capabilities to adapt energy storage management according to surrounding conditions and activity for a detailed control approach
Vehicle to Vehicle
For data sharing across fleets and models to improve AI responses in situations not yet encountered by each individual vehicle
Vehicle to Infrastructure
Traffic, weather, and condition predictions implemented into AI decisions in tandem with available IoT infrastructure
Vehicle to Grid
Grid communication improves charging decisions to appropriately time charging with grid activity and save energy cost
Adaptive Cell Model
Use programmed and current data to control module activity and predict battery failure.
Our cutting edge electrochemical model uses lab data, historical data, and programmed scenarios of simulation data to train recurring neural networks to make predictions and control module activity.
This cell model is updated to match the cell type, energy output rate (kWh), and specific vehicle and is continuously updated throughout the vehicle's lifetime.
The Adaptive Cell Model learns from all relevant connected systems. For example, combining data from other individual vehicles in a fleet of the same cell, pack, and vehicle type.