_edited.png){kind=small}
TECHNOLOGY EXPERTISE
The Electra Team is made up of experts in energy storage, AI, and electric vehicle sectors
TECHNOLOGICAL CONCEPTS BELOW:
APPLIED AI - ADAPTIVE BATTERY DIGITAL TWIN​ - BATTERY PACK SAFETY -
360 ADAPTIVE CONTROLS - CLOUD CONNECTIVITY - LEVERAGING AUTOMOTIVE BIG DATA - V2X COMMUNICATION (V2G, V2I, V2V) - STATE OF HEALTH PREDICTION -
APPLIED AI
​
Artificial intelligence is the foundation of Electra’s technology and is used in two core areas of Electra’s battery pack controls: learning from 360 degrees of external condition factors and modeling cells according to historical battery data to inform the Adaptive Battery Digital Twin. Traditional BMS strategies lack the adaptability and detection capabilities required to respond quickly to multiple outside stimuli, as well as to accurately predict battery faults in time to take corrective action. They are also highly specialized to a specific battery chemistry or application, making it necessary to create new management strategies for each new system application. Applied AI also provides flexibility and accommodates novel battery cells because of the continuous learning and adapting capabilities.
The machine learning capabilities of AI controls grow stronger when fed real driving data from fleets of vehicles to train individual battery pack controls for situations they have yet to individually encounter. Using adaptive controls allows for precise battery pack reactions to all activity inside the battery or outside the vehicle.
​
Learn more about the EVE-Ai™ 360 Adaptive Controls features supported by applied AI
ADAPTIVE BATTERY DIGITAL TWIN​
Traditional BMS strategies base battery fault prediction on static battery cell models that do not account for the ever-changing condition of individual cells over time. Only the most advanced detection systems in research facilities, each too large to install onboard a vehicle, have achieved accurate battery modeling through laboratory testing using acoustic modeling and pulse discharge. Electra takes this advanced battery modeling approach one step further by taking the data from laboratory testing and applying it to battery fault prediction models. By training Electra’s AI battery pack controls on detailed lab simulations of cell modeling before installing the controls onboard vehicles, we are able to bridge the gap between laboratory and vehicle capabilities. Once installed, the AI models continuously learn from on-the-road data to update the model according to real-time battery pack usage. This allows the AI software to monitor and predict battery fault events like extreme capacity fade in time to correct battery usage for safer, more efficient charge and discharge strategies.
BATTERY PACK SAFETY
​
Electra understands that ensuring battery safety is of utmost importance when designing battery packs and creating custom AI battery controls. Our battery pack solutions always defer safety decisions to preexisting BMS safety measures to be sure we are not overriding essential parameters of battery management. Additionally, Electra’s battery pack controls prioritize extending battery lifetime by decreasing battery wear and strain, both of which ultimately contribute to safer battery utilization. Finally, our advanced, adaptive battery model leads the industry in battery fault prediction strategies to predict and limit battery fault events before they occur, further extending battery safety capabilities.
360 ADAPTIVE CONTROLS METHODOLOGY
​
Current battery packs are managed according to static control algorithms that use only basic system inputs. These controls do not account for the dynamic environmental factors acting on the battery pack’s performance throughout driving activity. However, at Electra we recognize that battery packs and the vehicles they power do not operate within a vacuum. Instead, an electric vehicle’s performance needs are influenced by external environmental factors as they charge, drive, and repeat.
Electra’s active control strategy intakes multiple factors from the energy storage chemistry performance, the individual driver’s behavior, and multiple environmental factors like weather, temperature, traffic patterns, and road conditions to inform battery pack control strategies. With machine learning, the more inputs that are analyzed and learned from, the more each individual battery pack within a fleet is ready to adapt to situations it has yet to encounter. This holistic approach to battery management is essential to enabling precise, efficient, and superior battery packs as the electric vehicle market expands across vehicle applications.
CLOUD CONNECTIVITY AND OVER-THE-AIR UPDATES
​
Current onboard computing capabilities of electric vehicles are limited by processing power and data storage, leaving little room for the complex onboard software necessary to control innovative energy storage systems. Cloud connectivity overcomes these limitations by allowing an individual EV to obtain data from vehicles around the world that are experiencing varying operating conditions while surpassing the hardware’s minimal data storage capacities.
The key to utilizing the power of Cloud connectivity are over-the-air updates, or OTA updates, that reduce the amount that the ECU/BMS processes by taking the computation and exploration processes of novel controls out of the onboard system. This methodology in turn allows for a broader collection of data that can be shared throughout the network without increasing onboard hardware cost for excess storage functionality. At Electra, our active and adaptive battery pack controls learn from fleet data fed historically and in real time from the Cloud. This data utilization in the form of AI ensures that individual battery packs are prepared for any instance that has already arisen in historical data that the control model is given.
LEVERAGING AUTOMOTIVE BIG DATA
​
In an age where vehicles are software-controlled, there is no lack of automotive data being generated on vehicle utilization, driver data, and electric vehicle battery systems. However, gathering insights from this data to inform future product designs and business decisions is not yet streamlined due to the complexity and sheer magnitude of the collected data. This gap between data collection and data analytics is imperative to fill in the EV sector in order to make informed advances in energy storage design and controls based on historical data collection.
​
Luckily, simultaneous advances in AI technology and Cloud connectivity allow for automated collection, refinement, and usage of these automotive data goldmines. Once data is transmitted from pre-existing sensors to the Cloud, it informs machine learning control updates which can be added to battery pack control strategies through Over-The-Air (OTA) updates. This data is additionally passed on to the backend of fleet management, OEMs, and engineers through our Analytics software to analyze and adapt to current data trends in vehicle and energy storage usage. Leveraging automotive data is the key to creating informed energy storage progress.
V2X COMMUNICATION (V2G, V2I, V2V)
​
V2X communication, or Vehicle-to-Everything communication, gives vehicles the ability to communicate with surrounding objects such as vehicles, pedestrians, infrastructure, the grid, and cloud data. With V2X communication, driving becomes safer and more efficient. Electra’s EVE-Ai™ Onboard Controls feature V2G, V2I, and V2V connectivity.
​
VEHICLE-TO-GRID COMMUNICATION​
​
V2G communication, or vehicle-to-grid communication, describes technology that allows electric vehicles to connect with the power grid to exchange power in either direction. When a strategic charge and discharge approach is taken, vehicles can smart charge in a way that anticipates and considers current energy supplies in the grid to help balance energy consumption. By connecting to the grid, the energy storage systems within each vehicle become a part of the greater grid storage ecosystem that can both give and take energy instead of isolated storage units. Vehicle to grid communication is not just the future of the electric vehicle industry, but also the renewable energy industry as a key solution to fighting climate change.
​
VEHICLE-TO-INFRASTRUCTURE COMMUNICATION​
​
V2I communication, or vehicle-to-infrastructure communication, describes the connectivity between electric vehicles and surrounding infrastructure. This includes traffic signals, traffic flow, Google Maps, and road surfaces. V2I connectivity drastically improves safety on the road by providing live feedback on surrounding conditions. Additionally, vehicle to infrastructure technology has the potential to improve energy efficiency by shortening drive times through vehicle adaptation to traffic signal data. Vehicle to infrastructure connectivity is essential for route optimization features in that it anticipates traffic signals and traffic flow to optimize vehicle velocity without compromising the length of travel or battery power demands.
​
VEHICLE-TO-VEHICLE COMMUNICATION
​
V2V communication, or vehicle-to-vehicle communication, describes the real-time connectivity between vehicles as they share data and interact with one another on the road. Vehicles share information such as their speed, location, position, destination, and vehicle description and dimensions. Vehicle-to-vehicle communication will allow for additional pathways for data sharing in the near future, further improving road safety and efficiency.
STATE OF HEALTH PREDICTION
​
Electric vehicle battery State of Health (SOH) refers to a numerical value that estimates the condition of the battery by dividing the maximum cell capacity by the original cell capacity. While this calculation is fairly elementary, the ability to predict an accurate estimation of SOH is lacking in current electric vehicle BMS.
The SOH of a battery is affected by various internal and external conditions, both physical and chemical. State of Health monitoring is key to predicting and classifying what causes battery degradation. Electra works to improve the estimation and prediction of cell capacity using our Adaptive Battery Digital Twin that provides a better understanding of the inner workings of battery pack cells throughout their lifetimes. AI SOH monitoring unlocks the ability to preserve battery health by receiving and implementing known data into the BMS and actively making corrections within the system to avoid harming the battery. Electra works to improve the estimation of battery SOH as well as predict the cell capacity of the electric vehicle without ever overriding BMS safety measures.
​
Learn more about SOH Prediction using EVE-Ai™ 360 Adaptive Controls
CELL CHEMISTRY AND BATTERY PACK CONFIGURATION
​
While the EV market continues to grow, it is constantly plagued by energy storage systems that fall short in performance, efficiency, and range. The rechargeable energy storage market is attempting to overcome these obstacles by testing various cell chemistries and configurations in battery packs, and Electra is interested in enabling this technology from all angles. Traditional Li-Ion batteries are well-explored and continue to be the only chemistry powering electric vehicles currently on the road, thanks to their relative affordability and high energy capabilities. But while Li-Ion batteries are tried and true, Novel Batteries and new configurations could be the key to unlocking the power of the next great energy storage system. Other promising advances are taking place among dual chemistry systems, in which the electric vehicle industry can use the capabilities of multiple battery units at once to create an ideal system for any application instead of waiting for the perfect next-generation cell to emerge on the market.
Electra’s design, control, and analytics technologies enable current Li-Ion configurations to optimize the chemistry’s usage, but also allow for exploration of the many other options just making their way onto the EV market. At Electra, we believe that exploring the potential of Novel Batteries used in single and dual chemistry systems is essential to future battery pack success, which is why our technology remains hardware and chemistry agnostic.
​
Learn more about dual-chemistry systems by downloading EnPower™ case study results
