Power supply to electric vehicles

Abstract

A power system utilizing a hybrid architecture that allows low cycle life, high energy density chemistries to be used in the rechargeable battery to extend the range of the traction battery.

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims the benefit of priority of U.S. Application No. 63 / 089,990, filed Oct. 9, 2020, and U.S. Application No. 63 / 161,822, filed Mar. 16, 2021, each of which is incorporated herein by reference in its entirety.

[0002] This disclosure generally relates to systems, methods, and computer programs for powering an electric vehicle, and more particularly, to systems, methods, and computer programs for operating a power system of an electric vehicle via a high - energy - density battery configured to extend the driving range of a traction battery.

[0003] This disclosure also relates to methods, systems, and computer program products for intelligently determining the level of output power obtained from individual batteries of a hybrid architecture in order to achieve a driving range or distance goal while maintaining or maximizing the advantages provided by a hybrid architecture, including ensuring safety in an electric vehicle, maximizing battery life, and maximizing battery capacity.

Background Art

[0004] Power systems used in electric vehicles typically use a single battery pack or multiple battery packs connected in series. These batteries are usually rechargeable batteries, typically lithium - ion batteries.

[0005] Lithium - ion batteries have been widely used in electric vehicles and energy storage as green energy without environmental pollution due to their high output voltage, good cycle performance, low self - discharge rate, fast charging and discharging, and high charging efficiency.

[0006] Traditional battery parameter updates rely on a Battery Management System (BMS). The main functions of the BMS include monitoring battery voltage, current, and temperature among data points, estimating the battery's State of Charge (SOC), State of Health (SOH), State of Energy (SOE), State of Power (SOP), and Remaining Mileage (RM), performing diagnostics, protecting battery health, and executing battery balance management and battery temperature management processes.

[0007] To more accurately measure battery parameters, conventional technical solutions often involve pre-memorizing the OCV (Open Circuit Voltage)-SOC curve to check the estimated battery SOC. Some data may be uploaded to a cloud backup by the BWS so that the manufacturer or after-sales can obtain data analysis failure and battery history information.

[0008] Maintaining a precise balance of State of Charge (SOC) and balancing battery characteristics between battery cells and battery packs / modules is typically difficult. Old and new batteries, batteries of different capacities, or battery packs with different characteristics cannot be used together, and a failure in one battery core or pack can cause a failure of the entire battery system. These issues reduce efficiency and range, and significantly increase the cost of manufacturing and screening battery systems.

[0009] Another common problem in battery technology development involves a trade-off between energy density, the number of usable battery cycles during its lifespan, and battery performance. No known technology provides a battery solution or energy storage solution that offers a desirable energy density, high performance, and a large number of charge and discharge cycles during the battery's lifespan. [Overview of the Initiative] [Means for solving the problem]

[0010] To facilitate the identification of any particular element or action, the most significant digit of the reference number refers to the figure number in which the element is first introduced. Certain novel features that appear to be characteristic of the power supply system are outlined in the appended claims. However, the power supply system itself, as well as preferred modes of use, further non-limiting purposes, and its advantages, will be best understood by referring to the following detailed description of exemplary embodiments, when read in conjunction with the accompanying drawings. [Brief explanation of the drawing]

[0011] [Figure 1] This is a block diagram of a power supply system in which exemplary embodiments may be implemented. [Figure 2] This is a block diagram of a computer system in which exemplary embodiments may be implemented. [Figure 3] This is a sketch of an electric vehicle according to an exemplary embodiment. [Figure 4A] This is a chart illustrating an exemplary embodiment. [Figure 4B] This is another chart based on an exemplary embodiment. [Figure 5A] This is a sketch of a power supply system according to an exemplary embodiment. [Figure 5B] This is a chart of the discharge curve according to an exemplary embodiment. [Figure 6] This is another chart of a power supply system according to an exemplary embodiment. [Figure 7]This is another block diagram of a power supply system and vehicle chassis according to an exemplary embodiment. [Figure 8] This is a block diagram of a power supply system according to an exemplary embodiment. [Figure 9] This is a flowchart of an exemplary process for operating a power supply system in which exemplary embodiments may be implemented. [Figure 10] This is a block diagram of a network of a data processing system in which exemplary embodiments may be implemented. [Figure 11] This is a block diagram of a data processing system in which exemplary embodiments may be implemented. [Figure 12] This figure shows a configuration for an intelligent power output proposal according to an exemplary embodiment. [Figure 13] This is a block diagram of an exemplary configuration for training a machine learning model according to an exemplary embodiment. [Figure 14] This is a flowchart of an exemplary process according to an exemplary embodiment. [Figure 15] This is a block diagram of an exemplary attribute prioritization according to an exemplary embodiment. [Modes for carrying out the invention]

[0012] Exemplary embodiments recognize that currently available solutions do not fully address or provide suitable solutions to the problems discussed above. Electric vehicles typically rely on a single battery to power the vehicle. This limits the vehicle's range to only those chemistries that can meet cycle life, durability, and range requirements, which usually means the chemistries must be limited. Many chemistries may have higher energy densities (e.g., two to three times higher in energy density) than the conventional chemistries used for electric vehicle batteries but have insufficient cycle life. Considering the need to extend the range in electric vehicles, said chemistries can be utilized to significantly extend the range beyond the conventional capabilities when properly managed.

[0013] Exemplary embodiments recognize that most conventional cells in rechargeable batteries are connected in parallel, making it impossible to control the input and output currents passing through the cells. Exemplary embodiments also recognize that when an individual cell of said rechargeable battery fails, it is difficult to maintain the integrity and performance of the battery because the death of the cell is accelerated as the failure cannot be detected and / or mitigated in a timely manner. Further, in some configurations, when one cell fails, the entire battery may be rendered inoperable. Exemplary embodiments further recognize that conventional batteries do not utilize high-energy-density chemistries due to high cycle life requirements.

[0014] For the sake of clarity of explanation and without implying any limitation thereto, some exemplary configurations are used to describe the exemplary embodiments. From this disclosure, those skilled in the art can envision many changes, adaptations, and modifications to the described configurations to achieve the described purposes, and the same is intended to be within the scope of the exemplary embodiments.

[0015] Furthermore, in the figures and exemplary embodiments, simplified diagrams of the system are used. In an actual computing environment, additional structures or components not illustrated or described herein, or structures or components different from those illustrated but for the same functions as described herein, may exist without departing from the scope of the exemplary embodiments.

[0016] Furthermore, for the exemplary embodiments, by way of example only, specific actual or virtual components are described. The steps described by the various exemplary embodiments can be adapted to a power system for an electric vehicle using various components that can be targeted or reused to provide the described operations, and such adaptations are contemplated within the scope of the exemplary embodiments.

[0017] For the exemplary embodiments, by way of example only, specific types of steps, applications, processors, problems, and data processing environments are described. Any specific manifestations of these and other similar artifacts are not intended to limit the present invention. Any suitable manifestations of these and other similar artifacts can be selected within the scope of the exemplary embodiments.

[0018] The examples in this disclosure are used only for the purpose of clarifying the description and are not limitations on the exemplary embodiments. Any advantages recited herein are merely examples and are not intended to be limitations on the exemplary embodiments. Specific exemplary embodiments may achieve additional or different advantages. Furthermore, specific exemplary embodiments may have some, all, or none of the advantages recited above.

[0019] The exemplary embodiments described herein are directed to a power supply system 100 for an electric vehicle. Power supply system 100 (Figure 1) is configured to include low-cycle-life, high-energy-density chemicals in a hybrid architecture to enable the benefits of such chemicals, including a significant increase in range, while protecting the architecture from the disadvantages of such chemicals that have prevented them from being trusted in the automotive sector. The battery systems disclosed herein may be referred to as “hybrid” systems because they include multiple chemicals that act in parallel. Alternatively, to distinguish them from “hybrid” vehicles that use both electric and internal combustion power sources, the battery systems, vehicles, and related systems and components disclosed herein may be referred to as “range-extending, multi-chemical battery systems.”

[0020] The power supply system 100 disclosed herein may include a traction battery 102 (for example, containing lithium iron phosphate (LFP)) and a hybrid range extender battery 124 comprising one or more high-energy-density hybrid modules 112 having one or more hybrid chemicals and which can be controlled to provide power to charge the traction battery 102 and / or power the electric vehicle. One or more embodiments recognize that existing problems in rechargeable battery manufacturing require providing electric vehicles having batteries that have high energy density to increase the long-distance driving range available to electric vehicles beyond the conventional range, while taking into account the low cycle life introduced by said high energy density.

[0021] One or more embodiments include one or more processors 106 (or processor 120, computer processor 206, Figure 2), which are either onboard or located within an external computer system 126 (or computer system 200) or externally, in order to perform some of the steps described herein. A traction battery may monitor and determine its discharge and charge limits. An inverter may manage the flow of power, and a hybrid module controller may manage its own charge / discharge via a DC / DC converter. In one or more embodiments, the vehicle 302 (Figure 3) is configured as an electric vehicle (EV). In one or more embodiments, the vehicle 302 is configured as a plug-in hybrid electric vehicle (PHEV). The term electric vehicle is used below collectively for automatic vehicles, railway vehicles, ships, and aircraft that are configured to utilize a rechargeable electric battery as the primary energy source for supplying propulsion to the drive system, or that have an all-electric drivetrain.

[0022] Furthermore, as used herein, a sensor can be a system, apparatus, software, hardware, a set of executable instructions, an interface, a transducer, and / or various combinations thereof, comprising one or more sensors used to exhibit, respond to, detect, and / or measure physical properties and generate data relating to those physical properties.

[0023] Furthermore, battery energy density is generally used to refer to a measure of how much energy a cell contains in proportion to its volume.

[0024] Furthermore, as used herein, a high-energy-density module refers to a module having cells with a cell energy density of about 1000 Wh / L or more, for example, 1100 Wh / L or 1200 Wh / L. Those skilled in the art will recognize that conventional battery chemicals with automotive-level performance have a cell energy density measured at the cell level that is lower than or significantly lower than 1000 Wh / L, for example, between about 350 Wh / L and 500 Wh / L, as shown in Figure 4B. Using high-energy-density chemicals in the hybrid range extender battery 124 can ensure, for example, the provision of more than two or three times the energy provided by the traction battery 102.

[0025] In one or more embodiments, the power supply system 100 includes a traction battery 102 having one or more traction modules 122, a hybrid range extender battery 124 having one or more high energy density hybrid modules 112, and a partition between the traction battery 102 and the hybrid range extender battery 124.

[0026] Each module can be part of a battery stack. Those skilled in the art will understand that other types of battery devices can be used to supply power in the embodiments described herein, and therefore the enumeration of specific configurations is not intended to be limiting. As discussed herein with respect to Figure 1, the battery management system, BMS 104, may use, for example, an onboard computer system 126 to control relay 108 and report operating limits. It may also request power from one or more hybrid modules to meet the needs of the vehicle. The hybrid module controller of the hybrid range extender battery 124 can control its contribution to the high-voltage DC bus based on its own internal objectives (such as objectives defined by one or more preset or dynamically determined rules), energy state, and observed energy state and driving behavior of the traction battery, without centralized adjustment from the BMS. Thus, the power system 100 can operate in a more efficient and power-saving mode to increase the operating distance of the vehicle 302, or prevent module degradation caused by a single faulty cell. For example, while driving, one or more embodiments described herein include an onboard computer system 126 that estimates the power requirements for navigating to a destination and determines whether the vehicle 302 can safely reach the destination using the stored energy available to operate. If the computer system 126 determines that the vehicle cannot reach the given destination, the traction battery 102 may be charged using the hybrid range extender battery 124 to supply sufficient power for driving.

[0027] In one or more embodiments, the high energy density hybrid module 112 is configured to have a single chemical substance, while in one or more other embodiments, the high energy density hybrid module 112 is configured to have multiple chemical substances (for example, three chemical substances for daily, weekly, and monthly use).

[0028] In an exemplary embodiment, the traction battery 102 comprises a single traction module 122 or multiple traction modules 122 connected in series. In another exemplary embodiment, the hybrid range extender battery 124 has multiple high energy density hybrid modules 112 connected in parallel to each other and also in parallel with the traction battery 102, each of which is capable of managing the charging of the traction battery 102 or contributing to the power supply of the vehicle 302, and the hybrid module controller 118 of each high energy density hybrid module 112 includes a bidirectional DC-DC converter. More generally, the negative terminal is connected to the positive output of the bidirectional DC-DC converter.

[0029] In one or more embodiments, a battery that can be used in the hybrid range extender battery 124 described herein for supplying power to the vehicle 302 or for charging the traction battery 102 includes a battery having cells 114 having a cell energy density greater than 1000 Wh / L.

[0030] The battery system in electric vehicles is typically a traction battery, consisting of hundreds of cells packed together. These systems, for example, with voltage ratings of 300V to 400V, supply high currents of around 300A (e.g., 200-300A), and any mismanagement can trigger a serious disaster. Therefore, a battery management system is essential for the safe operation of high-voltage batteries in electric vehicles. A battery management system can be configured to monitor the battery's condition and prevent overcharging and discharging, which can reduce battery life and capacity, and even cause explosions. For example, a BMS can check the voltage and stop the charging process when the required voltage is reached. If an irregular pattern is detected in the power flow, the BMS can shut down the battery and issue an alarm. Furthermore, the BMS can be configured to relay information about the battery's condition to the energy and power management system. In addition, the BMS can also regulate the temperature of the battery cells and the overall health of the battery, ensuring the battery is safe and reliable under all conditions.

[0031] One feature of a BMS is its ability to estimate the state of charge (SOC) of a battery pack to the extent that it is desirable or, in some cases, important, in order to efficiently maintain the SOC of the battery pack to ensure that the battery voltage is not too high or too low. For example, a battery should not be charged beyond 100% or discharged to 0%, as this can sometimes reduce the capacity of the battery cells. In addition to providing accurate information on the battery voltage and temperature, a BMS can also provide indicators of the energy available for use and the remaining battery capacity.

[0032] In some embodiments, the State of Charge (SOC) can be estimated. Furthermore, in the Coulomb counting process, the currents entering and leaving the battery are integrated to generate a relative value of the charge. However, it can often be difficult for conventional systems to accurately determine the SOC and other characteristics of individual cells connected in parallel.

[0033] Therefore, the exemplary embodiment recognizes that conventional BMSs cannot accurately measure the individual characteristics of cells within a battery pack. Conventional solutions attempt to obtain estimates but lack a way to control the cell current to measure corresponding characteristic parameters such as cell voltage.

[0034] Returning to Figure 1, the traction battery 102 may include one or more traction modules 122 configured to supply power to the vehicle 302. The hybrid range extender battery 124 is designed to be modular, having one or more types of chemicals different from those of the traction battery 102, in order to provide the vehicle with its various power requirements when needed. As a specific example, the traction battery may have LFP chemicals, and the hybrid range extender battery 124 may have Gr (graphite) or Gr+SS (graphite + solid state) chemicals. Regardless of the specific chemicals used, the hybrid range extender battery 124 may be designed to have one or more high energy density hybrid modules 112 or packs, each having its own DC-DC converter to function as an independent battery. The high energy density hybrid modules 112 can be controlled independently, and the charge and discharge rates of the cells 114 can be adjusted by being able to independently measure the health or condition of its individual cells 114. In one embodiment, the cells 114 of the high energy density hybrid module 112 are arranged in series. By using a balancing device 128, such as a bleeder resistor, connected in parallel with each cell 114, the charging or discharging rate of the cells 114 can be controlled; that is, turning on a cell's bleeder resistor discharges the charge stored within the cell. In an exemplary embodiment, the bleeder resistor may be capable of creating an additional discharge current of up to several hundred (200) milliamperes, thereby allowing for fine-tuning of the cell's charge / discharge current and bringing the cells in the string to a common state. Furthermore, one or more sensors 116 are used to measure voltage and determine how long the bleeder resistor should remain activated to achieve equilibrium across all cells in a series string of cells.

[0035] The percentage of a battery's maximum capacity that it discharges is its C-rate. For example, a 1C rate means that the discharge current discharges the entire battery in one hour. Typically, a vehicle requires 4C peak and 1C average. By independently controlling the high-energy-density hybrid modules 112 using bidirectional DC-DC converters, rates of C / 5 (i.e., 0.2C) or less can be achieved. This prevents triggering failure events associated with high-energy-density chemicals due to overcharging and discharging. More specifically, the traction battery 102 can provide peak current according to the vehicle's load demands. The high-energy-density hybrid modules 112 can use their bidirectional DC-DC converters to discharge the traction battery and powertrain to the connected vehicle HV bus. In an exemplary embodiment having five high-energy-density hybrid modules 112, each contributing C / 5, their combined contribution is 1C. When the vehicle requires 4C, the traction is discharged at 3C. When the vehicle requires 1C, the traction battery 102 is stopped (0C). When the vehicle requires -1C (regenerative braking), the traction battery is recharged at 2C. In one embodiment, each high energy density hybrid module 112 also has an operablely coupled hybrid module controller 118 for measuring the health or condition of the cells 114. For example, the hybrid module controller 118 can be configured to measure voltage, current, temperature, SOC (state of charge), and SOH (state of health) for all cells of the corresponding high energy density hybrid module 112. It can also manage isolation and current, and has DC-DC converter control to adjust their contributions and absorb and supply energy to the main bus / high voltage DC bus of the power system 100. The system may also have a BMS 104 configured primarily to communicate with the traction battery 102.If the traction battery 102 fails, one or more of the high-energy-density hybrid modules 112 can act as a substitute (e.g., a temporary substitute) for the traction battery 102 by directly supplying power to the drive unit 110. One or more processors (processor 120, processor 106, or processors of computer system 126) are used in some configurations to enable the execution of one or more processes or operations described herein. Relay 108 is controlled to operably couple the vehicle's drive unit 110 to the power of the power system 100. The drive unit 110 may collectively refer to external devices of the power system 100, such as the propulsion motor, inverter, and HVAC (heating, ventilation, and air conditioning) system.

[0036] Having described the power supply system 100, we now refer to Figure 2, which shows a block diagram of a computer system 200 that may be used according to at least some of the exemplary embodiments described herein. While various embodiments of this exemplary computer system 200 may be described herein, those skilled in the art will likely see how the disclosure can be implemented using other computer systems and / or architectures after reading this description.

[0037] In exemplary embodiments herein, the computer system 200 may form part of the computer system 126 in Figure 1, or be independent of the computer system 126 in Figure 1. Furthermore, at least some components of the power supply system 100 may form the computer system 200 in Figure 2, or be included within the computer system 200 in Figure 2. The computer system 200 includes at least one computer processor 206. The processors 106 and 120 of the power supply system 100 may be part of the computer processor 206, form part of the computer processor 206, or be independent of the computer processor 206. The computer processor 206 may include, for example, a central processing unit (CPU), multiple processing units, an application-specific integrated circuit ("ASIC"), a field-programmable gate array ("FPGA"), and the like. The computer processor 206 may be connected to a communication infrastructure (e.g., a network) 202 (e.g., a communication bus, network). In exemplary embodiments herein, the computer processor 206 includes a CPU that controls the process of operating the power system 100, which includes controlling the state of the bidirectional DC-DC converter between the high energy density hybrid module 112 and the traction battery 102 or the drive unit 110 of the electric vehicle 302.

[0038] The display interface 208 (or other output interface) may transfer text, video graphics, and other data relating to the power system 100 from a communication infrastructure (e.g., a network) 202 or a frame buffer (not shown) for display on a display unit 214, which may be a display of the electric vehicle 302. For example, the display interface 208 may include a video card having a graphics processing unit, or it may provide an interface to an operator for controlling the power system 100.

[0039] The computer system 200 may also include an input unit 210, which can be used by an operator of the computer system 200 together with the display unit 214 to transmit information to the computer processor 206. The input unit 210 may include a keyboard and / or a touchscreen monitor. In one example, the display unit 214, the input unit 210, and the computer processor 206 may collectively form a user interface.

[0040] One or more computer implementation steps for operating the power supply system 100 may be stored on a non-temporary storage device in the form of computer-readable program instructions. To perform the procedure, the computer processor 206 loads the appropriate instructions stored on the storage device into memory and then executes the loaded instructions.

[0041] The computer system 200 may further include main memory 204 which may be random access memory ("RAM"), and may also include secondary memory 218. The secondary memory 218 may include, for example, a hard disk drive 220 and / or a removable storage drive 222 (e.g., a floppy disk drive, magnetic tape drive, optical disk drive, flash memory drive, etc.). The removable storage drive 222 reads from and / or writes to the removable storage unit 226 in a well-known manner. The removable storage unit 226 may be, for example, a floppy disk, magnetic tape, optical disk, flash memory drive, etc., which can be written to and read by the removable storage drive 222. The removable storage unit 226 may include a non-temporary computer-readable storage medium that stores computer executable software instructions and / or data.

[0042] In further exemplary embodiments, the secondary memory 218 may include other computer-readable media for storing computer executable programs or other instructions to be loaded into the computer system 200. Such devices may include a removable storage unit 228 and interface 224 (e.g., a program cartridge and cartridge interface), a removable memory chip (e.g., an erasable programmable read-only memory ("EPROM") or a programmable read-only memory ("PROM")) and associated memory sockets, as well as other removable storage units 228 and interfaces 224 that allow software and data to be transferred from the removable storage unit 228 to other parts of the computer system 200.

[0043] The computer system 200 may also include a communication interface 212 that enables software and data to be transferred between the computer system 200 and external devices. Such an interface may include a modem, a network interface (e.g., an Ethernet card or an IEEE 802.11 wireless LAN interface), a communication port (e.g., a USB or FireWire® port), a Personal Computer Memory Card International Association ("PCMCIA") interface, Bluetooth®, and the like. The software and data transferred via the communication interface 212 may be in the form of signals, which may be electronic signals, electromagnetic signals, optical signals, or other types of signals that can be transmitted and / or received by the communication interface 212. Signals may be provided to the communication interface 212 via a communication path 216 (e.g., a channel). The communication path 216 carries the signals and may be implemented using wires or cables, optical fibers, telephone lines, cellular links, radio frequency ("RF") links, and the like. The communication interface 212 may be used to transfer software, data, or other information between the computer system 200 and a remote server or cloud-based storage (not shown).

[0044] One or more computer programs or computer control logic may be stored in the main memory 204 and / or secondary memory 218. Computer programs may also be received via the communication interface 212. When executed by the computer processor 206, the computer programs include computer-executable instructions that cause the computer system 200 to perform the methods described below. Thus, the computer programs can control the computer system 200 and other components of the power supply system 100.

[0045] In another embodiment, the software may be stored in a non-temporary computer-readable storage medium using a removable storage drive 222, a hard disk drive 220, and / or a communication interface 212, and in main memory 204 and / or secondary memory 218. When the control logic (software) is executed by the computer processor 206, it causes the computer system 200, more generally the power supply system 100, to perform some or all of the methods described herein.

[0046] Finally, in another exemplary embodiment, hardware components such as ASICs and FPGAs may be used to perform the functions described herein. Implementations of such hardware arrangements for performing the functions described herein will be apparent to those skilled in the art in consideration of this description.

[0047] Figure 4A shows a chart according to an exemplary embodiment. The chart shows a percentage of driving days axis 402 and a daily driving distance axis 404 for an exemplary embodiment disclosed herein. By measuring the user's driving habits, it can be seen that a significant percentage of the days spent driving are spent driving relatively short distances and, therefore, utilizing the traction battery 102, as indicated by the traction battery portion 406 in the chart. The hybrid range extender portion 408, on the other hand, is used for a relatively much shorter time. In this case, as shown in Figure 4B, a range extender with a chemical that provides an energy density of 1000 Wh / L or more may offer a good trade-off between density and cycle life. The exemplary embodiment in Figure 4A can be achieved by determining the percentage of cycles that are out of routine use and selecting an appropriate chemical that can maintain many of those cycles.

[0048] The chart in Figure 4B includes an energy density axis 410 and a cycle life axis 412. As used herein, “cycle life” of a battery refers to the number of times a battery can be discharged to 100% depth of discharge (DoD) while retaining at least 80% of its original charge. For example, a battery with a cycle life of 100 cycles will retain 80% of its original charge after being charged and completely discharged 100 times.

[0049] Traction battery chemicals may be selected from the traction battery chemical region 414 to provide a cycle life of approximately 3000 cycles (e.g., at least 2500 or 3000 cycles). In conventional battery chemicals, this cycle life typically provides a corresponding cell energy density of approximately 400 Wh / L. To meet predetermined range requirements for non-traction applications, range battery chemicals may be selected from the exemplary hybrid range extender battery chemical region 418 (e.g., between 1000 and 1200 Wh / L). This typically provides a corresponding cycle life of approximately 2000 cycles or less (e.g., between 200 and 350 cycles). Depending on the vehicle's energy requirements, other chemicals 416 may be optionally available for intermediate range requirements, and the corresponding packs can be controlled independently.

[0050] More generally, embodiments disclosed herein may utilize multiple battery chemicals in a power system, each having a different expected cycle life and / or cell energy density. This may enable the use of battery chemicals and configurations that were previously considered unsuitable for electric vehicles and similar devices. For example, conventional systems have often been thought to require higher cycle life at the expense of higher energy density. In contrast, embodiments disclosed herein may utilize higher-density chemicals even when the associated battery may have a relatively low cycle life, as range extender or intermediate-range battery cells may not be subjected to charge / discharge cycles as frequently as traction batteries in normal use.

[0051] As a specific example, the hybrid power systems disclosed herein may include traction batteries having cell energy densities in the range of 300 to 500 Wh / L, not exceeding approximately 500 Wh / L, 450 Wh / L, 400 Wh / L, 350 Wh / L, 300 Wh / L, or less, but having relatively high cycle lives in the range of 2000 to 3200 cycles, or more, of 2000 to 3200 cycles.

[0052] The higher density battery cells used in the range extender batteries or intermediate batteries disclosed herein may have a relatively high cell energy density of 800 Wh / L, 1000 Wh / L, 1100 Wh / L, 1200 Wh / L, or more, or in the range of 800 to 1400 Wh / L, and a relatively low expected cycle life of 300, 400, or 500 cycles or less, or in the range of 100 to 500 cycles or less. In particular, in embodiments using three or more chemicals, other battery types and chemicals may be used. For example, any of the battery types shown between the traction region 414 and the range extender region 418 in Figure 4B may be used in an intermediate density battery having a cycle life in the range of 1000 to 2000 cycles and an energy density in the range of 500 to 800 Wh / L.

[0053] The numerical value of interest for the battery chemicals used in the embodiments disclosed herein is the energy density per cycle (EDC), which is determined as the ratio of the battery's cell energy density to its expected cycle life. For example, as shown in Figure 4B, an HE traction battery may have a cell energy density of approximately 400 Wh / L and a cycle life of 3000 cycles, resulting in an EDC of approximately 0.13 Wh / L / cycle. In contrast, the solid-state battery in the range extender region 418 of Figure 4B may have a cell energy density of approximately 1000 Wh / L and a cycle life of approximately 400 cycles, resulting in an EDC of approximately 2.5 Wh / L / cycle. Conventional battery chemicals with an EDC of 1.0 or higher were previously considered unsuitable for use in electric vehicles due to their relatively low cycle life. As previously disclosed, the embodiments provided herein enable such batteries to be used efficiently in electric vehicles when used in conjunction with other chemicals.

[0054] As a specific example, embodiments disclosed herein may use a traction battery having an EDC of about 0.12–0.16 Wh / L / cycle and a range extender battery having an EDC of 1.0 or greater, 2.0 or greater, 5.0 or greater, or any intermediate value. Other chemicals may be used similarly; for example, if three chemicals are used, the traction battery may have an EDC of 0.12–0.16 Wh / L / cycle, and the other batteries in the system may have an EDC between the EDC of the traction battery and the EDC of the highest density battery, with the highest density battery having an EDC of 1 Wh / L / cycle or greater.

[0055] More generally, any number of battery chemicals can be used in parallel, with “everyday” traction batteries having lower EDC values ​​and more specialized application battery chemicals having higher EDC values. As another example, a single battery chemical in the everyday traction region 414 can be used with any number of batteries in the range extender region 418 and / or any number of batteries in any intermediate range shown in Figure 4B. For example, a third battery chemical can be used with the previously disclosed traction battery and range extender battery, the third chemical having a cell energy density of 400 to 1200, 1300, or 1400 Wh / L or higher.

[0056] Figure 5A shows an exemplary embodiment of the power supply system 100. The system includes a traction battery 102, a plurality of high-energy-density hybrid modules 112 connected in parallel to the main traction bus / high-voltage DC bus, a plurality of traction modules 122, and a plurality of bidirectional DC-DC converters 502. In addition, the system has an onboard AC-DC charger 504 for recharging the power system from the grid, a 12V battery 512 for powering the vehicle's lights and ignition, and an auxiliary DC-DC converter 506 for maintaining the 12V battery 512 and powering the vehicle's 12V system. The embodiment also has a contactor 508 for switching various circuits on or off, and a control module 510 for controlling the power supply. By placing the 12V battery 512 inside the power supply system (inside the traction battery 102) rather than externally as in conventional systems, the contactor 508 can be controlled, for example, kept closed, even if other momentary problems exist in the 12V system. In an exemplary embodiment, a momentary loss of battery power (e.g., approximately 100 milliseconds or more) could cause a contactor to open. This loss could be caused by a single faulty wire outside the battery pack. This risk can be mitigated by drawing 12V into the pack.

[0057] In an embodiment as shown in Figure 5A, each high-energy-density hybrid module 112 has approximately 56 cells 114 connected in series. A particular number of cells is illustrative, and other numbers of cells may be used without departing from the scope of this disclosure. An operablely coupled hybrid module controller 118, such as an onboard hybrid module controller 118, is configured to measure the voltage, current, temperature, SOC, and SOH of each of the individual cells 114. Each of the 56 cells 114 may have an associated voltage sensor 116. By knowing the current and temperature passing through the cells 114 (such as the temperature at various points on the high-energy-density hybrid module 112), the SOH, SOC, and other parameters of the cells 114 can be calculated to determine whether the energy output of the corresponding high-energy-density hybrid module 112 can be connected to the traction battery 102, or optionally to the drive unit 110, via the corresponding bidirectional DC-DC converter 502. Furthermore, unlike conventional load-following power supplies that cannot control the changing drive power, a bidirectional DC-DC converter 502 for each high-energy-density hybrid module 112 can be used to precisely control the current input and output for each high-energy-density hybrid module 112. In an exemplary embodiment, charge pulses and discharge pulses are generated for the high-energy-density hybrid module 112. By controlling the amount of current for the series-connected cells 114 of the high-energy-density hybrid module 112 through the use of the bidirectional DC-DC converter 502 and measuring the voltage of each cell 114, the impedance of each cell 114 can be calculated and compared to reference data to identify any undesirable deviations in cell impedance and corresponding changes in cell health.

[0058] The current input to each high-energy-density hybrid module 112 may come from a charger after the traction battery has been charged or nearly charged. For maintenance and / or diagnostic purposes, hybrid modules may be discharged and / or recharged when not strictly required as a range extender. For example, if several months have passed since a hybrid module was used as a range extender, the hybrid module may be discharged and recharged during normal daily use to keep the cells working. The frequency of discharge and recharge of the range extender battery outside of normal use, or whether such discharge / charge is performed at all, may be selected based on one or more specific chemicals used in the range extender battery.

[0059] The hybrid module controller 118 can also manage the strain of cells 114 by monitoring and aligning them. For example, if one cell 114 (cell A) is determined to have a lower SOC (e.g., 20%) than another cell 114 (cell B) connected in series (70%), then cell B will reach full charge faster than cell A, and therefore, charging of cell B must be stopped to prevent overcharging. By using a bleeder resistor to reduce the SOC of cell B to the SOC of cell A, both cells A and B can be charged at the same rate to a predetermined full charge. Thus, the hybrid module controller 118 maintains the SOCs of the 56 cells 114 to be equal or substantially equal (e.g., + / -10%, or + / -5%, or + / -1%) so that the entire range of the module can be used. In another example, by determining which cell 114 has a lower self-discharge rate than the other cells 114, the hybrid module controller 118 then determines which cell 114 to selectively discharge to the determined charge in order to charge all 56 cells 114.

[0060] In another exemplary embodiment, the high-energy-density hybrid modules 112 are connected in parallel to one another and controlled independently, so that each high-energy-density hybrid module 112 can be individually removed for readjustment by slowly charging and discharging without affecting the normal operation of the power supply system 100.

[0061] Figure 5B shows an exemplary charge-discharge curve 500 of a cell, including a voltage axis 514, a capacitance axis 516, a discharge curve 518, and a charge curve 520. As shown in the discharge curve 518, at high discharge current / C rates 522 (e.g., 5C), the cell capacitance is not fully utilized, and the cell voltage drops due to internal resistance. The current flowing through the cell causes an IR voltage drop across the cell's internal resistance, which lowers the terminal voltage of the cell during discharge, increasing the voltage required to charge the cell, thus reducing the cell's effective capacity and also lowering the cell's charge / discharge efficiency. Higher discharge rates cause higher internal voltage drops, which explains the lower voltage discharge curve at high C rates 522 and the characteristically different shape of the curve. The ability to precisely control the current using the bidirectional DC-DC converter 502 allows for the inference and mitigation of impedance problems for any cell by comparing it to a reference profile, such as a previously stored profile of cell 114, by discharging and charging at various C rates. This can be achieved using a controlled step response to characterize the behavior of cell 114 over time. One mitigation action involves first discharging the high-energy-density hybrid module 112 that does not have an identified cell problem. Another mitigation action involves slowing down the discharge of the high-energy-density hybrid module 112 that has an identified cell impedance problem.

[0062] Figure 6 shows another exemplary configuration of the power supply system 100 disclosed herein, including an onboard energy management system 602. In this example, a traction battery 102 has a capacity of 44 kWh and provides a voltage of 320 V, and a hybrid range extender battery 124 has a capacity of 120 kWh through six 20 kWh high energy density hybrid modules 112, each having a voltage of 48 V. The onboard energy management system 602 has a battery management system (not shown) and is configured as a three-voltage system for handling 12 V, 48 V, and 320 V. Furthermore, the onboard energy management system 602 provides six bidirectional DC-DC converters (not shown), each operably coupled to the high energy density hybrid modules 112. For example, by configuring the bidirectional DC-DC converters to supply 10 kW of power, the energy management system 602 can provide 60 kW (6 × 10 kW) of bidirectional 48-500 V DC-DC with a peak efficiency of 98.5%. Of course, the specific arrangement of voltage, power capacity, and other features is not limiting, and other configurations can be obtained in light of this specification. The examples in this disclosure are for illustrative purposes only and are not limiting to exemplary embodiments. Additional operations, actions, tasks, activities, and operations can be conceived from this disclosure, and the same is contemplated within the scope of exemplary embodiments.

[0063] Conventional battery capacities in current electric vehicles range from a mere 17.6 kWh in some smart cars with a range of just 58 miles to up to 100 kWh in some Tesla models (Tesla is a trademark of Tesla Ltd. in the United States and other countries). As shown in Figure 7, by introducing a scalable architecture, various configurations can be provided to meet various range requirements. In an exemplary embodiment of Figure 7, by providing five additional high-energy-density hybrid modules 112 in configuration 2 704 compared to configuration 1 702, the available capacity increases from 130 kWh to 200 kWh, and by introducing another five additional high-energy-density hybrid modules 112 in configuration 2 704, a capacity of 270 kWh is obtained for configuration 3 706. Furthermore, the scalable architecture allows for unlimited placement of individual modules in various locations within the vehicle other than the conventional arrangement on the chassis 304 (Figure 3) of the vehicle 302, since each module only needs to be individually connected to a traction battery or a high-voltage DC bus. As disclosed herein, for example with respect to Figure 4, various high-energy-density modules may use different chemicals, allowing for additional flexibility in use cases, energy density, expected cycle life, etc.

[0064] Figure 8 shows another configuration of a power supply system having a traction battery 102, a plurality of high energy density hybrid modules 112, and a plurality of bidirectional DC-DC converters 502. In this configuration, a module is disabled by an SOH check indicating a problem with cell 114. The disabled hybrid module 802 is taken offline and can undergo formation recharging to extend its lifespan, where the module is slowly discharged, for example, over 20 hours at a defined temperature, and then slowly recharged, for example, over another 20 hours, to rebuild its chemistry. The modular nature of this configuration provides that the vehicle remains usable during formation recharging without the need to physically remove the disabled hybrid module 802. In the examples herein, a bleeder resistor of cell 114 is used in the charging and discharging operations.

[0065] This figure also shows a reduced-capacity hybrid module A 804, a reduced-capacity hybrid module B 806, and a normal-capacity hybrid module 808. The hybrid module controller 118 of the reduced-capacity hybrid module A 804 or the reduced-capacity hybrid module B 806 is configured to detect problems with cell 114 and independently make decisions regarding its discharge rate, for example, by reducing the power output from 2 kW to 1 kW.

[0066] In step 902, process 900 provides a traction battery comprising one or more traction modules connected to the high-voltage DC bus of the electric vehicle 302 and controlled by a battery management system (BMS) to be disconnected from the high-voltage DC bus of the electric vehicle 302. Here, the traction battery is configured to supply power to the electric vehicle 302. In step 904, process 900 provides a hybrid range extender battery 124 comprising a plurality of high-energy-density hybrid modules 112 connected in parallel to each other and connected to a high-voltage DC bus, to which the traction battery 102 is also connected. Each of the plurality of high-energy-density hybrid modules 112 includes a corresponding hybrid module controller (HMC) and a plurality of cells 114 connected in series. The health of each cell 114 is configured to be independently measurable by the corresponding HMC. The state of charge (SOC) of each cell can also be controlled via a balancing device 128, such as a bleeder resistor, connected in parallel with the cell 114. Therefore, each module's cell can be controlled independently and as a whole.

[0067] In step 906, multiple bidirectional DC-DC converters 502 are placed between multiple high-energy-density hybrid modules 112 and the high-voltage DC bus of the electric vehicle 302, and / or between multiple high-energy-density hybrid modules 112 and the traction battery 102.

[0068] Process 900 operably couples DC currents from one or more of the multiple high-energy-density hybrid modules 112 to the traction battery 102 (step 908) and / or the high-voltage DC bus of the electric vehicle 302 (step 910) in order to charge the traction battery and / or power the electric vehicle 302. In step 912, Process 900 controls the power generation mode of the power system by acquiring sensor information about independently measurable cells 114. In step 914, Process 900 uses each of the multiple corresponding HMCs to control the charge and discharge rates of the corresponding high-energy-density hybrid module through sensor information acquired about independently controllable cells.

[0069] Intelligent Power Control

[0070] The exemplary embodiments further recognize that conventional electric vehicle power systems configured to estimate the state of health (SOH) or state of charge (SOC) of component batteries are largely reactive, unable to predict energy consumption needs, and constrained to largely utilize the remaining available energy retrospectively. The exemplary embodiments recognize that while estimates can be obtained in the present according to the state of the recognized object, few or no mitigation measures are available to ensure battery safety or to preserve the battery's available lifespan and capacity. Furthermore, the load-following characteristics of conventional electric vehicle power computer systems, which cannot control changing drive power, mean that the current input and output of battery modules cannot be precisely controlled.

[0071] Existing conventional batteries charge and discharge all individual modules together, as long as the chemicals of the individual modules of the power system are managed. However, the embodiments disclosed herein recognize that monitoring the chemicals of individual battery modules within a larger power system and controlling them independently to ensure the safety of the system as a whole can offer additional advantages not available in conventional battery systems and electric vehicles. For example, in conventional systems, the safety of the power system cannot be guaranteed because individual modules cannot be disabled for recharging without having to disable the larger power system, and the available lifecycle of individual modules is excessively shortened by overcharging and over-discharging.

[0072] The embodiments disclosed herein recognize that currently available tools or solutions do not address the need to provide intelligent management of individual modules in a hybrid architecture to supply additional power when needed, while maintaining or maximizing the battery lifecycle and therefore the lifespan, safety, and maximum capacity of the individual modules, in a manner that enables the achievement of range and distance targets. The exemplary embodiments used to illustrate the present invention may address and solve the problems described above and other related problems by intelligently supplying power to an electric vehicle via a high-energy-density hybrid module 112 in a power supply system. The exemplary embodiments may solve these problems in a pre- and / or preparation process that forecasts the power demand of the electric vehicle and operates to meet said demand.

[0073] In one embodiment, certain actions are described as occurring in a particular component or location. Such locality of an action is not intended to be an limitation of the exemplary embodiments. Any action described herein as occurring in or being performed by a particular component, such as predictive analytics on battery data and / or natural language processing (NLP) analytics on contextual calendar data, can be implemented such that a function specific to one component causes or is performed in another component, for example, in a local or remote machine learning (ML) or NLP engine, respectively.

[0074] One embodiment monitors and manages the cumulative energy of a hybrid power system. Another embodiment monitors various profile sources configured for a user. A profile source is an electronic data source from which information can be obtained that is used to determine the user's profile characteristics. For example, a profile source could be the user's preference settings on a computer device, such as desired speed or route; a calendar application where the user's future events are planned and past events are recorded; a destination in a Global Positioning System (GPS) application where the user enters the current destination; or feedback from the user or community. A profile source can be a device, apparatus, software, or platform that provides information that can be used to derive the user's driving characteristics. For example, the dashboard of an electric vehicle can act as a profile source within the scope of an exemplary embodiment. Furthermore, a community, such as a fleet of electric vehicles, can be a profile source from which multiple driving characteristics of the user profile can be obtained to derive preferences, tastes, emotions, or usage of electric vehicles. Furthermore, measured health indices or parameters for individual modules of a vehicle fleet's battery packs can be a profile source from a community and can be used to learn from and derive patterns for powering electric vehicles in question. Therefore, batteries from a fleet of vehicles can adapt their predictions and share values ​​for the purpose of prediction / proposal.

[0075] User profile data, information, and preferences are terms used interchangeably herein to indicate constraints of one or more users that influence the power supply in the power system. Furthermore, information / data about the electric vehicle and power system 100 (such as vehicle speed, module current, temperature, voltage, impedance, health, charge status, average energy consumption, or other subject electric vehicle parameters 1220) may form part of the constraints, or be separate from the constraints, and may be acquired for use as input to the intelligent power control module for predictive analysis, as described below. Thus, the profile source information and electric vehicle and power system data (subject electric vehicle parameters 1220) collectively form input data 1202 or at least part of the constraints for the intelligent power control module to predict the level of output power to be obtained from individual batteries of the hybrid architecture to achieve a range or distance target, taking into account the safety, battery life, and battery capacity of the power system 100 in the electric vehicle, hereinafter referred to as the attributes of the power system 100.

[0076] Therefore, input data can be determined directly from measurements taken from the components of the electric vehicle. Input data can also be directly indicated in the profile source information. For example, a user may have explicitly stated preferences for destination arrival time or range targets for a specified period or until further changes in preferences occur.

[0077] Input data can also be derived from information collected from profile sources. For example, one embodiment may be configured to analyze the user's calendar to derive the time of arrival at a destination. Furthermore, information such as text or comments in the driving network may be analyzed contextually, for example, to determine future traffic conditions. In another example, the landscape of a geographical area may be obtained from an environmental profile and investigated to establish the nature of the terrain (e.g., the presence of steep slopes in mountainous areas obtained from an imaging device or database) and, therefore, the need to increase the battery power output.

[0078] The input data determined by one embodiment may change over time. For example, a user may prefer a predetermined route for short driving distances, such as commuting to and from work, and a route that optimizes energy consumption during long-distance vacation trips. Therefore, preferences may change as vacation driving characteristics derived from the user profile become prioritized. In such cases, the intelligent power control module may prioritize the use of the high-energy-density hybrid module 112 of the hybrid range extender battery 124 over the use of the traction module 122 of the traction battery 102 in order to extend the range of the traction battery 102.

[0079] Similarly, commuting by car may not require the use of the hybrid range extender battery 124. However, the user's "fast driving characteristics" may take precedence, for example, due to the judgment that the usual commute route is congested and the contextual establishment that the user has a meeting in an hour, and therefore the vehicle navigation system may choose a new route, even if it is mountainous, and abandon the usual route. Based on predictive analysis regarding whether the power or energy required to traverse the mountainous route in less than an hour is greater than, or at least greater than, the available traction battery power or energy, the power control module determines that the high energy density hybrid module 112 is required to complete the commute by car. Furthermore, the power control module may be configured to independently and automatically pre-charge the traction battery 102 to at least a threshold charge via a bidirectional DC-DC converter 502 connected to the high energy density hybrid module 112 in anticipation of driving when the context of the meeting is established.

[0080] Importantly, the power control module can control the output power obtained from one or more high energy density hybrid modules 112 while simultaneously ensuring that the safety, maximum lifecycle, and maximum capacity attributes of each individual high energy density hybrid module 112 are taken into consideration. For example, if the power control module determines that high energy density hybrid module 112A is faulty based on sensor information obtained with respect to independently measurable cells of high energy density hybrid module 112A, it may deactivate module A and utilize high energy density hybrid module 112B to pre-charge the traction battery 102, thus ensuring the safety of the battery pack and allowing the deactivated module A to eventually recover through re-charging. In another example, if the power control module determines that high energy density hybrid module 112C has 6 lifecycles remaining, it may prioritize the depleted module C before utilizing power from another module. User feedback indicating the accuracy of the output power to be obtained from the high energy density hybrid modules 112 as determined by the power control module is used to modify the power control module for better results.

[0081] Operating with profile information from one or more profile sources, one embodiment routinely evaluates constraints applicable to users of an electric vehicle. This embodiment may add new constraints / input data if found in the profile information analysis, modify existing constraints if justified by the profile information analysis, and reduce the use of past constraints depending on feedback, observed usage of constraints, and / or the presence of support for past constraints in the profile information. Past constraints can be reduced or made obsolete by lowering their priority to some extent, including removing / deleting / disabling past constraints. Sources of profile information may include, for example, calendar entries in a phone, tablet, or other device paired with the vehicle and / or management system, scheduling accounts associated with the user that the user may have access to, apps, and direct input of information by the user. More generally, profile information may be obtained from any source directly or indirectly available to the vehicle that can be associated with the user or owner of the vehicle.

[0082] Operating with profile information from one or more profile sources, one embodiment predicts a user's activity over a future period of time. For example, based on calendar information, an embodiment might determine, for instance, by NLP of calendar entries, that the user plans to work at location A at 9 a.m. tomorrow, have lunch from 12 p.m. to 1 p.m., and visit an out-of-state doctor after 3 p.m. This embodiment derives tomorrow's energy requirements based on the calendar activity and either pre-charges the traction battery 102 using one or more high energy density hybrid modules 112 or allocates the high energy density hybrid modules 112 to be used tomorrow. The allocation can also be done without using NLP to interpret the calendar data, as this is not intended to be limiting. Furthermore, these examples, such as input data / constraints, prioritization, and secondary considerations, are not intended to be limiting. From this disclosure, those skilled in the art can conceive of many other embodiments applicable to similar purposes, and these embodiments are construed within the scope of the exemplary embodiments.

[0083] The intelligent power control systems and techniques described herein are generally not available in conventional methods in the field of the art related to electric vehicles. One embodiment of the method described herein, when implemented to run on a device or data processing system, includes a substantial advance in the functionality of the device or data processing system in terms of power output by using a hybrid battery architecture that enables control of input and output currents while obtaining constraint proposals and ensuring the maximization of the safety, lifespan, and capacity attributes of the hybrid battery architecture modules.

[0084] In a further embodiment, a machine learning engine may be provided to enhance the resolution and effectiveness of predictions made by the controller based on a comparison of sensed and received information. The machine learning engine may detect patterns and, based on these patterns, compare possible outcomes with energy demand profiles. When a user engages with a vehicle, data on movement may be collected and stored for analysis by the controller or another networked computerized device. To enable additional resolution in detecting patterns and predicting behavior, data on movement by multiple users in multiple electric vehicles may be aggregated.

[0085] For example, a driver may be traveling on a county road that connects to an interstate highway. A geolocation sensor may detect that the vehicle is on the road and heading towards the interstate highway. Data collected by multiple vehicles may indicate that the majority of vehicles traveling on this county road in the direction of the interstate highway are likely to enter the interstate highway. Data collected by multiple vehicles may initially indicate that the driver is typically entering the interstate highway in a southbound direction, for example, towards a city or location with multiple workplaces.

[0086] A machine learning engine could use this information to predict an energy demand profile for a trip. This profile may be provided as a baseline because it may change as real-world conditions deviate from the predicted trip. Considering the example above, the driver of the electric vehicle may deviate from the predicted trip and drive past the entrance to the interstate highway. The machine learning engine could then determine the deviation from the predicted trip that occurred and update the energy demand profile to reflect the next most likely scenario, such as driving to a frequently visited relative's house 30 miles away from the interstate highway access point, or to another location.

[0087] In another example provided without limitation, sensors positioned in a vehicle may detect the amount of torque required to move the electric vehicle forward. In this example, it may be determined that substantially more torque is required to accelerate the vehicle. It may also be determined that more regenerative energy is generated as the vehicle decelerates. A machine learning engine may determine that the vehicle is towing another mass. The controller may perform calculations to modify the amount of stored energy required to complete the movement while towing the mass. These modifications may be applied and may affect the energy demand profile to reflect the additional energy demand of the movement. For example, the controller may adjust the energy demand profile to predict higher energy demand when towing mass.

[0088] The predictive profile of the machine learning engine may include an aggregated or baseline profile that can represent the general route characteristics and energy consumption needs of the user as a whole. The predictive profile of the machine learning engine may also include a local profile that shows the typical travel performed by the user, common destinations, and other general characteristics related thereto. In one embodiment, an energy demand profile may be generated for each user, and in this embodiment, the user may be identified by a dedicated key fob, a mobile computing device connected to the vehicle's entertainment system, voice recognition, a seat weight sensor, and / or other information indicating the operator's identity. Once the machine learning engine recognizes the operator, it may adjust its predictive model to fit statistically likely routes, driving habits, and other useful characteristics of the operator. The energy demand profile may be adjusted as appropriate with respect to the local profile associated with the operator.

[0089] A machine learning engine may operate by updating assumptions about vehicle parameters and predicting destination-weighted energy demand. Assumptions about the vehicle and expected movement may be made at the start of a journey. These assumptions may be supported by information determined by the vehicle and may be recorded as time-series data that can be used to calculate physical parameters. Exemplary physical parameters may include speed, net power of the battery system, traction motor power, geographical location, and other parameters that would be understood by a person skilled in the art after benefiting from this disclosure. Additional information may be derived, such as information related to geographical location, including latitude, longitude, bearing, altitude, speed, acceleration, inertia, and other information.

[0090] Vehicle-derived information can be supplemented by information supplied via the network. Such information may include wind speed, weather information, route, distance to destination, elevation and terrain profile of the route, traffic conditions, and other information that may influence the energy requirements of the journey.

[0091] The machine learning engine may perform analysis on time-series data collected in the vehicle, supplementary information such as that provided via a network, and / or other information to extract correlations. For example, the machine learning engine may perform linear algebraic regression analysis on time-series step data to find best-fit vehicle parameter values. Examples of best-fit vehicle parameter values ​​may include mass, rolling resistance coefficient, aerodynamic resistance coefficient, and other values ​​that a person skilled in the art would understand. The machine learning engine may additionally return vehicle parameters that can be used by the controller in energy management, such as mass, rolling resistance coefficient, aerodynamic resistance coefficient, average auxiliary power load, and other return parameters that a person skilled in the art would understand. An exemplary calculation that may be used to determine the average auxiliary electrical load power may be, but is not limited to, the sum of the net power of the battery system minus the sum of the traction motor power.

[0092] A machine learning engine can be advantageous in predicting destination-weighted energy needs. These energy needs can help determine whether to transfer electrical energy stored in a high-energy range battery to a high-power traction battery for use by an electric vehicle or other load. When making predictions, the machine learning engine can determine routes from the current location to various candidate charging locations. Travel information, such as directions provided by the user's mobile computing device and directions predicted based on the driver's behavior history, may be received by the vehicle's navigation system. The existence of charging locations, such as the user's home or public charging facilities, may be determined based on travel history, sources provided by the internet, navigation directions, and other sources.

[0093] Potential charging locations may be preferred if they are located within an acceptable proximity range to the travel destination. Preferred charging locations may be favored by the machine learning engine in calculating the expected energy demand. Similarly, undesirable charging locations may not be highlighted and / or removed when determining the expected energy demand for the trip.

[0094] Continuing the above example, a machine learning engine could calculate various route options to guide the operator from their starting point to their designated destination. These options might consider factors such as the presence of charging stations, anticipated road stoppages, acceptable and unacceptable distances between charging stations, changes in elevation, traffic conditions, and other characteristics associated with each route option. The machine learning engine might dislike route options that it deems inaccessible given the current state of the operator's vehicle's charging.

[0095] In another example provided without limitation, the machine learning engine might determine that a high-power traction battery is approximately 50% charged. In this example, the first route, such as the most direct route, might require at least 75% charge to reach a charging station under normal operation. An alternative route might be identified that presents the user with a charging station that requires only 25% charge to reach. The machine learning engine could then recommend a route that provides faster access to the charging station, thus avoiding the need to bring at least a portion of the high-energy range battery online.

[0096] Furthermore, in this example, the operator may choose to override the recommended route, for example, by driving on an alternative route. If the machine learning engine determines that the operator has chosen to take an undesirable route and has begun to move in a direction that indicates it will follow that undesirable route, it may instruct at least a portion of the high-energy range battery to come online to provide any supplemental energy that may be needed to reach a charging facility located outside the expected capacity remaining in the high-power traction battery.

[0097] As will be understood by those skilled in the art, the machine learning engine may provide varying weights to perceived information, conditions, parameters, movement details, and other factors that may influence the estimated energy consumption required to adhere to the expected energy consumption profile. Illustrative parameters that may be weighted to influence the expected energy consumption profile may include geographical location, GPS location, time, day of the week, vehicle mass, mass towed by the vehicle, temperature, auxiliary power demand, rolling resistance coefficient, aerodynamic coefficient area, time since the battery package was last charged, time since the last charging session occurred at the candidate location, and / or other factors and parameters that would be apparent to those skilled in the art after becoming interested in this disclosure.

[0098] The machine learning engine can then correlate these parameters to predict energy needs based at least partially on the weighted effects of the parameters considered. For example, the machine learning engine could apply a calculation that considers energy needs to be approximately equal to the sum of the mass contained by the vehicle and any other mass towed or transported by the vehicle. This value can then be multiplied by the expected energy required to complete the expected route. The machine learning engine can then analyze these factors and predict the expected energy needs profile associated with the expected movement. The battery package controller can then transfer power between the high-energy range battery and the high-power traction battery to compensate for the expected deficit in the charge state currently held by the high-power traction battery.

[0099] Here, we will discuss redundancy in more detail. In one embodiment, redundancy may be provided to mitigate the risk of one or more battery components experiencing complete depletion of stored energy and / or failure. Multiple energy management components may be included so that the failure of one energy management component is less likely to lead to a failure of the entire system. In one example, the battery package may include modular components, including connected electric battery management components, high-power traction battery modules, high-energy range battery modules, cooling functions, and / or other embodiments that support storage and power supply. In this example, if one modular component fails, the remaining modular components may continue to provide power from their connected embodiments.

[0100] In one embodiment, an independent observer module may be included to provide backup functionality otherwise provided by the energy management components. In this example, the independent observer module can continue to power the vehicle or other connected loads from the electrical energy stored in the battery packages, even in the event of a failure of an energy management component otherwise connected to the respective battery packages. For example, in the event of a failure of a connected energy management component, the redundant functionality of the independent observer module may take over the operation of the energy management so that the connected load, e.g., the vehicle, can continue to operate substantially safely until such a problem that would cause intervention by the independent observer module can be investigated and / or repaired. By providing such redundant and safety features to mitigate system failures, the systems enabled by this disclosure may be certified as an ASIL D architecture.

[0101] Exemplary embodiments are described merely as examples with respect to specific types of data, functions, algorithms, equations, model configurations, locations of embodiments, additional data, devices, data processing systems, environments, components, and applications. Any particular manifestation of these and other similar artifacts is not intended to be a limitation to the present invention. Any suitable manifestation of these and other similar artifacts can be selected within the scope of the exemplary embodiments.

[0102] Furthermore, exemplary embodiments may be implemented with respect to any type of data, data source, or access to a data source via a data network. Within the scope of the invention, any type of data storage device may provide data to embodiments of the invention locally in a data processing system or via a data network. When describing an embodiment using a mobile device, within the scope of the exemplary embodiment, any type of data storage device suitable for use in a mobile device may provide data to such embodiment locally in the mobile device or via a data network.

[0103] Exemplary embodiments will be described using specific code, designs, architectures, protocols, layouts, schematics, and tools, only as examples and not as limitations to the exemplary embodiments. Furthermore, exemplary embodiments will be described using specific software, tools, and data processing environments, where applicable, only as examples for the sake of clarity of explanation. Exemplary embodiments may be used in conjunction with other equivalent or similar structures, systems, applications, or architectures. Thus, for example, other equivalent mobile devices, structures, systems, applications, or architectures may be used in conjunction with such embodiments of the invention that fall within the scope of this invention. Exemplary embodiments may be implemented in hardware, software, or a combination thereof.

[0104] The examples in this disclosure are used solely for illustrative purposes and are not limiting to exemplary embodiments. Additional data, behaviors, actions, tasks, activities, and operations can be conceived from this disclosure, and the same is contemplated within the scope of the exemplary embodiments.

[0105] Any advantages listed herein are merely examples and are not intended to be limitations to exemplary embodiments. Additional or different advantages may be realized by specific exemplary embodiments. Furthermore, specific exemplary embodiments may have some or all of the advantages listed above, or none at all.

[0106] Referring to the figures, and in particular to Figures 10 and 11, these figures are illustrative diagrams of a data processing environment in which exemplary embodiments may be implemented. Figures 10 and 11 are not intended to assert or imply any limitations regarding environments in which different embodiments may be implemented. Certain implementations may involve many modifications to the illustrated environment based on the following description.

[0107] Figure 10 shows a block diagram of a network of a data processing system in which an exemplary embodiment may be implemented. The data processing environment 1000 is a network of computers in which an exemplary embodiment may be implemented. The data processing environment 1000 includes a network / communication infrastructure 1002. The network / communication infrastructure 1002 is a medium used to provide communication links between various devices, databases, and computers connected within the data processing environment 1000. The network / communication infrastructure 1002 may include connections such as wires, wireless communication links, or fiber optic cables.

[0108] The clients or servers are merely illustrative roles of specific data processing systems connected to the network / communication infrastructure 1002 and are not intended to exclude other configurations or roles for these data processing systems. Servers 1004 and 1006 are coupled to the network / communication infrastructure 1002 together with the storage unit 1008. Software applications may run on any computer within the data processing environment 1000. Clients 1010, 1012, and the dashboard 1014 are also coupled to the network / communication infrastructure 1002. Client 1010 may be a remote computer with a display. Client 1012 may be a mobile device configured with applications that transmit or receive information, such as receiving the charge status of the power system 100 or transmitting information about the user's calendar. The dashboard 1014 may be located inside an electric vehicle and may be configured to transmit or receive any of the information discussed herein. A data processing system such as server 1004 or server 1006, or a client (client 1010, client 1012, dashboard 1014), may contain data and may have software applications or software tools that run on it.

[0109] For illustrative purposes only, and without implying any limitations on such architectures, Figure 10 shows specific components that can be used in an exemplary implementation of one embodiment. For example, the servers and clients are merely examples and do not imply any limitation to a client-server architecture. As another example, one embodiment can be distributed across several data processing systems and data networks, as shown. In contrast, another embodiment can be implemented on a single data processing system within the scope of the exemplary embodiment. The data processing system (server 1004, server 1006, client 1010, client 1012, dashboard 1014) also represents exemplary nodes in a cluster, partition, or other configuration suitable for implementing one embodiment.

[0110] The power supply system 100 includes a traction battery 102 containing one or more traction units and a hybrid range extender battery 124 containing one or more high energy density hybrid modules 112. As discussed, one or more high energy density hybrid modules 112 are constructed using chemicals that prioritize high energy density over available cycle life, and each of the high energy density hybrid modules 112 includes a corresponding hybrid module controller 118 and cells connected in series, and each of the cells is configured to be independently measurable by the corresponding hybrid module controller 118.

[0111] Any other application, such as client application 1020, dashboard application 1022, or server application 1016, implements one embodiment described herein. Any of the applications can use data from the power system 100 and profile sources to predict power or energy requirements. The application can also retrieve data from storage unit 1008 for predictive analytics. The application can also run on any data processing system (server 1004 or server 1006, client 1010, client 1012, dashboard 1014).

[0112] Servers 1004, 1006, 1008, 1010, 1012, and 1014 can be connected to the network / communication infrastructure 1002 using wired connections, wireless communication protocols, or other suitable data connections. Clients 1010, 1012, and 1014 may be, for example, mobile phones, personal computers, or network computers.

[0113] In the illustrated example, server 1004 may provide data such as boot files, operating system images, and applications to clients 1010, client 1012, and dashboard 1014. Clients 1010, client 1012, and dashboard 1014 may be clients to server 1004 in this example. Clients 1010, client 1012, and dashboard 1014, or any combination thereof, may include their own data, boot files, operating system images, and applications. The data processing environment 1000 may include additional servers, clients, and other devices not shown.

[0114] The server 1006 may include a search engine configured to retrieve information such as terrain conditions, speed limits, user feedback, alternative profile sources, GPS information, traffic conditions, or other driving characteristics, as well as battery measurements (e.g., real-time battery measurements from individual cells of the high-energy-density hybrid module 112), in response to requests from operators for power supply as described herein in relation to various embodiments.

[0115] In the illustrated example, the data processing environment 1000 could be the Internet. The network / communication infrastructure 1002 could represent a collection of networks and gateways that use the Transmission Control Protocol / Internet Protocol (TCP / IP) and other protocols to communicate with one another. At the heart of the Internet is the backbone of data communication links between major nodes or host computers, including thousands of commercial, governmental, educational, and other computer systems that route data and messages. Of course, the data processing environment 1000 could be implemented as several different types of networks, such as an intranet, a local area network (LAN), or a wide area network (WAN). Figure 10 is intended as an example and not as an architectural limitation to different exemplary embodiments.

[0116] Among other uses, the data processing environment 1000 can be used to implement a client-server environment in which exemplary embodiments can be implemented. The client-server environment allows software applications and data to be distributed across a network so that applications function by using interactivity between a client data processing system and a server data processing system. The data processing environment 1000 can also utilize a service-oriented architecture in which interoperable software components distributed across a network can be packaged together as a consistent business application. The data processing environment 1000 may also take the form of a cloud and utilize a cloud computing model of service delivery to enable convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly distributed and released with minimal administrative effort or interaction with service providers.

[0117] Figure 11 shows a block diagram of a data processing system in which an exemplary embodiment may be implemented. The data processing system 1100 is an example of a computer, such as client 1010, client 1012, dashboard 1014, or server 1004, server 1006 in Figure 10, or another type of device in which computer-readable program code or instructions that implement the processes relating to the exemplary embodiment may be located.

[0118] The data processing system 1100 will be described as a computer, for example only, and not limited to this. Implementations in the form of other devices shown in Figure 10 may modify the data processing system 1100, such as by adding a touch interface, without deviating from the general description of the operation and function of the data processing system 1100 described herein, and certain illustrated components may even be removed from the data processing system 1100.

[0119] In the illustrated example, the data processing system 1100 uses a hub architecture that includes a northbridge and memory controller hub (NB / MCH) 1102 and a southbridge and input / output (I / O) controller hub (SB / ICH) 1104. The processing unit 1106, main memory 1108, and graphics processor 1110 are coupled to the northbridge and memory controller hub (NB / MCH) 1102. The processing unit 1106 may include one or more processors and may be implemented using one or more heterogeneous processor systems. The processing unit 1106 may be a multicore processor. In certain implementations, the graphics processor 1110 may be coupled to the northbridge and memory controller hub (NB / MCH) 1102 via an accelerated graphics port (AGP).

[0120] In the illustrated example, a local area network (LAN) adapter 1112 is coupled to the southbridge and input / output (I / O) controller hub (SB / ICH) 1104. An audio adapter 1116, a keyboard and mouse adapter 1120, a modem 1122, read-only memory (ROM) 1124, a universal serial bus (USB) and other ports 1132, and a PCI / PCIe device 1134 are coupled to the southbridge and input / output (I / O) controller hub (SB / ICH) 1104 via bus 1118. A hard disk drive (HDD) or solid state drive (SSD) 1126a and a CD-ROM 1130 are coupled to the southbridge and input / output (I / O) controller hub (SB / ICH) 1104 via bus 1128. The PCI / PCIe device 1134 may include, for example, an Ethernet adapter, an add-in card, and a PC card for a notebook computer. PCI uses a CardBus controller, while PCIe does not. Read-only memory (ROM) 1124 may be, for example, a flash binary input / output system (BIOS). Hard disk drives (HDDs) or solid-state drives (SSDs) 1126a and CD-ROMs 1130 may use, for example, Integrated Drive Electronics (IDE), Serial Advanced Technology Attachment (SATA) interfaces, or variations such as External SATA (eSATA) and Micro SATA (mSATA). Super I / O (SIO) devices 1136 may be coupled to the Southbridge and Input / Output (I / O) Controller Hub (SB / ICH) 1104 via bus 1118.

[0121] Memory such as main memory 1108, read-only memory (ROM) 1124, or flash memory (not shown) are some examples of computer-usable storage devices. Hard disk drives (HDDs) or solid-state drives (SSDs) 1126a, CD-ROMs 1130, and other similar usable devices are some examples of computer-usable storage devices, including computer-usable storage media.

[0122] The operating system runs on the processing unit 1106. The operating system coordinates and provides control over the various components within the data processing system 1100 in Figure 11. The operating system could be a commercially available operating system for any type of computing platform, including, but not limited to, server systems, personal computers, and mobile devices. An object-oriented or other type of programming system may work in conjunction with the operating system, providing calls to the operating system from programs or applications running on the data processing system 1100.

[0123] Instructions for an operating system, an object-oriented programming system, and applications or programs such as application 1016 and client application 1020 in Figure 10 are located on a storage device, such as in the form of code 1126b on a hard disk drive (HDD) or solid-state drive (SSD) 1126a, and can be loaded into at least one of one or more memories, such as main memory 1108, for execution by the processing unit 1106. The processes of an exemplary embodiment can be executed by the processing unit 1106 using computer implementation instructions that may be located in memory, for example, main memory 1108, read-only memory (ROM) 1124, or one or more peripheral devices.

[0124] Furthermore, in some cases, code 1126b may be downloaded from a remote system 1114b via network 1114a, and a similar code 1114c may be stored on storage device 1114d; in other cases, code 1126b may be downloaded to a remote system 1114b via network 1114a, and the downloaded code 1114 may be stored on storage device 1114d.

[0125] The hardware in Figures 10 and 11 may vary depending on the implementation. Other internal hardware or peripheral devices, such as flash memory, equivalent non-volatile memory, or optical disc drives, may be used in addition to or instead of the hardware shown in Figures 10 and 11. Furthermore, the processes of the exemplary embodiments may be applied to multiprocessor data processing systems.

[0126] In some exemplary cases, the data processing system 1100 may be a personal digital assistant (PDA) typically configured with flash memory providing non-volatile memory for storing operating system files and / or user-generated data. The bus system may comprise one or more buses, such as a system bus, an I / O bus, and a PCI bus. Of course, the bus system may be implemented using any type of communication fabric or architecture that provides data transfer between different components or devices attached to the fabric or architecture.

[0127] The communication unit may include one or more devices used to transmit and receive data, such as a modem or network adapter. Memory may be, for example, main memory 1108, or a cache such as the cache found in the northbridge and memory controller hub (NB / MCH) 1102. The processing unit may include one or more processors or CPUs.

[0128] The illustrations in Figures 10 and 11, as well as the examples described above, do not imply any architectural limitations. For example, the data processing system 1100 may also be a tablet computer, laptop computer, or telephone device, in addition to taking the form of a mobile or wearable device.

[0129] When a computer or data processing system is described as a virtual machine, virtual device, or virtual component, the virtual machine, virtual device, or virtual component operates like the data processing system 1100 using virtualized manifestations of some or all of the components represented in the data processing system 1100. For example, in the virtual machine, virtual device, or virtual component, processing unit 1106 is manifested as a virtualized instance of all or some of the hardware processing units 1106 available in the host data processing system; main memory 1108 is manifested as a virtualized instance of all or some of the main memory 1108 that may be available in the host data processing system; and hard disk drive (HDD) or solid state drive (SSD) 1126a is manifested as a virtualized instance of all or some of the hard disk drive (HDD) or solid state drive (SSD) 1126a that may be available in the host data processing system. In such a case, the host data processing system is represented by the data processing system 1100.

[0130] With respect to Figure 12, this figure shows an exemplary configuration for intelligent power control according to an exemplary embodiment. Intelligent power control can be implemented using application 1204 in Figure 12. Application 1204 is an example of server application 1016, client application 1020, or dashboard application 1022 in Figure 10. Application 1204 receives or monitors a set of input data 1202, for example, in real time. The input data includes target electric vehicle parameters 1220 such as the current of the high energy density hybrid module, the temperature of individual cells 114 and the temperatures of adjacent cells, the voltage of cell 114, the impedance of cell 114, the health of cell 114, the capacity of cell 114, the calculated polarization curve or charge / discharge curve 500 of cell 114 that identifies the graphitization plateau, the maximum speed / acceleration of the vehicle, the total mass of the vehicle, the aerodynamic drag of the vehicle, location, and the nearest charging station. The input data also includes driving characteristics from profile source 1226 (user profile 1222, community profile 1224, environmental profile 1230), such as user preferences, planned number of stops during travel, average distance of daily driving, historical driving energy consumption per mile, duration of stops, calendar data, and environmental data such as terrain data, road slope angle, drag coefficient, and road rolling resistance coefficient.

[0131] In one or more embodiments described herein, characteristics, traits, and / or preferences related to users, communities, environments, target electric vehicles, power systems, etc., are referred to as “features.” In one or more embodiments, Configuration 1200 defines and configures algorithms and / or rules to drive feature selection results. In certain embodiments, the algorithm may include, for example, determining a minimum common value for features and determining whether that value satisfies the best match among users within a feature threshold (e.g., 90%). In one embodiment, the system may prioritize certain features such that features such as battery module safety, or time to arrival, or SOH, or driving distance, carry different weights. In one embodiment, after commonalities in a fleet of vehicles are found, Configuration 1200 extracts and derives the best feature values ​​to understand the problems of individual vehicles and help control the power of the target electric vehicles.

[0132] In one embodiment, the feature extraction component 1214 is configured to generate relevant features for a target electric vehicle using data from all different available features (e.g., target electric vehicle parameters 1220, user profile 1222, community profile 1224, environment profile 1230) based on the content of a request from application 1204. In this embodiment, the feature extraction component 1214 receives a request from application 1204 that includes at least identification information of the target electric vehicle 1232 and / or its user or location, as well as instructions proposing power output to be obtained from one or more high energy density hybrid modules 112 to complete a 10-mile journey. Using the target electric vehicle 1232 and / or user information, the feature extraction component 1214 obtains a combination of specific target electric vehicle parameters 1220, user profile information from user profile 1222, community profile information from community profile 1224, and environment data from environment profile 1230. In this embodiment, the feature extraction component 1214 uses a defined prioritization algorithm to generate features as feature profiles. In a particular embodiment, the feature profile includes each feature (e.g., 1. current in cell 114, 2. temperature of cell 114, 3. voltage of cell 114, 4. impedance of cell 114, 5. user calendar, 6. GPS location, 7. destination, 8. range requirements, 9. health audit report indicating the safety, capacity, and remaining lifecycle of cell 114, and 10. weights assigned to each feature). Using the extracted features and a trained M / L model trained with a number of different datasets, the power control module 1216 determines a power output proposal 1212 for the target electric vehicle 1232. The main advantage of the hybrid architecture, in which cell 114 uses a chemical composition that prioritizes high energy density over the number of available cycles that can be charged and / or discharged, is that the range of the traction battery 102 of the power system 100 is significantly increased.Furthermore, by individually controlling the current input and output of the high-energy-density hybrid module 112 having series-connected cells 114, an advanced modular architecture is obtained that enhances the safety of individual cells 114 or modules through the ability to control which modules are activated or deactivated, for example, in the event of a short circuit detection, to prevent localized faults from causing further damage. By modularly controlling the high-energy-density hybrid module 112 based on measurements obtained with respect to those component cells 114, the maximum lifecycle of each high-energy-density hybrid module 112 can be ensured by simply preventing the rapid degradation of cells that is typically associated with the undetected cell problem in parallel-connected cells of conventional solutions. For example, if one cell overheats and goes undetected, a chain reaction may begin that affects other cells. The ability to modularly control the current of individual series-connected cells and their charge / discharge rates via the balance device 128 ensures that the maximum capacity of the cells 114 and their available lifecycle are maintained. Therefore, by using a power control module 1216 based on a machine learning model that takes into account preferences and target electric vehicle health parameters, the output of individual high-energy-density hybrid modules 112 can be intelligently and in real time to efficiently address the vehicle's changing energy demands while enabling the user to achieve their range or destination goals without compromising the benefits provided by the hybrid architecture. In one embodiment, the power control module 1216 is trained to maximize the aforementioned benefits, as discussed herein.

[0133] Returning to Figure 12, the feature extraction component 1214 may be integrated into the deep neural network. Alternatively, the feature extraction component 1214 may be outside the deep neural network. The power control module 1216 uses features obtained from the feature extraction component 1214 to generate a power output proposal 1212, which may include information about the power or energy or C rate 522 required to operate one or more bidirectional DC-DC converters 502 to meet immediate or extended distance or range targets, based on a request from application 1204, for example. The power output proposal 1212 may also include information indicating the predicted state of one or more components of the power supply system 100 and instructions to mitigate predicted / potential failure modes. Furthermore, the power output proposal 1212 may include information on which one or more high-energy-density hybrid modules 122 to obtain defined power output from, the charge or discharge rate of the cell 144 or traction battery via one or more bidirectional DC-DC converters 502, the time to initiate such charging or discharging, an optimized route, etc. These examples are not limiting, and any combination of these and other exemplary power output proposals is possible as described. The power control module 1216 is not limiting, but may be based on a neural network, such as a recurrent neural network (RNN) and a dynamic neural network (DNN). An RNN is a type of artificial neural network designed to recognize patterns in a sequence of data and generate image descriptions and text summaries, such as numerical time series forecasting or prediction and numerical time series anomaly detection using data emitted from a sensor. RNNs use recurrent connections (moving in the opposite direction to the “normal” flow of signals) that form cycles within the network topology. Calculations derived from previous inputs are fed back into the network, which gives the RNN “short-term memory”. Feedback networks such as RNNs are dynamic, and their "states" change continuously until they reach an equilibrium point.For this reason, RNNs are particularly well-suited for detecting relationships over time in a given set of data. A recurrent network takes in not only the input examples it is currently viewing, but also those previously perceived in time. A decision reached by the recurrent network at time step t-1 influences a decision reached a little later at time step t. Thus, a recurrent network has two input sources: the current and the recent past, which are combined to determine how to respond to new data. DNNs rely on the dynamic declaration of the network structure. In traditional static models, a computation graph (where the symbolic representation of the computation by the neural network is usually defined) and then examples are fed to an engine, which performs this computation and calculates its derivative. However, with a static graph, the input size must first be defined, which can be inconvenient for applications where the input changes. In DNNs, however, a dynamic declaration policy is used, and the computation graph is implicitly constructed by executing procedural code that computes the network output, using the ability to use a different network structure for each input. Therefore, during the training process, the computation graph can be newly defined for each training example. Therefore, the computation graph is dynamically constructed immediately after the input variables are declared. Thus, the graph is flexible and allows for modification and inspection of its internal workings at any time. Thus, instead of having to maintain relationships between all inputs to the neural network and the layers of the neural network, we can decide that when a defined parameter exceeds a threshold for increasing its priority, the structure of the neural network is dynamically modified to cause a corresponding change in the output to address the new functional requirements of the power system 100 caused by the increase in priority, and vice versa. Thus, in a dynamic neural network, the output depends on the current and past values ​​of the inputs, outputs, and network structure.Neural networks with such feedback are well-suited for modeling, identifying, controlling, and filtering systems, and are particularly important for nonlinear dynamic power supply systems. Of course, these examples are non-limiting, and other examples can be obtained in light of this specification.

[0134] In an exemplary embodiment, the power output proposal 1212 may be presented by the presentation component 1208 of application 1204. The adaptive component 1210 is configured to receive user input to adapt the power output proposal 1212 as needed. For example, changing the route proposed by the power control module 1216 triggers a recalculation of the proposed power output, taking into account the terrain and distance of the new route.

[0135] The feedback component 1218 optionally collects user feedback 1224 regarding the power output proposal 1212. In one embodiment, the application 1204 is configured not only to calculate the power output proposal 1212 but also to provide a way for the user to input feedback, which indicates the accuracy of the calculated power output proposal 1212. The feedback component 1218 applies feedback in machine learning techniques to profiles 1222, 1224, 1230, or M / L model 1206, etc., to modify the M / L model 1206 for better proposals. In an exemplary embodiment, the application analyzes the feedback input and the application enhances the M / L model 1206 of the power control module 1216. If the feedback is positive or unsatisfactory regarding the accuracy of the proposal, the application enhances or weakens the parameters of the M / L model 1206, respectively. In one example, the proposal was to turn on the hybrid range extender battery 124 30 miles before reaching the mountain so that sufficient battery capacity and power would be available at the mountaintop, eliminating the need to limit power at the summit. However, if it is determined that power was actually limited at the mountaintop and therefore a lower speed could be maintained than expected, feedback is provided to the power control module 1216 regarding the inaccuracy of the proposal / prediction.

[0136] The input layer of the neural network model can be, for example, a vector representing the current, voltage, or impedance values ​​of cell 114, pixels from a 2D image of terrain data, or context calendar data provided by the NLP engine 1228. In one example, a CNN (Convolutional Neural Network) uses convolution to extract features from the input image. In one embodiment, upon receiving a request to provide a proposal, application 1204 creates an array of values ​​to be input to the input neurons of M / L model 1206 to generate an array containing the power output proposal 1212.

[0137] The neural network M / L model 1206 is trained using various types of training datasets, including memorized profiles and measurements from numerous sample vehicles and cells. As shown in Figure 13, which illustrates a block diagram of an exemplary training architecture 1302 for machine learning-based recommendation generation according to an exemplary embodiment, the program code extracts various features 1306 from the training data 1304. The components of the training data 1304 have labels L. The features are used to develop a predictive function H(x) or hypothesis, which the program code utilizes as the M / L model 1308. When identifying various features in the training data 1304, the program code may utilize various techniques, including, but are not limited to, mutual information, which are examples of methods that can be used to identify features in one embodiment. Other embodiments may utilize various techniques for feature selection, including, but are not limited to, principal component analysis, diffusion mapping, random forest, and / or recursive feature removal (a brute-force method for feature selection). "P" is an output that can be obtained (e.g., a power output value, a high energy density hybrid module 112 for obtaining the power output value, etc.), and when this output is received, it can further trigger the power system 100 or the vehicle to perform other steps, such as steps of stored instructions. The program code may utilize a machine learning m / l algorithm 1312 to train the M / L model 1308, including providing weights for the output so that the program code can prioritize various changes based on a prediction function including the M / L model 1308. The output can be evaluated by a quality metric 1310.

[0138] By selecting a diverse set of training data 1304, the program code trains an M / L model 1308 to identify and weight various features such as the target electric vehicle 1232, the driver, the vehicle fleet, and environmental conditions. To utilize the M / L model 1308, the program code acquires (or derives) input data or features to generate an array of values ​​to input to the input neurons of the neural network. In response to these inputs, the output neurons of the neural network generate an array containing power output proposals 1212 to be presented or used simultaneously.

[0139] Referring to Figure 14, this figure shows a flowchart of an exemplary process 1400 for providing a power output proposal for an electric vehicle according to an exemplary embodiment. Process 1400 can be implemented using application 1204 in Figure 12.

[0140] In step 1402, process 1400 independently measures the parameters of each of several cells in at least one high-energy-density hybrid module of the power supply system using at least one hybrid module controller (HMC). The several cells are connected in series in at least one high-energy-density hybrid module.

[0141] In step 1404, process 1400 receives measured parameters of a cell as at least part of a set of subject electric vehicle parameters that represent one or more characteristics of the subject electric vehicle for use by a power control module. The parameters may include at least current, temperature, and voltage. Other parameters, including capacitance, polarization curves with graphitization plateaus, and impedance (DC IR, AC IR), may be derived from single or time-series measurements of current, temperature, and voltage. For example, a polarization curve with graphitization plateaus (resulting in iron interpolation) may be used by process 1400 to interpret the type of failure occurring in cell 114, such as lithium loss, loss of active sites for lithium storage, etc.

[0142] In step 1406, process 1400 generates input data using at least a set of target electric vehicles. In step 1408, process 1400 extracts one or more features from the input data, one or more features representing requests to complete a power output proposal operation, such as a calendar of users with upcoming meetings scheduled. Feature extraction may be separate from the model or may be included in one or more layers of the model that have been adjusted during training. One or more features may also represent attributes obtained from the attribute prioritization step 1502, as shown in Figure 15. In attribute prioritization 1502, one or more attributes 1510 are obtained to be considered in the power output proposal operation. One or more attributes may be assigned different priorities or weights, or they may be assigned the same priority or weight, or even have no priority or weight assigned. Different scenarios can be handled by the power control module 1216 by training the M / L model 1206 with a large set of different datasets that consider the attributes 1510. In exemplary and non-limiting embodiments, attribute 1510 includes instructions for maximizing or enforcing safety attribute 1504, maximizing lifespan attribute 1506, and maximizing capacity attribute 1508. In step 1410, process 1400 uses a power control module to propose at least one power output proposal for the electric vehicle in question.

[0143] At least one power output proposal may take into account attributes of the power supply system 100 by attribute prioritization 1502, maximum safety 1504. In an exemplary embodiment, maximizing safety means taking into account possible or observed activity in the cell chemistry (e.g., a short circuit between the anode and cathode manifesting as self-discharge), and the power control module 1216 proposes and implements a temporary suspension of operation of the battery pack segment / high energy density hybrid module 112 without affecting other modules / high energy density hybrid modules 112, a step otherwise unavailable in conventional battery packs. This implementation may also include, for safety benefits, distancing energy from the high energy density hybrid module 112, or discharging and turning off the high energy density hybrid module 112 to insulate it. Furthermore, by observing an abnormal temperature rise without any corresponding current change, the power control module 1216 may infer a fire event, circuit board failure, etc., and therefore discharge the corresponding module adjacent to the temperature rise to avoid the propagation of fire or failure. In another example, by observing the loss of insulation between the chassis 304 and the high-voltage bus, the power control module reduces the charge state of one or more modules, provides a service warning, and thus maximizes the safety of the power supply system 100 and, consequently, the safety of the electric vehicle.

[0144] At least one power output proposal may take into account attribute prioritization 1502 and maximum lifespan 1506 of the power supply system 100. In an exemplary embodiment, maximizing lifespan includes maximizing the health of cell 114, i.e., the cell's ability to discharge current. By observing an increase in the battery's impedance, the power supply system 100 causes a change in the maximum power of cell 114 to avoid overheating or "excessive stress" of cell 114 in order to maximize the lifespan of cell 114. Thus, a defined discharge power is determined to complement the health of cell 114. In this embodiment, impedance is measured based on the discharge and recharge of the cell and a comparison with an ideal standard for discharge parameters and recharge, and cell 114 is graded in SOH grading operation. Cell 114 is graded, for example, as A, B, C, D, and E, where A represents high SOH and E represents low SOH. Therefore, in this embodiment, all modules having cells 114 graded D and E may be operated by the power control module 1216 at a C rate of C / 10 522, modules having cells graded B and C may be operated at a C rate of C / 5 522, and modules having cells graded A may be operated at a C rate of C / 3 522, where the operating C rate 522 is the discharge power limit of each high energy density hybrid module 112. The power control module 1216 continues to learn and adjust according to these limits, along with safety and capacity attributes. Therefore, if a cell 114 graded A and its module are taken offline due to a safety issue, another cell 114 may be upgraded from B to A, or its module may be set to a harder duty cycle, due to the absence of the offline cell 114.

[0145] At least one power output proposal may take into account the maximum capacity 1508 of the power supply system 100 by attribute prioritization 1502. In an exemplary embodiment, maximizing capacity takes into account the impedance problem of the cells. For cells with high impedance, the power control module 1216 can operate the corresponding high energy density hybrid module 112 at the lowest C rate 522, providing energy for the longest time and thus maximizing capacity, even if it is not highly likely to operate it first based solely on the lifetime attribute 1510. Furthermore, in the case of a series chain of cells in a group, the capacity of the group is limited by the weakest cell. If all cells have 100 AH and the weakest cell has 60 AH, once they reach zero charge, the remaining cells in the series chain cannot discharge further to avoid damaging the weakest cell, thus the weakest cell limits the other cells. The power control module 1216 operates to avoid capacity differences between cells in order to protect and prevent the weakest cell from degrading. Furthermore, the power control module 1216 may perform a slow charge by discharging the weakest cells during formation charging in order to restore the cell capacity.

[0146] Therefore, in an exemplary embodiment, the power control module 1216 operates on a merits-and-demerits system that works to maximize lifespan, safety, capacity, and other attributes, while also considering input data such as geography, maximum current, and speed, and predicting how all inputs will benefit attribute targets. Performing SOH checks frequently / periodically allows cell / module grading to continue tracking health for decision-making. For example, a calendar may be used to check upcoming moves, and SOH checks may be performed to identify weak battery modules to determine if they can be improved. Identified weak battery modules may be charged very slowly before the move to resolve health issues for use during the move.

[0147] Accordingly, computer implementation methods, systems, or devices and computer program products are provided exemplarily for the power supply of electric vehicles and other related features, functions, or operations. Where a part of an embodiment thereof is described with respect to a certain type of device, the computer implementation method, system, or device, computer program product, or a part thereof, is adapted or configured for use in a suitable equivalent manifestation of that type of device.

[0148] Where an embodiment is described as being implemented in an application, the delivery of the application in a Software as a Service (SaaS) model is intended to be within the scope of the exemplary embodiment. In a SaaS model, the ability of an application implementing an embodiment is provided to the user by running the application on a cloud infrastructure. Users can access the application using a variety of client devices via a thin client interface such as a web browser (e.g., web-based email) or other lightweight client applications. Users do not manage or control the underlying cloud infrastructure, including the network, servers, operating system, or storage of the cloud infrastructure. In some cases, users may not even manage or control the capabilities of a SaaS application. In some other cases, the SaaS implementation of an application may allow possible exceptions to limited user-specific application configuration settings.

[0149] The present invention may be a system, method, and / or computer program product at any possible level of technical detail of integration. The computer program product may include a computer-readable storage medium having computer-readable program instructions for causing a processor to perform aspects of the present invention.

[0150] A computer-readable storage medium can be a tangible medium capable of holding and storing instructions for use by an instruction execution device. A computer-readable storage medium may, for example, be an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. A non-exhaustive list of more specific examples of computer-readable storage media includes, but are not limited to, portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital multipurpose disks (DVDs), memory sticks, floppy disks, mechanically encoded devices such as punch cards or grooved raised structures on which instructions are recorded, and any suitable combination thereof. Computer-readable storage media, including computer-readable storage devices used herein, should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides, or other transmission media (e.g., light pulses passing through optical fiber cables), or electrical signals transmitted through wires.

[0151] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may include copper transmission cables, optical transmission fibers, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and transfers the computer-readable program instructions for storage in a computer-readable storage medium within each computing / processing device.

[0152] The computer-readable program instructions for performing the operations of the present invention may be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk and C++, and procedural programming languages ​​such as the "C" programming language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, via the Internet using an Internet service provider). In some embodiments, to carry out aspects of the present invention, an electronic circuit including, for example, a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA) may execute a computer-readable program instruction by utilizing state information of the computer-readable program instruction to personalize the electronic circuit.

[0153] Aspects of the present invention will be described herein with respect to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It will be understood that computer-readable program instructions can implement each block in the flowchart and / or block diagram, and combinations of blocks within the flowchart and / or block diagram.

[0154] These computer-readable program instructions may be provided to the processor of a general-purpose computer, a dedicated computer, or other programmable data processing device to generate a machine such that instructions executed via the computer's processor or other programmable data processing device create means for implementing functions / activities specified in one or more blocks of a flowchart and / or block diagram. These computer-readable program instructions may also be used to instruct computers, programmable data processing devices, and / or other devices to function in a particular way such that a computer-readable storage medium storing the instructions contains a product containing instructions that implements a mode of functions / activities specified in one or more blocks of a flowchart and / or block diagram.

[0155] Computer-readable program instructions can also be loaded onto a computer, another programmable device, or another device to cause a set of operational steps to be executed on the computer, another programmable device, or another device, in order to generate a computer implementation process in which instructions executed on the computer, another programmable device, or another device implement a function / activity specified in one or more blocks of a flowchart and / or block diagram.

[0156] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of instructions containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions shown within a block may occur out of the order shown in the figure. For example, two consecutively shown blocks may actually be executed substantially simultaneously, or sometimes these blocks may be executed in reverse order depending on the functionality they contain. It should also be noted that each block in a block diagram and / or flowchart, as well as combinations of blocks in a block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs a specified function or activity, or by executing a dedicated combination of hardware and computer instructions. [Explanation of Symbols]

[0157] 100 Power Systems 102 Traction Battery 104 Battery Management System (BMS) 106 Processors 108 Relay 110 Drive Unit 112 High Energy Density Hybrid Module 112A High Energy Density Hybrid Module 112B High Energy Density Hybrid Module 114 cells 116 Sensors 118 Onboard Hybrid Module Controller 120 processors 122 Traction Module 124 Hybrid Range Extender Battery 126 Onboard Computer System 128 Balance Device 200 Computer Systems 202 Communications Infrastructure 204 Main Memory 206 Computer Processors 208 Display Interfaces 210 Input Units 212 Communication Interfaces 214 Display Units 216 Communication Path 218 Secondary memory 220 hard disk drives 222 Removable Storage Drives 226 Removable Storage Unit 302 vehicles 304 Chassis 402 Percentage of operating days axis 404 Daily driving distance axis 406 Traction Battery Section 408 Hybrid Range Extender Section 410 Energy density axis 412-cycle life axis 414 Traction Battery Chemicals Area 418 Hybrid Range Extender Battery Chemicals Area 500 charge-discharge curve, charge-discharge curve 502 Bidirectional DC-DC Converter 504 Onboard AC-DC Charger 506 Auxiliary DC-DC Converter 508 Contactor 510 Control Module 512 12V Battery 514 Voltage axis 516 Capacity axis 518 Discharge curve 520 charging curve 522 C rating 602 Onboard Energy Management System 702 Configuration 1 704 Configuration 2 706 Configuration 3 802 Disabled Hybrid Module 804 Reduced Capacity Hybrid Module A 806 Reduced Capacity Hybrid Module B 808 Standard Dose Hybrid Module 1000 Data Processing Environments 1002 Network / Communication Infrastructure 1004 Server 1006 Server 1008 Storage Units 1010 Client 1012 Clients 1014 Dashboard 1016 Server Application 1020 Client Application 1022 Dashboard Application 1100 Data Processing System 1102 Northbridge and Memory Controller Hub (NB / MCH) 1104 Southbridge and Input / Output (I / O) Controller Hub (SB / ICH) 1106 Hardware Processing Unit 1108 Main Memory 1100 Graphics Processor 1112 Local Area Network (LAN) Adapter 1114a Network 1114b Remote System 1114c code 1114d Storage Device 1116 Audio Adapter 1118 Bus 1120 Keyboard and Mouse Adapter 1122 Modem 1124 Read-only memory (ROM) 1126a Hard disk drive (HDD) or solid state drive (SSD) 1126b Code 1128 Bus 1130 CD-ROM 1132 Universal Serial Bus (USB) and other ports 1134 PCI / PCIe devices 1136 Super I / O (SIO) Devices 1200 configuration 1202 Input Data 1204 Application 1206 M / L model 1208 Presentation Components 1210 Adaptive Components 1212 Power output proposal 1214 Feature Extraction Components 1216 Power Control Module 1218 Feedback Component 1220 Target Electric Vehicle Parameters 1222 User Profiles 1224 Community Profiles 1226 Profile Source 1228 NLP Engine 1230 Environmental Profile 1232 Target electric vehicles 1302 Training Architecture 1304 Training Data 1306 Features 1308 M / L model 1310 Quality Metrics 1312 Machine Learning m / l Algorithms 1502 Attribute Prioritization 1504 Safety Attributes 1506 Lifespan attribute 1508 Capacity attribute 1510 Lifetime Attributes

Claims

1. A power supply system for electric vehicles, A traction battery configured to be connected to or disconnected from the high-voltage DC bus of the electric vehicle in order to supply power to the electric vehicle, A hybrid range extender battery comprising one or more high energy density hybrid modules connected in parallel, wherein each high energy density hybrid module includes a corresponding hybrid module controller (HMC) and a plurality of cells connected in series, One or more bidirectional DC-DC converters disposed between the one or more high-energy-density hybrid modules and the high-voltage DC bus of the electric vehicle. Equipped with, A power supply system in which each of the one or more arranged bidirectional DC-DC converters operably couples a direct current from the corresponding high-energy-density hybrid module to the traction battery and / or powertrain via the high-voltage DC bus of the electric vehicle, for the purpose of charging the traction battery and / or supplying power to the electric vehicle.

2. The power supply system according to claim 1, wherein each of the one or more high energy density hybrid modules is constructed using a chemical that prioritizes high energy density over usable cycle life.

3. The power supply system according to claim 1, wherein each of the plurality of cells is configured to be independently measurable by the corresponding HMC.

4. The power supply system according to claim 1, wherein one or more high-energy-density hybrid modules are configured to manage charging and / or discharging via a corresponding bidirectional DC-DC converter.

5. The power supply system according to claim 1, wherein the corresponding HMC of the high energy density hybrid module is configured to further manage the power generation mode of the power supply system by controlling the charge and discharge rates of the high energy density hybrid module via sensor information acquired for the independently measurable cells.

6. The power supply system according to claim 1, further comprising a cell-by-cell balancing device of the high-energy-density hybrid module configured to selectively discharge the charge accumulated in the cell.

7. The power supply system according to claim 1, wherein the hybrid range extender battery comprises a plurality of chemical substances.

8. The power supply system according to claim 1, wherein at least one cell of a high-energy-density hybrid module has a cell energy density of about 1000 Wh / L or more.

9. The traction battery has a first chemical type and a cell energy density of 500 Wh / L or less. The power supply system according to claim 1, wherein the hybrid range extender battery has a second chemical type different from the first chemical type and a cell energy density of 1000 Wh / L or more.

10. The power supply system according to claim 1, wherein the traction battery is separated from the hybrid range extender battery.

Images