Battery management system

1. A method comprising:

receiving, at a data processing hardware, current measurements from at least one current sensor configured to measure current of a battery system in communication with a power distribution network having a power plant distributing power to one or more consumers;

receiving, at the data processing hardware, voltage measurements from at least one voltage sensor configured to measure voltage of the battery system;

receiving, at the data processing hardware, temperature measurements from at least one temperature sensor configured to measure temperature of the battery system;

determining, by the data processing hardware, an impedance parameter of the battery system based on the measurements received at the data processing hardware;

determining, by the data processing hardware, a temperature parameter of the battery system based on the measurements received at the data processing hardware;

determining, by the data processing hardware, a predicted voltage parameter based on the impedance parameter;

determining, by the data processing hardware, a predicted temperature parameter based on the temperature parameter; and

commanding, by the data processing hardware, the battery system to charge power from the power plant if an amount of power consumed in the power distribution network is below an amount of power produced in the power plant, or discharge power from the battery system if the amount of power consumed in the power distribution network exceeds the amount of power produced in the power plant based on the predicted voltage parameter and the predicted temperature parameter,

wherein determining one of the predicted voltage parameter or the predicted temperature parameter includes executing a time series analysis implementing an auto-regressive model AR(p), wherein an optimal solution of the auto-regressive model AR(p) is determined by implementing a Yule-Walker approach.

2. The method of claim 1 , wherein the auto-regressive model AR(p) is defined as:

X t =c+Σ i=1 p φ i X t−i +∈ i , and

wherein φ 1 -φ p are parameters of the auto-regressive model AR(p), c is a constant, and ϵ t is white noise.

3. The method of claim 1 , wherein commanding the battery system to charge power from the power plant includes commanding the battery system to store power from the power plant.

4. The method of claim 1 , further comprising updating an impedance profile, a voltage profile, or a temperature profile based on the voltage measurements or the temperature measurements.

5. The method of claim 1 , wherein determining the predicted voltage parameter includes:

training the data processing hardware to generate a best fit of the voltage measurements or the temperature measurements; and

predicting, by the data processing hardware, the predicted voltage parameter or the predicted temperature parameter based on the best fit of the voltage measurements or the temperature measurements, respectively.

6. The method of claim 1 , further comprising:

tracking, by the data processing hardware, a remaining available capacity of the battery system; and

determining, by the data processing hardware, one of a charge state or life cycle of the battery system.

7. A system comprising:

a data processing hardware; and

a memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:

receiving current measurements from at least one current sensor configured to measure current of a battery system in communication with a power distribution network having a power plant distributing power to one or more consumers;

receiving voltage measurements from at least one voltage sensor configured to measure voltage of the battery system;

receiving temperature measurements from at least one temperature sensor configured to measure temperature of the battery system;

determining an impedance parameter of the battery system based on the measurements received at the data processing hardware;

determining a temperature parameter of the battery system based on the measurements received at the data processing hardware;

determining a predicted voltage parameter based on the impedance parameter;

determining a predicted temperature parameter based on the temperature parameter; and

commanding the battery system to charge power from the power plant if an amount of power consumed in the power distribution network is below an amount of power produced in the power plant, or discharge power from the battery system if the amount of power consumed in the power distribution network exceeds the amount of power produced in the power plant based on the predicted voltage parameter and the predicted temperature parameter,

wherein determining one of the predicted voltage parameter or the predicted temperature parameter includes executing a time series analysis implementing an auto-regressive model AR(p), wherein an optimal solution of the auto-regressive model AR(p) is determined by implementing a Yule-Walker approach.

8. The system of claim 7 , wherein the auto-regressive model AR(p) is defined as:

X t =c+Σ i=1 p φ i X t−i +∈ i , and

wherein φ 1 -φ p are parameters of the auto-regressive model AR(p), c is a constant, and ϵ t is white noise.

9. The system of claim 7 , wherein commanding the battery system to charge power from the power plant includes commanding the battery system to store power from the power plant.

10. The system of claim 7 , wherein the instructions, when executed on the data processing hardware, cause the data processing hardware to perform operations comprising:

updating an impedance profile, a voltage profile, or a temperature profile based on the voltage measurements or the temperature measurements.

11. The system of claim 7 , wherein determining the predicted voltage parameter includes:

training the data processing hardware to generate a best fit of the voltage measurements or the temperature measurements; and

predicting, by the data processing hardware, the predicted voltage parameter or the predicted temperature parameter based on the best fit of the voltage measurements or the temperature measurements, respectively.

12. The system of claim 7 , wherein the instructions, when executed on the data processing hardware, cause the data processing hardware to perform operations comprising:

tracking a remaining available capacity of the battery system; and

determining one of a charge state or life cycle of the battery system.