Battery Informatics, Inc. (Bii) is developing the next generation Battery Management System (BMS) built using a model-based and application-aware approach. Our solutions optimize energy storage batteries to increase value and lifetime by using physics-based dynamic models of the internal battery behavior. The BMS is also tailored to the specific needs of the applications.

 

Our Core Technologies

 

icon-battery

Real-Time Models

Bii's BMS uses physical models to compute the internal state of the batteries; e.g., state of charge, internal temperature, and degree of degradation.

 

icon-bulb

Battery Degradation

The model is used to optimize the trade-off between cost of battery degradation versus the value from provided services.

icon-diagnostic

Diagnostics

Bii is developing automated testing procedures to experimentally determine battery health for comparison with, and calibration of, model-based results.

Our Features

Model-Based

The Li-ion batteries are modeled by solving a P2D (pseudo-two-dimensional) set of equations. The model estimates the state of charge, capacity fade, side reactions, electrode utilization efficiency and temperature distribution. UW has developed a method of model reduction, simplification and reformulation. This significantly reduces the computational cost of implementation on small chips.

 

Cost versus Benefits

Battery degradation is computed in terms of marginal cost of degradation which Bii expresses by Levelized Cost of Storage (LCOS). As result, we make a comparison between the benefits of the service performed by the battery versus the cost of delivering that service. The degradation cost is computed by our P2D model.

 

State Estimation

You normally only measure voltage, current, and external temperature. Information like internal temperature, SoC, SoE, and SoH need to be derived from a model. Some people use simple models combined with a Kalman Filter. Bii uses a detailed P2D model combined with self-learning to achieve the same effect with higher accuracy.

 

Optimized Charging

For those applications (e.g., e-bus) where you have a non-standard charger, our BMS will automatically determine a path that results in much less degradation than standard CC-CV (traditional method), step-wise CC-CC (common Super Fast charging), and other charging protocols.

 

Faster than Real-Time

To be useful for state estimation and control, the model must run faster than the data sampling interval. For use in EV drivetrains, the sampling interval is sub-seconds. For grid applications the time interval is typically 15 minutes.

 

Self-Learning

P2D model needs up to 28 parameters to adequately characterize the particular battery chemistry. Some of these parameters change during the life-time of the batteries. We are developing a self-learning capability to automatically estimate these parameters.

 

Runs on Small Micro-Processors

If you only need to monitor the average cell, you can run all the necessary calculations on a small processor. That will only add a few dollars to the overall cost of a BMS solution. If you want to monitor many cells, you presumably do it because your application warrants the cost of advanced control.

 

 

Superior Solution

If you use all the capabilities provided by Bii's BMS, you will be have the most economic and safe operation of Li-ion batteries that state-of-the-art battery control can provide. The solutions run on commonly available chips like TI and Xilinx.

 

Get in touch