System and method for detecting degradation of battery energy storage system and warranty tracking thereof

Abstract

The invention relates to a system and method for detecting degradation of a battery energy storage system and warranty tracking thereof. A system and method for detecting degradation of a storage battery of a battery energy storage system (BESS) container includes a battery management system (BMS) coupled to the storage battery, the battery management system (BMS) configured to collect battery operation data from the storage battery. A battery data repository coupled to the BMS receives and stores battery operation data. A learning agent coupled to the battery data repository uses the stored battery operation data to estimate prognostic degradation of the storage battery and train a degradation estimation model. A state of charge (SOC) estimation agent receives and uses the trained degradation estimation model to generate a real-time estimate of SOC of the storage battery for scheduling charge / discharge cycles of the BESS container. The system and method includes updating storage battery model parameters based on data generated during battery charge and discharge cycles, the system and method checking and verifying the battery model parameters using expert users, and a system and method for protecting the integrity of battery data for warranty tracking of storage batteries for BESS.

Description

Technical Field

[0001] The present disclosure relates generally to battery energy storage systems. More specifically, the present disclosure relates to a system and method for detecting degradation of a battery energy storage system and tracking the battery energy storage system for replacement under warranty. Background Art

[0002] At present, most electricity is generated by large centralized power plants, such as nuclear power plants, hydroelectric power plants, and fossil fuel power plants. These large facilities often use non-renewable energy sources to generate electricity, such as coal or natural gas. Such power plants usually have good economies of scale, but due to various economic and operational reasons, it may not be possible to provide all the power required for the loads served by the service grid through centralized power plants. Battery energy storage systems that store electricity can be connected at power plants, substations, transmission lines, or customer sites to selectively use the stored battery energy to supplement or provide all the electricity required by the grid, thereby preventing service interruptions. Battery energy storage systems use chemical energy storage batteries that store energy chemically, such as, for example, lithium iron (LiON) batteries, lead-acid batteries (Pb) or sodium sulfur (NAS) batteries.

[0003] The degradation of battery energy storage systems, especially those based on LiON cells, is an irreversible process that depends on the current and temperature to which the battery cell is subjected when a voltage or potential difference is applied across its terminals. The battery cell voltage depends on the state of charge in the cell. When the voltage across the battery terminals falls below a minimum threshold, the LiON battery chemistry becomes unstable. In this case, the cell impedance will also increase, and moderate to high charging currents may cause a temperature rise, which exacerbates the degradation of the battery cell. A similar situation exists at high voltages when overcharging leads to faster degradation of the battery cell.

[0004] Protecting the integrity of battery data regarding battery degradation is also desirable for suppliers of battery energy storage systems because the data is susceptible to tampering, especially by users who may subject the battery system to abuse cycles and still require warranty costs or replacement. The battery warranty tracking system is intended to make battery data reflecting the health of the battery energy storage system tamper-proof, making any tampered data unusable for warranty claim purposes.

[0005] The present disclosure describes systems and methods for detecting degradation of a battery energy storage system by detecting the state of charge of the battery management system to enable automatic calibration of the battery management system and tracking battery degradation for replacement of the battery energy storage system under warranty. Summary of the invention

[0006] The present disclosure relates to a system and method for detecting degradation of a battery energy storage system by detecting the state of charge of a battery management system and tracking battery degradation for replacement under warranty.

[0007] In a first embodiment, a system for detecting degradation of a storage battery of at least one battery energy storage system (BESS) container is disclosed. The system includes a battery management system (BMS) coupled to the storage battery, configured to collect battery operation data from the storage battery. A battery data repository coupled to the BMS receives and stores the battery operation data. A learning agent coupled to the battery data repository uses the stored battery operation data to estimate prognostic degradation of the storage battery and train a degradation estimation model, wherein a state of charge (SOC) estimation agent receives and uses the trained degradation estimation model to generate a real-time estimate of the SOC of the storage battery for scheduling a charge / discharge cycle for the BESS container.

[0008] The system also includes a check and verification of battery parameters by an expert user, including a battery digital twin configured to receive operational data including charge / discharge current and ambient temperature from a BESS container to generate a simulated voltage response and a simulated temperature rise. A mean square error model receives an actual voltage response and an actual temperature rise from the BMS and a simulated voltage response and a simulated temperature from the battery digital twin, and is configured to output a mean square error output. A hybrid parameter identification model receives the mean square error output to generate an estimate of a change in the battery parameter based on the charge / discharge current and ambient temperature input to the hybrid parameter identification model, wherein the change in the battery parameter is uploaded to the expert user for checking and verification of the changed battery parameter.

[0009] The warranty tracking system includes storing battery operation data as a plain text file in a BESS container, wherein a private key generator coupled to the RNN battery model is configured to send simulated battery data and a unique identifier to the private key generator, which generates a private encryption key coupled to an encryption agent. The plain text file is encrypted into an encrypted text file by the encryption agent using the private encryption key. A data storage platform located away from the BESS container has a data repository coupled to the encryption agent and is arranged to receive and store the encrypted text file. A key generator located on the data storage platform coupled to the RNN battery model generates a public encryption key using the simulated battery data from the RNN battery model. Based on a request from a user, a decryption agent coupled to the data repository receives the encrypted text file and the public encryption key and decrypts the encrypted text file into a plain text file.

[0010] In a second embodiment, a method for detecting degradation of a storage battery of at least one battery energy storage system (BESS) container is disclosed. The method includes collecting battery operation data from a battery management system (BMS) coupled to the storage battery and storing the battery operation data in a data repository. A degradation estimation model is trained using the stored battery operation data to estimate a prognostic degradation of the storage battery. The method also includes using the trained degradation estimation model to generate a real-time estimate of the state of charge (SOC) of the storage battery for scheduling a charge / discharge cycle of the storage battery.

[0011] The method also includes checking and verifying the battery parameters by an expert user, including receiving operational data from the BESS container through the battery digital twin, the operational data including charge / discharge current and ambient temperature. Using the operational data to generate a simulated voltage response and a simulated temperature rise through the battery digital twin. Generating a mean square error output using the actual voltage response and actual temperature rise from the BMS and the simulated voltage response and simulated temperature from the digital twin to generate an estimate of the change in the battery parameter based on the charge / discharge current and ambient temperature using the mean square error output, wherein the change in the battery parameter is presented to the expert user for checking and verifying the changed battery parameter.

[0012] A warranty tracking method includes storing battery operation data as a plain text file in a BESS container. A private key generator is coupled to an RNN battery model, the RNN battery model is configured to send simulated battery data and a unique identifier to the private key generator to generate a private encryption key, wherein the private encryption key is used by an encryption agent to encrypt the plain text file. The encrypted text file is stored in a data repository of a data storage platform located away from the BESS container. A key generator located on the data storage platform is coupled to the RNN battery model and is configured to generate a public encryption key using the simulated battery data from the RNN battery model. The method also includes sending the encrypted text file and the public encryption key to a decryption agent upon a user's request, the decryption agent decrypting the encrypted text file into a plain text file.

[0013] In a third embodiment, a computer program product including a non-transitory data storage medium is disclosed, comprising program instructions executable by a processor to enable the processor to perform a method for detecting degradation of a battery of at least one battery energy storage system, the method comprising collecting battery operation data from the storage battery and storing the battery operation data in a data repository, training a degradation estimation model using the stored battery operation data to estimate prognostic degradation of the storage battery, and using the trained degradation estimation model to generate a real-time estimate of a state of charge of the storage battery.

[0014] Other technical features will be apparent to those skilled in the art from the following drawings, descriptions and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] For a more complete understanding of the present disclosure, reference is now made to the following description taken in conjunction with the accompanying drawings, in which:

[0016] Figure 1 is a diagram schematically illustrating a multi-level control system for controlling a battery energy storage system;

[0017] Figure 2 is a diagram schematically illustrating a battery energy storage system container;

[0018] Figure 3 is a diagram schematically illustrating an energy control system for controlling a battery energy storage system;

[0019] Figure 4 is a diagram schematically illustrating an embodiment of an exemplary system for detecting degradation of a battery energy storage system according to the present disclosure;

[0020] Figure 5 is a diagram schematically illustrating an exemplary battery rack digital twin for detecting degradation of a battery energy storage system according to the present disclosure;

[0021] Figure 6 is a diagram schematically illustrating an exemplary battery module digital twin for detecting degradation of a battery energy storage system according to the present disclosure;

[0022] Figure 7 is a diagram schematically illustrating an embodiment of an exemplary system for detecting degradation of a battery energy storage system according to the present disclosure, the exemplary system using an expert user to check and validate battery model parameters;

[0023] Figure 8 is a diagram schematically illustrating an embodiment of an exemplary system for detecting degradation of a battery energy storage system according to the present disclosure, the exemplary system including a warranty tracking system for protecting the integrity of battery data of the battery storage system; and

[0024] Fig. 9 is a diagram schematically illustrating a data storage platform according to the present disclosure. DETAILED DESCRIPTION

[0025] These figures discussed below and the various embodiments for describing the principles of the present invention in this patent document are carried out by way of example only and should not be construed as limiting the scope of the present invention in any way. Those skilled in the art will appreciate that the principles of the present invention can be implemented in any type of suitably arranged device or system.

[0026] The degradation of storage batteries of a battery energy storage system (BESS) is not linear over the entire range of voltage, current, and temperature of the battery cells contained in the BESS. Assuming linear approximations in the estimation of the state of charge (SOC) based on the battery cell voltage and current measured at different temperatures may lead to errors in the estimation of the SOC of the battery cells of the BESS.

[0027] The battery cell SOC is typically calibrated when the BESS is installed and commissioned and before it is used. Calibration is also required periodically, such as once every three to six months. The calibration process requires the BESS to be taken out of service and cycled from a low SOC limit to a high SOC limit multiple times. This is time consuming and renders the BESS unusable while the calibration is performed. The calibration training also uses some of the remaining useful life of the BESS battery and therefore reduces the life of the BESS storage battery. Accurate estimates of the SOC between calibrations are useful to avoid degradation due to inaccurate estimates of the SOC. When the reduction in the BESS energy capacity is not accurately estimated and an incorrect state of charge is calculated, shallow battery cycling becomes prone to errors in the estimate when a higher or lower SOC limit is reached. For example, if the battery cells of the BESS are regularly cycled between 30% and 70% SOC and only a few cycles extend to 95% or 20%, an incorrect estimate of the energy capacity may cause the BESS to be charged to a higher than expected 95% or lower than the expected 20% due to a reduction in energy capacity. If the battery cells of the BESS storage battery are stored at high or low SOC limits, and if the temperature during storage is high, they also degrade faster than usual.

[0028] Field data collected during operation of the BESS can be used to provide accurate estimates of the storage battery SOC and the overall state of health (SOH) of the BESS storage battery between calibrations. The battery management system (BMS) does not use field data generated between calibration cycles during BESS operation to make accurate estimates of battery cell SOC. Accurate BESS SOH and energy capacity estimates will also guide early maintenance, which ensures that the battery is operating in a safe range for the measured operating voltage or estimated SOC.

[0029] A battery energy control system (BECS) can be used to schedule BESS operation charging and discharging cycles to meet a variety of use cases. BECS can also be tasked with providing power control from the BESS to the grid or microgrid. BECS can provide foresight operations using predictions of different variables that affect the operation of the BESS. For example, such as generation, power demand, the cost of power from different power sources at different times of the day, time-of-use tariffs from the grid, peak power costs due to demand charges or lower cost power from renewable energy sources and even higher cost power from diesel or gas turbine generators.

[0030] The BECS needs to know all of the economic costs associated with each power source required to charge the BESS. It is also necessary to know the economic costs associated with cycling the BESS. More specifically, the BECS needs to know the costs associated with the degradation of the BESS cells when the BESS is charged from another power source or discharged at another time to meet a different load (including when it is idle for a short or long duration at some other time). Accurate estimates of degradation costs require battery prognostic methods and systems that estimate degradation based on a future schedule prepared by the BECS. Degradation and associated costs are considered in an iterative optimization algorithm that runs at the BECS to generate a schedule for charging and discharging the BESS.

[0031] Suppliers and manufacturers of battery cells included in the storage batteries of BESS provide warranty commitments regarding maximum degradation at the end of a specific operating cycle, such as every 6 to 12 months. For example, the degradation in the first year may be about 5%, while the degradation in the 2nd, 3rd and 4th years may be about 2.5%. In the 5th and 6th years, it may be different, and the degradation is non-linear most of the time, with a higher degradation rate in the first year, followed by a lower degradation rate in the next few years and a higher degradation rate in the following years. Early detection of battery cell degradation is more important because the cell is unlikely to show typical symptoms associated with long-term degradation during early cycles, except when the cell has a lower life due to manufacturing defects or changes that need to be identified early. It may be uneconomical to subject all battery cells of the BESS to highly accelerated life cycle testing because it may degrade more cells in order to identify a small percentage of cells with manufacturing defects or changes that cause them to have a lower life. If a battery cell is found to be degraded faster than expected, the user can ask the supplier for early replacement. Until replacement is completed, the ECS may schedule such batteries to be not cycled at high charge and discharge currents taking into account their higher than normal degradation cost, to also have tighter temperature control when such batteries are cycled at higher C-rates. The warranty tracking system is designed to track all battery cells within the battery modules of the BESS storage battery. The battery module is the lowest replaceable unit in the BESS storage battery, and tracking of battery cell health becomes important to identify weaker outlier cells within the battery module that may limit the life of the entire battery module. The warranty tracking system provides a warning to the user when faster than normal degradation is detected and a higher than normal degradation cost is incurred to cycle such battery cells.

[0032] Figure 1 An exemplary multi-level control system for controlling a battery energy storage system (BESS) is shown. Figure 1 As shown, the multi-level control system 100 consists of four hierarchical control levels. At the first control level, a BESS unit controller 110 is located in a BESS container 120. The BESS unit controller 110 is used to control the functions of the BESS container 120 and its power conversion assets. Each BESS container 120 is organized as an independent package that may include at least a power conversion system, a battery system, a heating ventilation and air conditioning (HVAC) system, a fire protection system, and components and sensors required to monitor the BESS container 120. Each BESS container 120 can be used to power an independent deployment of a BESS 120 container and its associated multi-level control system 100, such as, for example, a building or commercial enterprise or a microgrid deployment, where a single BESS container 120 or multiple BESS containers 120 can provide power to a residential area or a commercial area.

[0033] At the second control level, an energy control system (ECS) 130 is communicatively coupled to one or more BESS unit controllers 110, 110'. The ECS 130 includes a BESS ECS controller 135 and a microgrid ECS (MECS) controller 140 that controls the operation of one or more BESS containers 120. The ECS 130 may be connected to a standalone BESS container 120 deployment or a plurality of BESS containers 120 deployments, or to a grid-connected multi-container BESS deployment. For example, in Figure 1 , the ECS 130 is shown connected to the BESS unit controller 110 and the BESS unit controller 110 ′ of the BESS container 120 ′.

[0034] The MECS controller 140 comprises a third control level of the multi-level control system 100. The MECS controller 140 is communicatively coupled to the BESS ECS controller 135 and manages alternative power generation assets, such as, for example, solar, wind, hydroelectric, which are connectable to and available on the grid for use by the BESS container 120. The MECS controller 140 is arranged to provide alternative power capabilities to a standalone BESS container 120 deployment or a plurality of microgrid-connected BESS container deployments.

[0035] The fourth control level of the control system 100 includes a virtual power plant (VPP) 160. The VPP 160 includes distributed small and medium-sized power generation units, loads, and energy storage systems that perform functions equivalent to centralized physical power plants when aggregated and coordinated using software. A software operating program executed on, for example, a server 162 serves as a controller for controlling the VPP 160. The VPP 160 also includes an operator station 164 and an interface to a cloud 168. The server 162 can be any device that provides resources, data, services, or software programs to other processing devices or clients over a network. The operator station 164 can be any computing device that provides functions for power plant operation and monitoring, including displaying graphics such as diagrams, systems, BESS container 120 deployments, and data to a user or operator. The operator station 164 can also receive input from a user or operator to adjust or input configurable parameters for the BESS unit controller 120, the BESS ECS controller 135, and the MECS controller 140. The cloud 168 may be any computing device or technology that provides services over the Internet, including information, data storage, servers, access to databases, networking, and software. The VPP 160 may control multiple BESS containers 120 connected to the VPP 160 via the communication network 108. Figure 1 The VPP 160 shown controls BESS containers 120, 120' in a multi-grid deployment, such as Figure 1 Microgrid 1 and microgrid 2 are shown.

[0036] The task of the BESS unit controller 110 is to provide safe and reliable operation for the BESS container 120. The BESS unit controller 110 monitors the operation of the BESS container 120, thereby preventing operation during fault conditions, shutting down faulty subsystems, and / or sending notifications and alarms to the operator station 164 or mobile device 240. If a component, sensor, or subsystem of the BESS unit container 120 fails or becomes faulty, alarms may be sent using different priority levels. The BESS unit controller 110 interfaces with all subsystems within the container, such as the power conversion system, battery system, HVAC subsystem, fire protection system, etc. The main components of the BESS unit container 120 include the BESS unit controller 110, the battery rack 305 housed within the BESS container 120, the power container 310 housed on a separate transportable skid, and the ECS 130.

[0037] like Figure 2 As shown, the BESS unit controller 110 includes at least one processor 301, at least one memory device 302, and at least one I / O interface 320. The processor 301 executes instructions that may be loaded into the memory 302. The processor 301 may include any suitable number and type of processing or other devices in any suitable arrangement. Exemplary types of processing devices include microprocessors, microcontrollers, digital signal processors, field programmable gate arrays, application specific integrated circuits, and discrete circuits. The processor 301 executes various programs for operating various operating modes, states, and safety systems of the BESS container 120.

[0038] Memory 302 represents any structure capable of storing information (such as data, program code, and / or other suitable information, whether temporary or permanent) and facilitating retrieval of the information. Memory 302 may represent random access memory or any other suitable volatile or non-volatile storage device. Memory may also include one or more components or devices that support longer-term storage of data, such as read-only memory, a hard disk, flash memory, or an optical disk.

[0039] The I / O interface supports communication with other systems or devices contained in the BESS container 120. For example, the communication interface 320 may include I / O modules and network interface cards that facilitate communication between the BESS unit controller 110 and the subsystems and sensors of the BESS container 120 and communication with levels 2-4 of the BESS control system 100. The I / O unit 320 may support communication through any suitable physical wired communication link or bus. For example, the I / O interface 320 may include an I / O module that can transmit control signals through the interface to the connected HVAC systems HVAC-1 and HVAC-2 through the comstat 4HVAC controller 350 using serial digital outputs. The I / O interface 320 may also include an analog module that can receive 4mAh-20mAh current loop signals from various analog sensors located in the BESS container 120, such as, for example, a temperature sensor 352, an air speed sensor 354, a pressure sensor and transmitter 356, and a relative humidity (RH) sensor 358. The I / O interface 320 also includes an Ethernet interface for bidirectional communication of control signals and data between the battery management unit (BAU) 360, the ECS 130, and various fire safety equipment, such as a fire detection panel 362, a LiON tamer 364, and a gas detector 366, which detects hydrogen gas that may be generated when lithium batteries degrade.

[0040] The BESS container 120 includes a storage battery, hereinafter referred to as a BESS battery 304. The BESS battery includes a plurality of battery modules that are electrically interconnected in series and also in parallel. Each battery module includes a plurality of battery cells that are electrically interconnected in series and also in parallel. The plurality of battery modules form a plurality of battery racks 305 stacked within the BESS container 120. The charging and discharging of the BESS battery 304 takes into account the state of charge (SOC) of the battery modules that comprise the BESS battery 304, and ensures that charging does not result in increased power dissipation and heating of the individual cells.

[0041] Each BESS container 120 is also connected to a power container 310, which is communicatively coupled to the BESS container 120. The power container 310 typically includes a bidirectional power conversion system (PCS) 315 and its associated components. The PCS 315 converts the AC voltage supplied by the grid into a DC voltage to charge the BESS storage battery, or converts the DC power provided by the BESS battery 304 battery rack 305 into an AC voltage to provide power to a connected building, residence, or microgrid. The power container 310 may also include a low voltage (LT) switchgear 327 and a transformer 319 to provide power to a low voltage or low voltage electrical network.

[0042] The BESS unit controller 110 is also operatively connected to a BESS ECS controller 135 of the ECS 130. The BESS ECS controller 135 acts as a supervisory controller for one or more BESS unit controllers 110. Figure 1 As shown, the ECS 130 may control one or more BESS containers using the control network 109. The BESS unit controller 110 of each BESS container 120 collects operating parameters of the connected BESS units 120 and sends the data to its supervisory BESS ECS controller 135 for controlling the charging requirements and discharging requirements of the BESS container 120. For example, the BESS ECS controller 135 calculates a power reference for each PCS 315 attached to one or more BESS containers 120, taking into account the current operating status of the BESS container 120, such as alarms related to subsystem failures or failures and diagnostic data of key subsystems (such as battery racks 305, PCS 315, and HVAC).

[0043] When multiple BESS containers 120 and their power containers 310 are used at any site to provide power at an independent site or a microgrid, the BESS ECS controller 135 determines the total charging power or discharging power that should be provided to the deployment and distributes the charging power requirement or discharging power requirement to the connected multiple BESS unit controllers 110. For example, the BESS ECS controller 135 calculates the power references of the different PCS 315 units, taking into account the power capacity of each PCS 315 and the power and energy capacity of the BESS container 120. The power and energy capacity of the BESS container 120 is determined by the number of battery racks 305 in operation. When there are multiple BESS containers 120 deployed at an independent site or at a microgrid site, and the multiple BESS containers 120 are connected to a single PCS 315, but the deployment has more than the capacity of a single PCS 315, then the BESS ECS controller 135 calculates the power references that can be used for all connected PCS 315. The BESS ECS controller 135 communicates the status of the plurality of BESS containers 120 to the next level in the BESS control hierarchy, namely the MECS controller 140. The BESS ECS controller 135 also communicates the available power reference of the PCS 315 to the BESS cell controllers 135 associated and connected to the BESS containers 120.

[0044] In an installation with multiple BESS containers 120, the balance state of the battery rack 305 of one BESS container 120 with the battery rack of another BESS container is not taken into account by the BMS 308 or the battery management unit 360 of the BESS container 120. In such a multiple BESS container 120 installation, the BESS ECS controller 135 manages each BAU 360 of the BESS container 120 through its corresponding BESS unit controller 110 through the network switch 200. The ECS controller 135 provides the state of charge (SOC) balance to the BMS 308 of each battery rack in the battery rack 305 of the multiple connected BESS containers 120 through the BAU 360.

[0045] In addition, the BESS ECS controller 135 is arranged to take into account the microgrid loads that the BESS container 120 needs to power, and manage the available power containers 310 to draw sufficient power from the battery rack 305 of each BESS container 120 so as not to over discharge any battery rack 305. When multiple BESS containers 120 and PCS 315 are used at any deployment site, the BESS ECS controller 135 is also arranged to determine the total charging power or discharging power of the multiple BESS containers 120. The BESS ECS controller 135 controls the allocation of charging power to multiple BAU units 360 and their associated battery racks 305 to multiple BESS power containers 310.

[0046] Figure 3 Components of an exemplary ECS 130 are shown. The ECS 130 includes a BESS ECS controller 135 and a MECS controller 140. The BESS ECS controller 135 and the MECS controller 140 are logically separate, however, they may be located on and executed within a common physical hardware / software controller, or communicatively coupled to different physical hardware / software controllers. The BESSECS controller 135 includes at least one processor 401, at least one memory device 410, at least one ECS server interface 415, and at least one MODBUS TCP interface 420. The processor 401 executes instructions that may be loaded into the memory 410. The processor 401 may include any suitable number and type of processing or other devices in any suitable arrangement. Exemplary types of processing devices include microprocessors, microcontrollers, digital signal processors, field programmable gate arrays, application specific integrated circuits, and discrete circuits.

[0047] Memory 410 represents any structure capable of storing information (such as data, program code, and / or other suitable information, whether temporary or permanent) and facilitating retrieval of the information. Memory 410 may represent random access memory or any other suitable volatile or non-volatile storage device. Memory may also include one or more components or devices that support longer-term storage of data, such as read-only memory, a hard disk, flash memory, or an optical disk.

[0048] The processor 401 executes various programs stored in the memory 410 that operate the BESS ECS controller 135 to provide a reference for power balance between the BESS container 120 and the PCS 315 attached to the BESS container 120. The programs also operate to distribute the power balance reference to the BESS unit controller 110. When calculating the power reference of the BESS container 120, the BESS ECS controller 135 also inputs the number of battery racks 305 that have been submitted within the BESS battery 304 of the BESS container 120. The BESS ECS controller 135 executes a program that calculates the energy balance taking into account the aggregate SOC and cycle counts of different BESS containers 120 that have the same or different numbers of battery racks 305 that are available for discharge or cut off and unavailable for use.

[0049] The ECS server interface 415 provides a communication portal to the network 108 to the VPP 160 through the network switch 400 and the firewall 220 using the DPN3 or MQTT protocol. This communication portal from the server interface 415 acts as a BESS container 120 connection to the VPP 160 and the level 4 of the BESS control system 100. Figure 4 As shown, MECS controller 140 also includes a direct communication connection to ECS server interface 415 via bidirectional line 401, which enables MECS controller 140 to directly access VPP 160. MECS controller 140 may be located and contained within ECS 130, however, it is logically separate from BESSECS controller 135 as described above.

[0050] The MODBUS TCP interface provides a Modbus TCP / IP communication portal that provides Ethernet intranet communications between the BESS ECS controller 135 and the BESS cell controllers 110 in single or multiple BESS container 120 deployments using the network switch 200 .

[0051] The ECS 130 functions as a DNP3 substation that interoperates with a DNP3 master station on a server 162 running on the VPP 160. The BESS ECS controller 135 allows selection of four different operating modes to provide a power reference for the connected BESS unit controller 110. The operating modes include: (1) a configurable power reference configured for test mode (this mode is primarily used for testing and certification); (2) a local human machine interface (HMI) configuration of a schedule for charging or discharging according to the power reference configured by the BESS ECS controller 135. The HMI (not shown) runs at an operator station 164 or a handheld mobile device; (3) a local HMI configuration of the use cases that the BESS battery 304 needs to support and a configurable schedule for the use cases configured by the HMI; and (4) a charge / discharge schedule downloaded from the cloud 168 from a remote operations center. Operation Mode 3 enables the BESS control system 100 to operate within a microgrid, where a function block running an algorithm for each of the use cases generates a power reference for the PCS 315 to manage charging or discharging of the BESS battery 304 .

[0052] The VPP 160 performs centralized coordination of distributed microgrids. The task of the VPP 160 is to calculate reference power for supplying power to a microgrid or drawing power from a main grid to which multiple microgrids are connected. The MECS controller 140 may receive an operation schedule from the VPP 160. For example, a schedule for power exchange between a microgrid and the grid, or a schedule for grid electricity prices associated with power input or output in the event that there is a difference in price set points between the input and output of power, or simply a schedule for use cases to which one or more BESS containers 120 together with generation assets and loads will be submitted. For example, renewable smoothing of output between 09:00 and 12:00, frequency regulation support between 12:00 and 17:00, and grid peak demand support between 17:00 and 20:00.

[0053] Using information from VPP 160, MECS controller 140 calculates a schedule for charging or discharging one or more BESS containers 120. MECS controller 140 calculates the schedule for charging or discharging, taking into account the schedule provided by VPP 160, and also taking into account local generation load, frequency, and voltage within the microgrid.

[0054] When calculating the schedule, the MECS controller 140 uses the prospective operation to predict local generation and demand / load. When local generation is controllable, such as when using diesel generators, the forecast of such local generation is the same as the committed schedule of the controllable generator. When the generation is uncontrollable and is variable and intermittent in nature, such as when using renewable generation sources, the forecast is calculated using historical time series data and inputs from meteorological sites and other weather forecast sources provided by the VPP 160. If there is a deviation, the deviation needs to be explained and the schedule is modified. For example, in the prospective operation using the forecast, short-term accurate forecasts are necessary because it takes a finite amount of time to charge or discharge the battery and depends on the level rate at which the battery provides energy or its C-rating. For example, it takes about 1 hour to fully charge a 1C rated battery system, while it takes 4 hours to fully charge or discharge a 0.25C rated battery system. Therefore, for a 0.25C rated battery system, in order to meet the grid peak support use case at any given time, battery charging should start at least four hours before the time when the PCS 315 is intended to discharge the battery. Even for a 1C rated system, charging with a high power is feasible, but charging with a higher power may attract a higher demand charge, or alternatively may result in a greater rise in battery temperature, which may cause the battery to degrade more quickly, so charging with a lower power may be necessary. A lower charging power will mean that the duration of charging is longer, and at least that duration needs to be predicted to ensure that the battery is charged when it is needed to release the stored energy.

[0055] Another task of VPP 160 is to balance the power supply and demand in multiple microgrids using economic optimization objectives, peak demand forecasts, and renewable energy generation forecasts. For example, VPP 160 can receive information from energy trading data from cloud 168, which generates market bids and market clearing prices from energy market operators. The information used on demand can be sent as an information response signal to MECS controller 140 to reduce the demand for diesel generation sources relative to other generation sources, for example due to the higher cost of diesel fuel.

[0056] Figures 4 to 5A system 500 for detecting degradation of a battery energy storage system of the present disclosure is shown in FIG. The system and method detect battery degradation of a BESS battery 304 and any economic costs associated with BESS battery degradation that may affect future cycles of the battery. The BECS includes a BESS digital twin (DT) 510 installed on an energy control system of a SCADA server 162 along with a VPP 160. For the purposes of the present disclosure, a digital twin is defined as a virtual model that uses real-time data to simulate the behavior of an asset or system and its operation, including monitoring the performance of an asset or system to identify potential failures and make more informed decisions about operational and life cycle performance. The energy control system SCADA server 162 utilizes the physical state of the battery racks 305a, 305b to 305n contained in the battery 304 to schedule future charge / discharge cycles of the BESS battery 304. For each battery rack installed in a BESS container 120, battery operation data is collected from the BESS container 120 installed on site. The collected battery operation data includes, for example, current, voltage, temperature, SOC of the battery modules 306 of the battery racks 305a, 305b, 305n. The battery operation data is coupled to the degradation estimation model in the learning agent 515. The degradation estimation model is continuously improved using any new data collected from the BESS containers 120 in the field. When the battery is not fully cycled, accurate estimation of SOC needs to take into account the reduced energy capacity of the battery rack between SOC calibration test cycles. The learning agent 515 trains a recurrent neural network using a physics-based model that is parameterized using a data-driven approach of past time series of voltage, current, and temperature. Estimation of battery degradation reduces the need to run or even avoid calibration tests altogether, which in turn reduces the downtime of the BESS for such tests, thereby increasing the availability of the BESS to the microgrid. Prognosis of battery degradation and improved battery charge / discharge schedules improve cost and energy efficiency. Early identification of degradation, even before typical symptoms of degradation are visible, can be used to track the warranty of BESS storage batteries.

[0057] The BESS unit controller 110 receives battery module 306 and battery rack 305a, 305b, 305n level data sent from the battery rack BMS 518a, 518b, 518n associated with the battery racks 305a, 305b, 305n, respectively, from the multi-rack battery management system 520. The read data includes the voltage, current, temperature, SOC and SOH of all battery cells contained in each battery rack module 306. When the voltage, current and temperature are measured directly from sensors attached to the battery cells, the SOC and SOH are inferred by the battery management system 520. If the battery has not undergone a calibration test or a full cycle of battery charging and discharging, the BMS520's estimate of SOC and SOH may be inaccurate. The battery prognostic model based on hybrid physics and neural networks learned from battery operation data solves the problem of accurately estimating battery SOC and SOH using data obtained from partial battery cycles between calibration tests and full cycles. The energy control system SCADA server 162 may also be programmed to avoid complete battery cycling, thereby avoiding stretching the battery to its limits in order to increase its lifespan.

[0058] The SOC estimation agent 525 running in the BESS unit controller 110 uses the DT 526 within a containerized software application executed by the processor 301 of the BESS unit controller 110. The containerized software 530 includes all binaries (BINs) 532 and libraries (LIBS) 534 required to run the SOC estimation agent 525 application in the container runtime. The processor 301 also runs the controller 110 host operating system 535. The host operating system 535 in the controller 110 can run multiple container engines. Battery operating data from the battery racks 305a, 305b, 305n and battery modules 306 is sent from each rack BMS 518a, 518b, 518n to the multi-rack battery management system 520. It should be noted that the BESS battery 304 can also be deployed using a compact, modular and scalable cubic battery cell configuration organized as strings and modules of lithium iron phosphate (LFP) battery cells that are connected to the same Figure 4 The rack BMSs 518a, 518b, 518n shown are similar to cubic BMSs on the cell, module, and string levels for monitoring.

[0059] Battery operation data such as, for example, current, voltage, temperature, SOC, and SOH are collected periodically during a time interval defined by the user, for example, every 15 minutes, the rate of change of the parameter is calculated, and if the rate of change of the parameter exceeds a configurable threshold, the collected data is sent to a battery data repository 915 hosted by the BESS unit controller and ECS 130 on the energy control system SCADA server 162. The battery data repository may also be hosted in the cloud 168. The battery data repository 915 aggregates battery data from multiple BESS containers. A remote or cloud-hosted battery data repository 915 is beneficial in enabling newer BESS devices to utilize data generated by older BESS container 120 devices, so that models can be trained using data generated by older BESS devices. This provides improved accuracy of SOC and charge estimation because more data is collected from the BESS container 120.

[0060] The battery data stored in the battery data repository 915 is used to train the digital twin associated with and coupled to the individual BESS container 120. For example, a DT parameter learning agent 515 including the digital twin 510 will be associated with each BESS container 120. The DT 510 is an aggregation of multiple physical models based on the Doyle-Fuller-Newman (DFN) multi-physics model of the battery cell. The parameter learning agent 515 learns from historical time series data using a recurrent neural network (RNN). The trained digital twin is downloaded to the estimation agent of the BESS unit controller 110. The DT 526 running in the BESS unit controller 110 generates a real-time estimate of the SOC that is more accurate than the SOC estimate generated by the BESS container BMS 308. This is because the BESS container BMS 308 does not have visibility of data generated by other BESS container 120 deployments, and the ability to learn from data generated by other BESS containers 120 deployed at multiple microgrid sites controlled by the VPP 160.

[0061] The real-time SOC estimation performed within the BESS unit controller 110 at the microgrid site is useful in generating a short-term schedule of battery charging and discharging cycles based on short-term forecasts of power demand, generation, and power costs from different sources at the microgrid site. However, long-term degradation and reduction of the state of health parameters is a much slower process. Therefore, the health estimation does not need to be updated as frequently as the state of charge estimation and can be run centrally on a cloud-hosted energy control system computer platform. The energy and power capacity estimated by the model is used by the SOC estimation agent when the battery has not undergone a full charge and discharge cycle or long after the calibration test is performed.

[0062] Figure 5 and Figure 6 An exemplary embodiment of a DT 526 for battery scheduling operations for charging / discharging a BESS battery 304 is shown. The DT 526 executed in the BESS unit controller 110 is an aggregation of multiple battery rack digital twins 550a-550n associated with each battery rack included in the BESS battery 304, such as Figure 3 305a, 305b, 305n shown in FIG. Multiple battery digital twins are also installed inside cubes in the battery cube cluster, which can be connected in parallel to form the BESS battery 304. Each battery rack digital twin 550a, 550n is an aggregation of multiple battery module digital twins arranged in series for each battery rack included in the BESS battery 304. Each battery rack digital twin 550a, 550n is identical in its functional structure and is referred to in detail for clarity. Figure 5 Only the system elements of the battery rack digital twin 550a are shown.

[0063] Each battery rack digital twin 550a is an aggregation of battery module digital twins 555a, 555n, which is a further aggregation of multiple battery cell digital twins 560a, 560b, 560n, such as Figure 6 As shown. The battery cell digital twins 560a, 560b, 560n are arranged in series and in parallel for each battery module 306. The battery cell digital twins 560a, 560b, 560n are the smallest units used in the hybrid physics and RNN battery model 565. The Doyle-Fuller-Newman multi-physics model is one of the options for the physics-based model of the battery cells. The data-driven RNN-based model within 565 is also used to estimate the degradation of the energy capacity and power capacity of the battery cells of the BESS battery 304. If the physical model parameter estimation fails, or if the parameter estimation is rejected by the human expert because the parameter estimation model needs to be retrained with input from the human expert user, RNN estimation is used. Expert user inspection and verification of the battery model parameters will be performed in Figure 7 explained in the discussion.

[0064] The cell digital twins 560a, 560b, 560n use the battery thermal model to generate estimates of the surface temperature of the battery cells. The battery cell surface temperature estimates are input to the cell parameter identification model 562. The battery thermal model includes a battery cell heat transfer model 563 as a lumped element model that generates estimates of heat generation and heat transfer in a single battery cell due to current flowing in either direction at different cell voltages or SOC levels. The battery thermal model also includes an RNN thermal model 566 trained using past temperature time series data.

[0065] The battery module ambient temperature is an input to the battery heat transfer model 563 of the battery thermal model used for each battery thermal model digital twin 560a, 560b, 560n. The module ambient temperature is derived from the ambient temperature estimation model 556, as Figure 5 As shown. The output of the ambient temperature estimation model 556 is coupled to the battery heat transfer model 563 and used to estimate the battery cell surface temperature. The battery cell current is estimated by the 1:N scaling module 558 by dividing the module current by the scaling factor 1:N, where N is the number of parallel cell groups arranged within the battery module. The estimated current output from the module 558 is coupled to the hybrid battery parameter identification model 562. The ambient temperature estimation model 556 also considers forced air cooling using a heating, ventilation and air conditioning (HVAC) unit and liquid cooling using a chiller that can be installed in the BESS container 120 to cool the BESS battery 304 battery cells. Models for the HVAC system 552 and the liquid cooling chiller system 553 are used as input to refine ambient temperature data from temperature sensor inputs installed near each battery module 306.

[0066] The energy and power capacity estimates generated by the RNN model due to training using historical data from multiple BESS container 120 devices from the hybrid physics and RNN battery model 565 provide more accurate capacity estimates than estimates from multi-physics models that are not correctly parameterized. This energy and power capacity estimate by the RNN model is used in the interim until the parameter estimates are improved with input from human expert users to reflect the physical condition of the BESS battery 304. The parameter identification model 562 uses a data-driven approach, including the use of a neural network to estimate the physical parameters of the BESS battery. Since each parameter is different, a different model is used for each parameter. Only parameters with good sensitivity and identifiability are estimated using this method. Parameters with low sensitivity are omitted to reduce computational requirements, and they are fixed in the physical model and do not change during the life of the battery. Lookup tables are used for parameters with low identifiability but good sensitivity to degradation caused by age and cycling.

[0067] The cost aggregation module calculates the total cost of degradation of the BESS battery 304 at different levels of the BESS system. The battery module 306 is considered the lowest replaceable unit of the BESS battery 304. The aggregated cost of degradation of the battery module 306 is not the sum of the degradation costs of the individual cells, but the battery module degradation cost is calculated by scaling the degradation cost of each battery cell in the battery module with the highest or most degradation within the battery module by a scaling factor equal to the number of battery cells within the battery module. The data-driven RNN based model 565 is used to estimate the degradation of the energy and power capacity available from each cell of the battery module 306 and the degradation scaling factor applied to the cell degradation cost aggregation module 567.

[0068] Since all battery modules are replaceable, the degradation cost of the battery rack 305a, 305b, 305n is the degradation cost of all battery modules 306 connected in series within the battery rack, and the cost is applied from the per-battery cell degradation cost aggregation module 567 to the battery rack degradation cost aggregation module 559. Similarly, the degradation cost of the BESS battery rack 305 in the BESS battery 304 in the BESS container 120 or cube cluster is the sum of the degradation costs of multiple racks within the BESS container 120 or cube cluster. The total degradation cost of the BESS container 120 from all battery racks 305a, 305b, 305n is coupled to the BESS rack degradation cost aggregation module 561. The BESS container degradation cost aggregation module 570 also receives the total degradation cost from other BESS containers 120 deployed in the field as input. The output of the BESS container degradation cost aggregation module 570 is coupled to the battery energy control system 513, where the battery energy control system 513 runs energy cost and carbon optimization calculations. The battery energy control system 510 considers the degradation cost in the schedule generation algorithm of the battery scheduler model 512 to minimize the total weighted cost of energy to include the battery degradation cost associated with the charging and discharging of the BESS battery 304. The battery energy control system 510 using the battery scheduler model 512 can also perform opportunistic full cycles for automatic calibration in an energy optimization schedule over a period of time, which can eliminate downtime for calibration testing of the BESS battery 304.

[0069] Figure 7 An embodiment 700 for updating BESS battery model parameters based on data generated during battery charge and discharge cycles is schematically illustrated, which uses an expert user to check and verify the battery model parameters. The physical BESS system 701 of the BESS container 120 includes battery modules 306 installed in battery racks 305a, 305b, 305n within the BESS container 120. For simplicity, the battery racks of the BESS container 120 are Figure 71 is shown as a single rack 710. The battery modules of the battery rack 710 are controlled by a hierarchical BMS 715, which in turn is under the control of the BESS unit controller 110 through the PCS 315 associated with the BESS container 120. The BESS unit controller 110 receives a power reference from the BESS ECS controller 130 and verifies the power reference relative to the available power and energy capacity of the BESS battery 304 installed at the microgrid site. The BESS ECS controller 130 couples the power reference to the PCS 315, which draws a discharge current from the BESS battery 304 or supplies a charge current to charge the BESS battery.

[0070] In response to the charge or discharge current, the battery terminals of the battery cells contained in the battery rack 710 produce changes in the voltage at the battery terminals, which represents the actual voltage response of the battery to the charge or discharge current. The current flowing through the battery cells of the battery rack 710 also generates heat that causes the temperature to rise, which is referred to as the actual rise in the temperature of the battery cells of the battery rack 710. The BMS 715 calculates the estimated SOC, SOH, cycle count, battery energy capacity, and battery power capacity of the battery rack 710 using an internal model located in the BMS 715. The estimate from the BMS 715 is accurate only immediately after the SOC calibration, which is accompanied by downtime of the BESS system.

[0071] The battery digital twin 704 can be used to increase the time period between calibration tests. This reduces the downtime of the BESS container 120 and increases the availability of the BESS for its intended use case. This goal is achieved by using the digital twin 704 simultaneously with the physical system 701 to generate more accurate estimates of SOC, SOH, battery energy capacity, battery power capacity, and cycle count. The digital twin 704 takes into account all field data generated from BESS operation, as well as field data from multiple BESS container deployments downloaded from the VPP controller 160 to the ECS controller 130. The digital twin 704 produces a simulated voltage response and simulated temperature rise that is as close as possible to the actual voltage response and actual temperature rise derived by the physical system 701.

[0072] The mean square error model 720 receives the simulated voltage response and the actual voltage response, as well as the simulated temperature rise and the actual temperature rise, and calculates the mean square error between the simulated quantity and the actual quantity. The output of the mean square error is input to the hybrid battery parameter identification model 730. The parameter identification model 730 uses a data-driven method including a neural network to generate an estimate of the parameter, or more specifically to generate changes in the battery parameters based on the actual charge and discharge current and ambient temperature input to the parameter identification model. The parameter identification model 730 for each parameter is different and is derived by considering the identifiability and sensitivity of the parameter to changes in the battery state estimation. The battery model parameters are listed in Table 1 below.

[0073] The generated parameters are uploaded to the human expert user 750 for analysis. The human expert user 750 analyzes and confirms the changes in the battery parameters before the battery parameters are downloaded to the hybrid physical and RNN battery model 760. The battery model 760 is supplemented by a fully data-driven RNN model that is trained with past time series battery operation data and generates as outputs estimates of simulated voltage response, simulated temperature rise, SOC, SOH, and cycle counts during partial cycles of the BESS battery 304. When there is evidence that the model parameters are inaccurate, or when the parameter identification model needs to be improved, the human expert user 750 may decide to select the actual estimate of the battery state from the physical system 701 instead of the estimate of the battery state derived by the battery model 760 of the digital twin 704. Templates for parameter identification models are provided to the expert user 750 to experiment with the model for parameter identification. Such templates will include options for formulating algebraic equations, differential algebraic equations, ordinary differential equations, and expression trees for formulating parameters.

[0074] Since each parameter is different, a different model is used for each parameter. Only parameters with good sensitivity and identifiability are estimated using this method. Parameters with low sensitivity are left out to reduce computational needs, and they are fixed in the physical model and do not change during the life of the battery.

[0075] The model parameters of the parameter identification model 730 with high sensitivity and identifiability are listed in Table 1 below.

[0076]

[0077] Table 1

[0078]

[0079] Table 1 (continued)

[0080]

[0081] Table 1 (continued)

[0082]

[0083] Table 1 (continued)

[0084]

[0085] Table 1 (continued)

[0086] The cell level anomaly detection model 770 detects any abnormal values ​​in the cell current, voltage, or temperature. Higher than normal cell currents can indicate physical defects that may lead to short circuits, which can be catastrophic if not detected. Deviations in cell voltage (particularly when charging or discharging) can indicate higher cell impedance due to degradation. If this degradation remains undiagnosed, it can lead to higher temperatures, which can further exacerbate degradation. Higher cell temperatures, especially cell temperatures close to the SOC limit, can indicate reduced energy capacity, which will require calibration of the SOC. If not detected, this deviation can accelerate degradation, which is preventable if detected early. The cell level anomaly detection model identifies outliers in the cell data and issues an alert for user intervention to update model parameters associated with the outlier cells. This provides an effective means for identifying voltage, current, and temperature data at the cell level.

[0087] Figure 8 and Fig. 9 An embodiment of the present disclosure of a warranty tracking system 800 for protecting the integrity of BESS battery data for warranty tracking of BESS batteries 304 is shown. The BESS warranty tracking system 800 uses battery data such as current, voltage, and temperature, as well as reports that a BESS system battery, such as the BESS battery 304, has degraded faster than expected in any given time period. The warranty tracking system 800 uses actual BESS battery data stored in a plain text file 810 in the BESS unit controller 110. The battery data contained in the plain text file 810 is encrypted by an encryption agent 815 using a private key generated by a private key generator 820. As shown in FIG. Figure 7As described in the illustrated embodiment, a private key is generated by the digital twin 704 using a hybrid physics and RNN battery model 760 using simulated battery data derived from actual battery data. The private key generator 820 receives the simulated battery data from the digital twin 704. The battery data includes energy capacity, power capacity, cycle count, a unique identifier of the physical battery, a maximum value of current (MAX 1), a maximum value of voltage (Max V), and a maximum value of temperature (MAX T), and a minimum value of voltage (MIN V). A plain text file containing battery data from file 810 is encrypted using a private key generated by the private key generator 820, and the encrypted battery data file is uploaded to a data storage platform 910 in the cloud and stored in a data repository 915.

[0088] The digital twin 704 generates data related to battery cycling, energy capacity, and power capacity that is closely correlated with historical battery data collected and received from the VPP 160 by the battery data repository 915 for warranty tracking purposes. The private key generated by the private key generator 820 is stored within the BESS unit controller 110 of the BESS container 120.

[0089] The data storage platform 910 includes a key generator 920 that can generate both public encryption keys and private encryption keys. When a user 950 at a remote operation center 955 requests battery data, an encrypted battery file from the data repository 915 is downloaded to the user 950 at the remote operation center 955. The public encryption key generated by the key generator 920 at the data storage platform is also sent with the encrypted file for use by the decryption agent 960. Using the public key, the decryption agent 960 decrypts the encrypted data file into a plain text battery data file.

[0090] Once decrypted back to plain text, the battery data cannot be encrypted back to the downloaded encrypted form without the originally generated private encryption key. Since the private key remains in the BESS unit controller 110, the private key is not available to the user 950 at the remote operation center 955. Any attempt to generate a private key will require the battery data to be available and passed as input to the battery model that generated that data along with the unique identifier. In order for the identifier to match, any attempt to generate the same private key requires the same battery data to be provided as input to the battery model 760 at the BESS system. Tampered data will result in a mismatch and the same private key will not be generated for use by the encryption agent 815.

[0091] The battery model may be updated by a human expert user 750 using battery parameters verified by the expert, such as Figure 7The parameters are updated to reflect changes due to the age and use of the physical battery. The decryption agent 930 uses the public key generated by the key generator 920 using data from the existing battery model to decrypt the encrypted data file that needs to be updated and stored in the data repository 915. The decrypted or plain text data is coupled to the battery digital twin learning agent 515.

[0092] The digital twin learning agent 515 also receives battery parameters verified by a human expert user 750. The updated battery model is also coupled as an input to the hybrid physics and RNN battery model 760 to generate a new private key for the updated battery model. The simulated battery model data from the battery model 760 is coupled to a key generator 920, which generates a new private encryption key that is input to an encryption agent 945, which encrypts the updated battery data for storage in a data repository 915. The cycle is repeated each time the battery model is updated. The method ensures that the private and public key pairs used to encrypt the battery data remain periodically changed. In addition to being able to detect data tampering, the method also ensures that newer data is not passed as older data for warranty replacement purposes.

[0093] It may be advantageous to set forth the definitions of certain words and phrases used throughout this patent document. The term "communication" and its derivatives cover both direct communication and indirect communication. The terms "include" and "comprising" and their derivatives mean including but not limited to this. The term "or" is inclusive, meaning and / or. The phrase "associated with..." and its derivatives may mean including, included within, interconnected with, included within, connected to, or connected with, coupled to, or coupled with, can communicate with, collaborate with, interlace, juxtapose, approach, be coupled to, or be combined with, have, have the property of, have a relationship with, or have a relationship with. When used with a list of items, the phrase "at least one of..." means that different combinations of one or more items in the listed items may be used, and only one item in the list may be needed. For example, "at least one of A, B, and C" includes any combination of A, B, C, A and B, A and C, B and C, and A and B and C.

[0094] The description in this application should not be construed as implying that any particular element, step, or function is an essential or critical element that must be included within the scope of the claims. The scope of subject matter protected by a patent is limited only by the claims as allowed. In addition, none of the claims is intended to invoke 35 U.S.C. §112(f) with respect to any of the appended claims or claim elements, unless the exact words "means for..." or "step for..." followed by a participle phrase identifying the function are expressly used in a particular claim. The use of terms such as (but not limited to) "mechanism", "module", "device", "unit", "component", "element", "member", "device", "machine", "system" or "controller" within the claims is understood and intended to refer to structures known to those skilled in the relevant art, as further modified or enhanced by the features of the claims themselves, and is not intended to invoke 35 U.S.C. §112(f).

[0095] Although the present disclosure has described certain embodiments and generally associated methods, changes and permutations of these embodiments and methods will be apparent to those skilled in the art. Therefore, the above description of the exemplary embodiments does not limit or restrict the present disclosure. Other changes, substitutions and variations are also possible without departing from the spirit and scope of the present disclosure as defined in the following claims.

Claims

1. A system for detecting degradation of a storage battery of at least one battery energy storage system (BESS) container, the system comprising: a battery management system (BMS) coupled to the storage battery, the battery management system (BMS) being configured to collect battery operation data from the storage battery; a battery data repository coupled to the BMS, the battery data repository for receiving and storing the battery operation data; and a learning agent coupled to the battery data repository, the learning agent using the stored battery operating data to estimate prognostic degradation of the storage battery and train a degradation estimation model, Therein, a state of charge (SOC) estimation agent receives and uses the trained degradation estimation model to generate a real-time estimate of the SOC of the storage battery for scheduling charge / discharge cycles for the BESS container.

2. The system according to claim 1, wherein: The storage battery of the BESS container includes a plurality of battery cells organized into a plurality of battery modules, and the battery modules are further organized into a plurality of battery racks, each of the plurality of battery racks includes a battery rack BMS, and the battery rack BMS collects operating data of the plurality of battery cells contained in the battery rack, wherein the operating data includes voltage, current, and temperature of all battery cells contained in the battery rack.

3. The system according to claim 2, wherein: Each battery rack BMS is coupled to a multi-level BMS configured to receive the operation data from each of the battery racks of the BESS container and estimate the SOC and state of health (SOH) of each battery rack.

4. The system of claim 1, wherein the learning agent is associated with the at least one BESS container and comprises: A digital twin comprising a multi-physics model of a battery organization of the storage batteries simulating the BESS container, wherein the digital twin is trained using time series operational data from at least one or more BESS containers stored in the battery data repository, and wherein the digital twin uses a recurrent neural network (RNN) to generate the degradation estimation model.

5. The system of claim 2, wherein the at least one BESS container comprises: a BESS unit controller executing the SOC estimation agent, the SOC estimation agent comprising: a first level aggregation of a digital twin for each battery module, the battery module digital twin receiving thermal parameters from a battery thermal model associated with each battery cell included in each battery module, the battery thermal model generating an estimate of a surface temperature of the battery cell; a second level aggregation of the digital twin having a digital twin for each battery rack, the digital twin configured to provide an ambient temperature and an estimated battery current scaling factor for the battery cells contained in each battery module; and The thermal model uses an RNN battery model to estimate the degradation of the energy and power capacity of each individual battery cell.

6. The system according to claim 5, wherein: Each battery module digital twin includes: a battery cell degradation cost aggregation module that receives an estimate of the degradation of the power and energy capacity available from each RNN battery model and outputs an aggregated degradation cost of the battery cells contained in each battery module, and each BESS battery rack digital twin includes: a battery module degradation cost aggregation module that receives the aggregated degradation of each battery cell included in the battery module and outputs an aggregated degradation cost of each battery module included in the battery rack, and Each BESS container includes: a container degradation cost aggregation module that receives aggregated degradation of the battery racks for a BESS container and outputs aggregated degradation costs to a battery energy control system, wherein the battery energy control system considers the battery degradation costs from the container degradation cost aggregation module to minimize costs associated with charging / discharging of the BESS storage batteries.

7. The system according to claim 2, wherein: The BMS collects battery operation data including actual voltage response and actual temperature rise from the storage battery, and the system further includes: a battery digital twin configured to receive operational data including charge / discharge current and ambient temperature from the BESS container to generate a simulated voltage response and a simulated temperature rise; a mean square error model receiving the actual voltage response and the actual temperature rise from the BMS and the simulated voltage response and the simulated temperature rise from a battery digital twin, the mean square error model being configured to output a mean square error output; a hybrid parameter identification model that receives the mean square error output to generate an estimate of a change in a battery parameter based on the charge / discharge current and ambient temperature input to the hybrid parameter identification model, Wherein, before downloading the changes in the battery parameters to the RNN battery model associated with the battery digital twin, the changes in the battery parameters are uploaded to an expert user, and the expert user confirms and verifies the changes in the battery parameters. The RNN battery model is trained using past time series operation data and is configured to generate estimates of simulated voltage response, simulated temperature rise, SOC, SOH, and cycle counts during a partial cycle of the storage battery as output.

8. The system according to claim 7, wherein: The battery operation data is stored in the BESS container as a plain text file, and the system further comprises: a private key generator coupled to the RNN battery model, the RNN battery model being configured to send simulated battery data and a unique identifier to the private key generator, the private key generator generating a private encryption key coupled to an encryption agent, wherein the plain text file is encrypted by the encryption agent into an encrypted text file using the private encryption key; a data storage platform located remotely from the BESS container, having a data repository coupled to the encryption agent and arranged to receive and store the encrypted text file; a key generator located on the data storage platform and coupled to the RNN battery model, the key generator being configured to generate a public encryption key using the simulated battery data from the RNN battery model; and A decryption agent is coupled to the data repository and the key generator, and receives the encrypted text file and the public encryption key upon user request, and decrypts the encrypted text file into a plain text file.

9. The system of claim 8, wherein using the learning agent, the encrypted text file stored in the data repository is configured to be updated with the changed battery parameters from the expert user to the RNN battery model to generate a new private key for the updated battery model, wherein the simulated battery data from the battery model is coupled to the key generator, the key generator generates a new private encryption key, the new private encryption key is input to an encryption agent located on the data storage platform, and the encryption agent uses the new private encryption key to encrypt the changed battery parameters into an encrypted text file for storage in the data repository.

10. A method for detecting degradation of a storage battery of at least one battery energy storage system (BESS) container, the method comprising: collecting battery operation data from a battery management system (BMS) coupled to the storage battery; storing the battery operating data in a data repository; using the stored battery operation data to train a degradation estimation model to estimate prognostic degradation of the storage battery; as well as A real-time estimate of a state of charge (SOC) of the storage battery is generated using the trained degradation estimation model for use in scheduling charge / discharge cycles of the storage battery.

Images