Abstract: A battery stack includes a plurality of battery management system (BMS) nodes and a controller. Each BMS node includes a battery, an isolation switch configured to selectably isolate the battery of the BMS node from the batteries of the other BMS nodes, and a bypass switch configured to selectably provide a path for electrical current flowing through the battery stack to bypass the battery of the BMS node. The batteries of the BMS nodes are electrically coupled in series. The controller is configured to control the isolation switch and the bypass switch of each BMS node such that the battery of each BMS node can be individually connected to and disconnected from an electrical power source/sink.

Abstract: A machine learning model may be trained for a target battery system. Reference initialization data may be generated by a reference battery system sharing one or more characteristics with the target battery system. Synthetic training data may be generated for the target battery system by determining simulated input parameters and simulated output parameters by supplying the simulated input parameters to a physics model. Upon determining that a subset of the reference initialization data matches a subset of the synthetic training data, a trained machine learning model based on the synthetic training data is determined. The trained machine learning model may be trained to predict a state of the target battery system based at least in part on one or more observed input parameters generated by the target battery system.

Abstract: A method for thermal management performed by a controller of an energy storage system, where the energy storage system includes at least a first battery module, a second battery module, a first battery management system (BMS) node, and a second BMS node. The first BMS node is configured to control operation of the first battery module, and the second BMS node is configured to control operation of the second battery module. The method includes (a) determining a first temperature profile difference representing a difference between an actual temperature profile of the first battery module and a desired temperature profile of the first battery module, (b) determining a first operation adjustment representing a desired change in operation of the first battery module for decreasing the first temperature profile difference, and (c) controlling the first BMS node to change operation of the first battery module according to the first operation adjustment.

Abstract: Described herein are methods and systems for updating SOC estimates of individual cells in battery packs. Specifically, SOC estimates are updated in-situ, e.g., while the battery packs remain operational. For example, a cell is charged or discharged, independently from other cells, until the cell OCV is at a set value, corresponding to one of target zones. The target zones have more prominent correlations between the OCV and SOC than other parts of the OCV profile. A new SOC value, corresponding to the cell OCV, is used to update the SOC estimate. In some examples, a set of voltages is obtained while the cell is charged or discharged at a constant current/power, e.g., outside of the target zones. One or more differential capacities are determined from this voltage set, and a new SOC value is obtained based on these differential capacities.

Abstract: Described methods and systems provide in-situ leakage current testing of battery cells in battery packs even while these packs operate. Specifically, an external electrical current is discontinued through a tested battery cell using a node controller, to which the tested battery cell is independently connected. Changes in the open circuit voltage (OCV) are then detected by the node controller for a set period time. Any voltage change, associated with taking the tested cell offline, is compensated by one or more other cells in the battery pack. The overall pack current and voltage remains substantially unchanged (based on the application demands), while the in-situ leakage current testing is initiated, performed, and/or completed. The OCV changes are then used to determine the leakage current of the tested cell and, in some examples, to determine the state of health of this cell and/or adjust the operating parameters of this cell.

Abstract: A method for managing a plurality of stacks of electrochemical cells, where the plurality of stacks are electrically coupled in parallel in a battery. The method includes (a) operating the plurality of stacks to execute a global operating strategy, (b) controlling one or more first power converters to change operation of one or more first stacks of the plurality of stacks to execute a first local operating strategy of operating the one or more first stacks at one of a constant power and a constant current, and (c) controlling one or more second power converters to change operation of one or more second stacks of the plurality of stacks to compensate for change in operation of the one or more first stacks caused by executing the first local operating strategy, and thereby maintain the global operating strategy of the battery while executing the first local operating strategy.

Abstract: Battery node diagnostic data may be received from a battery system. One or more outlier detection machine learning models may be selected based on profile information included in the battery node diagnostic data. The profile information may identify a battery node operation profile associated with some or all of the battery node diagnostic data. One or more of the battery nodes may be identified as outliers by applying the one or more outlier detection machine learning models to identify one or more differences between first diagnostic data for the designated subset of the battery nodes and a population-level representation of the battery node diagnostic data. Outcome values may be determined by applying one or more predetermined rules to the battery node diagnostic data. A battery node may be identified as exhibiting a fault based on the designated subset of the battery nodes and the plurality of outcome values.

Abstract: A method for adaptive electrochemical cell management in an energy storage system including a plurality of battery management system (BMS) nodes, the method including (1) obtaining a first signal identifying one or more degradation mechanisms of a first cell assembly of a first BMS node of the plurality of BMS nodes, the first cell assembly including one or more first electrochemical cells, and (2) controlling a first BMS node controller of the first BMS node in response to the first signal, to change a state of operation of the first cell assembly to mitigate the one or more degradation mechanisms of the first cell assembly, independently of operation of a second BMS node of the plurality of BMS nodes.

Abstract: Signal data may be received from battery node controllers corresponding with battery nodes in a battery system operating in a charging or discharging state. A battery node controller may selectively couple a power bus in the battery system with a battery node including a respective one or more battery cells. Candidate events may be identified as anomalous based on a discontinuous changes in the signal data. The candidate events may correspond to a respective battery node and a respective period of time. When one or more candidate events are identified as anomalous, diagnostic data may be sent to a remote computing system over the internet via a communication interface. An indication of a designated battery node as defective based on one or more machine learning models applied to the diagnostic data at the remote computing system may be received. A battery node operation profile associated with the designated battery node may be updated.

Abstract: Battery system diagnostic data may be generated at battery node controllers in a battery system. A battery node controller may selectively couple a power bus with a battery node including one or more battery cells. An unsupervised machine learning model may be trained based on the battery system diagnostic data. Outlier values may be determined by applying the machine learning model to the battery node diagnostic data. A battery node may be identified as anomalous based on a comparison of a designated outlier value associated with the designated battery node with the plurality of outlier values. A battery node operation profile controlling operation of the designated battery node may be updated.

Abstract: A method for managing a plurality of batteries that are electrically coupled together includes (1) monitoring respective voltages of the plurality of batteries and (2) in response to a respective voltage of a first battery of the plurality of batteries reaching a first threshold value at a first time, reducing a charge or discharge rate of the first battery, relative to at least a second battery of the plurality of batteries. Charge and discharge rates may be adaptively managed such that each battery reaches the first threshold value at substantially the same time.

Abstract: A battery stack includes a plurality of battery management system (BMS) nodes and a controller. Each BMS node includes a battery, an isolation switch configured to selectably isolate the battery of the BMS node from the batteries of the other BMS nodes, and a bypass switch configured to selectably provide a path for electrical current flowing through the battery stack to bypass the battery of the BMS node. The batteries of the BMS nodes are electrically coupled in series. The controller is configured to control the isolation switch and the bypass switch of each BMS node such that the battery of each BMS node can be individually connected to and disconnected from an electrical power source/sink.

Abstract: Described methods and systems provide in-situ leakage current testing of battery cells in battery packs even while these packs operate. Specifically, an external electrical current is discontinued through a tested battery cell using a node controller, to which the tested battery cell is independently connected. Changes in the open circuit voltage (OCV) are then detected by the node controller for a set period time. Any voltage change, associated with taking the tested cell offline, is compensated by one or more other cells in the battery pack. The overall pack current and voltage remains substantially unchanged (based on the application demands), while the in-situ leakage current testing is initiated, performed, and/or completed. The OCV changes are then used to determine the leakage current of the tested cell and, in some examples, to determine the state of health of this cell and/or adjust the operating parameters of this cell.

Abstract: Described methods and systems are used for in-situ impedance spectroscopy analysis of battery cells in multi-cell battery packs. Specifically, the cell impedances are determined while the pack continues to operate, such as being charged or discharged. For example, the pack voltage/power output remains unchanged while this analysis is initiated, performed, and ended. Cell impedance is determined based on the cell's response to the signal applied to the cell. For example, a current through the cell is charged while monitoring cells' voltage response. Although the power output of the changes during this testing, but the operation of the pack is not impacted due to the power compensation provided by one or more other cells in the pack thereby ensuring uninterrupted operation of the pack. This in situ testing is provided by the unique architecture of the pack, comprising multiple nodes and individual node controllers.

Abstract: A system and method for hierarchical arc fault monitoring in an energy storage system, where the energy storage system includes a plurality of stacks that are electrically coupled together. Each stack includes a plurality of battery management system nodes that are electrically coupled together. The method includes (1) obtaining respective electrical measurement values for each stack; (2) determining, for each stack, that the stack is free of arc faults, using the respective electrical measurement values for the stack; (3) obtaining electrical measurement values for the energy storage system; and (4) determining that the energy storage system is free of arc faults outside of the plurality of stacks, using (a) the electrical measurement values for the energy storage system and (b) a subset of the respective electrical measurement values for each stack.

Abstract: A method for managing a plurality of stacks of electrochemical cells, where the plurality of stacks are electrically coupled in parallel in a battery. The method includes (a) operating the plurality of stacks to execute a global operating strategy of the battery, (b) changing respective operating points of one or more first stacks of the plurality of stacks to execute a local operating strategy, and (c) changing respective operating points of one or more second stacks of the plurality of stacks to maintain the global operating strategy of the battery while executing the local operating strategy.

Abstract: Described methods and systems provide in-situ leakage current testing of battery cells in battery packs even while these packs operate. Specifically, an external electrical current is discontinued through a tested battery cell using a node controller, to which the tested battery cell is independently connected. Changes in the open circuit voltage (OCV) are then detected by the node controller for a set period time. Any voltage change, associated with taking the tested cell offline, is compensated by one or more other cells in the battery pack. The overall pack current and voltage remains substantially unchanged (based on the application demands), while the in-situ leakage current testing is initiated, performed, and/or completed. The OCV changes are then used to determine the leakage current of the tested cell and, in some examples, to determine the state of health of this cell and/or adjust the operating parameters of this cell.

Abstract: Described methods and systems are used for in-situ impedance spectroscopy analysis of battery cells in multi-cell battery packs. Specifically, the cell impedances are determined while the pack continues to operate, such as being charged or discharged. For example, the pack voltage/power output remains unchanged while this analysis is initiated, performed, and ended. Cell impedance is determined based on the cell's response to the signal applied to the cell. For example, a current through the cell is charged while monitoring cells' voltage response. Although the power output of the changes during this testing, but the operation of the pack is not impacted due to the power compensation provided by one or more other cells in the pack thereby ensuring uninterrupted operation of the pack. This in situ testing is provided by the unique architecture of the pack, comprising multiple nodes and individual node controllers.

Abstract: Described herein are methods and systems for detecting variation in minor total-impedance contributors in sets of electrochemical cells. For example, a method comprises maintaining a substantially constant current through the set of electrochemical cells and obtaining multiple voltage readings of the cells while the substantially constant current is maintained. The method then proceeds with determining multiple differential capacity values from the multiple voltage readings, characterizing one or more peaks in the multiple differential capacity values, and determining the variation in the minor total-impedance contributor based on one or more peaks. More specifically, partial capacitance values can be assigned to different impedance channels based on these peaks or, more specifically, based on the separation of adjacent peaks.

Abstract: Described herein are methods and systems for updating SOC estimates of individual cells in battery packs. Specifically, SOC estimates are updated in-situ, e.g., while the battery packs remain operational. For example, a cell is charged or discharged, independently from other cells, until the cell OCV is at a set value, corresponding to one of target zones. The target zones have more prominent correlations between the OCV and SOC than other parts of the OCV profile. A new SOC value, corresponding to the cell OCV, is used to update the SOC estimate. In some examples, a set of voltages is obtained while the cell is charged or discharged at a constant current/power, e.g., outside of the target zones. One or more differential capacities are determined from this voltage set, and a new SOC value is obtained based on these differential capacities.