Estimating a temperature or a state of charge of an external battery of a power supply device

Abstract

A method is provided and includes receiving a temperature of a body of a power supply device and generating a temperature of an external battery of the power supply device using a model. A second method includes accessing a dataset that each includes a first voltage across a first external battery of a first power supply device, a first current flowing through the first battery, a first state of charge of the first battery, and a first temperature of the first battery. The second method includes training a model using the dataset to receive a second voltage across a second battery of a second power supply device, a second current flowing through the second battery, and a second temperature of the second battery, and generate a second state of charge of the second battery based on the second voltage, the second current, and the second temperature.

Description

Technical Field

[0001] This disclosure generally relates to estimating the temperature or state of charge of an external battery of a power supply device using a computational model, and more specifically, to estimating the temperature or state of charge of an external battery based on one or more temperatures of the body of the power supply device, the voltage across the battery, and / or the current flowing through the battery. Background Technology

[0002] Some power supply devices are configured to operate in normal mode and battery standby mode. In normal mode, the power supply device receives input power, regulates or modulates that input power, and provides that power to the load. When the input power is offline, battery standby mode can be triggered. An external battery is connected to the power supply device via a power cable. In battery standby mode, the power supply device receives power from the battery, regulates or modulates that power, and provides that power to the load.

[0003] It may be beneficial to reliably estimate the battery's state of charge, making it possible to assess the battery's readiness for operation in battery standby mode or its general condition. However, performing a direct measurement of the battery's state of charge, which would yield an accurate state of charge, is generally impractical during the operation of a power supply device. Summary of the Invention

[0004] A first example is a method for accessing a first dataset, each of which includes one or more first temperatures of a first body of a first power supply device and a second temperature of a first battery of the first power supply device. The first battery is located outside the first body. The method also includes using the first dataset to train a first computational model to perform a function. The function includes receiving one or more third temperatures of a second body of a second power supply device and generating a fourth temperature of a second battery of the second power supply device based on the one or more third temperatures. The second battery is located outside the second body.

[0005] The second example is a non-transitory computer-readable medium storing instructions that, when executed by a computing device, cause the computing device to perform the method of the first example.

[0006] The third example is a computing device, which includes: one or more processors; and a computer-readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform the method of the first example.

[0007] The fourth example is a method comprising: accessing datasets, each of which includes a first voltage across a first battery of a first power supply device, a first current flowing through the first battery, a first state of charge of the first battery, and a first temperature of the first battery. The first battery is located outside a first body of the first power supply device. The method further comprises using the datasets to train a computational model to perform a function. The function includes receiving a second voltage across a second battery of a second power supply device, a second current flowing through the second battery, and a second temperature of the second battery, and generating a second state of charge of the second battery based on the second voltage, the second current, and the second temperature.

[0008] The fifth example is a non-transitory computer-readable medium storing instructions that, when executed by a computing device, cause the computing device to perform the method of the fourth example.

[0009] The sixth example is a computing device comprising: one or more processors; and a computer-readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform the method of the fourth example.

[0010] The seventh example is a method comprising: receiving one or more first temperatures of the body of a power supply device, and using a first calculation model, generating a second temperature of a battery of the power supply device using the one or more first temperatures. The battery is located outside the body.

[0011] The eighth example is a non-transitory computer-readable medium storing instructions that, when executed by a computing device, cause the computing device to perform the method of the seventh example.

[0012] The ninth example is a computing device comprising: one or more processors; and a computer-readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform the method of the seventh example.

[0013] The use of the terms “approximately” or “substantially” in relation to the quantities or measurements described herein means that the recorded characteristics, parameters, or values ​​need not be precisely achieved, but rather include deviations or variations that may occur in a quantity that does not preclude the effects that the characteristic is intended to provide, such as tolerances, measurement errors, measurement accuracy limitations, and other factors known to those skilled in the art.

[0014] The features, functions, and advantages already discussed can be implemented independently in various examples, or combined in yet another set of examples, further details of which can be seen in the description and figures below.

[0015] This invention provides a set of technical solutions, as follows: Technical Solution 1. A method, comprising: Access a first dataset, each of the first datasets including one or more first temperatures of a first body of a first power supply device and a second temperature of a first battery of the first power supply device, wherein the first battery is external to the first body; and The first dataset is used to train a first computational model to perform a function, which includes: One or more third temperatures are received from the second body of the second power supply device; and A fourth temperature is generated for the second battery of the second power supply device based on the one or more third temperatures, wherein the second battery is located outside the second body.

[0016] Technical Solution 2. The method according to Technical Solution 1, wherein the first main body includes the control module of the first power supply device.

[0017] Technical solution 3. The method according to technical solution 2, wherein the one or more first temperatures are detected within the control module.

[0018] Technical Solution 4. The method according to Technical Solution 1, wherein the one or more first temperatures include the temperature of the inner surface of the outer shell of the first body.

[0019] Technical Solution 5. The method according to Technical Solution 1, wherein the one or more first temperatures include the temperature of a normal mode bypass circuit, the normal mode bypass circuit being configured to receive a main power input and generate a power output.

[0020] Technical Solution 6. The method according to Technical Solution 1, wherein the one or more first temperatures include the temperature of a battery mode bypass circuit, the battery mode bypass circuit being configured to receive auxiliary power input from the first battery and generate power output.

[0021] Technical Solution 7. The method according to Technical Solution 1, wherein the one or more first temperatures include the temperature of a charger circuit configured to charge the first battery.

[0022] Technical Solution 8. The method according to Technical Solution 1, wherein the first battery is connected to the first body via a power cable.

[0023] Technical Solution 9. The method according to Technical Solution 1, wherein using the first dataset to train the first computational model includes training the first computational model as a multinomial regression algorithm.

[0024] Technical Solution 10. The method according to Technical Solution 1 further includes: Access the second datasets, each of which includes a first voltage across the first battery, a first current flowing through the first battery, a first state of charge of the first battery, and a second temperature of the first battery; and The second dataset is used to train a second computational model to perform a function, which includes: Receives a second voltage across the second battery, a second current flowing through the second battery, and the fourth temperature; and The second charge state of the second battery is generated based on the second voltage, the second current, and the fourth temperature.

[0025] Technical Solution 11. The method according to Technical Solution 10, wherein training the second computational model using the second dataset includes training the second computational model using a multinomial regression algorithm.

[0026] Technical Solution 12. A non-transitory computer-readable medium storing instructions, which, when executed by a computing device, cause the computing device to perform the method according to any one of Technical Solutions 1-11.

[0027] Technical Solution 13. A computing device, comprising: One or more processors; and A computer-readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform the method according to any one of claims 1-11.

[0028] Technical Solution 14. A method comprising: Access datasets, each including a first voltage across a first battery of a first power supply device, a first current flowing through the first battery, a first state of charge of the first battery, and a first temperature of the first battery, wherein the first battery is located outside a first body of the first power supply device; and The dataset is used to train a computational model to perform a function, which includes: Receives a second voltage across the second battery of the second power supply device, a second current flowing through the second battery, and a second temperature of the second battery; and The second charge state of the second battery is generated based on the second voltage, the second current, and the second temperature.

[0029] Technical Solution 15. The method according to Technical Solution 14, wherein using the dataset to train the computational model includes using a multinomial regression algorithm to train the computational model.

[0030] Technical Solution 16. A non-transitory computer-readable medium storing instructions, which, when executed by a computing device, cause the computing device to perform the method according to any one of Technical Solutions 14-15.

[0031] Technical Solution 17. A computing device, comprising: One or more processors; and A computer-readable medium storing instructions that, when executed by one or more processors, cause the computing device to perform the method according to any one of claims 14-15.

[0032] Technical Solution 18. A method comprising: One or more first temperatures of the main body of the power supply device; and A second temperature of the battery of the power supply device is generated using a first calculation model and the one or more first temperatures, wherein the battery is located outside the body.

[0033] Technical Solution 19. The method according to Technical Solution 18, wherein the main body includes the control module of the power supply device.

[0034] Technical solution 20. The method according to technical solution 19, wherein the one or more first temperatures are detected within the control module.

[0035] Technical Solution 21. The method according to Technical Solution 18, wherein the one or more first temperatures include the temperature of the inner surface of the outer shell of the body.

[0036] Technical Solution 22. The method according to Technical Solution 18, wherein the one or more first temperatures include the temperature of a normal mode bypass circuit, the normal mode bypass circuit being configured to receive a main power input and generate a power output.

[0037] Technical Solution 23. The method according to Technical Solution 18, wherein the one or more first temperatures include the temperature of a battery mode bypass circuit, the battery mode bypass circuit being configured to receive auxiliary power input from the battery and generate power output.

[0038] Technical Solution 24. The method according to Technical Solution 18, wherein the one or more first temperatures include the temperature of a charger circuit configured to charge the battery.

[0039] Technical Solution 25. The method according to Technical Solution 18, wherein the battery is connected to the main body via a power cable.

[0040] Technical solution 26. The method according to technical solution 18, wherein a multinomial regression algorithm is used to train the first computational model.

[0041] Technical solution 27. The method according to technical solution 18 further includes: Receives the voltage across the battery, the current flowing through the battery, and the second temperature; and Using a second calculation model, the state of charge of the battery is generated based on the voltage, the current, and the second temperature.

[0042] Technical Solution 28. The method according to Technical Solution 27, wherein the second computational model is trained as a multinomial regression algorithm.

[0043] Technical Solution 29. A non-transitory computer-readable medium storing instructions, which, when executed by a computing device, cause the computing device to perform the method according to any one of Technical Solutions 18-28.

[0044] Technical solution 30. A computing device, comprising: One or more processors; and A computer-readable medium storing instructions that, when executed by one or more processors, cause the computing device to perform the method according to any one of claims 18-28. Attached Figure Description

[0045] The appended claims set forth the features of the illustrative examples that are considered novel. However, the illustrative examples, along with preferred modes of use, their further purposes and descriptions, will be best understood by reading in conjunction with the accompanying drawings and with reference to the following detailed description of the illustrative examples of this disclosure.

[0046] Figure 1 It is a block diagram of a computing device, a power supply device, and a battery, based on the example.

[0047] Figure 2 It is a block diagram of a computing device, a power supply device, and a battery, based on the example.

[0048] Figure 3 The example shows a training dataset that includes temperature, voltage, current, and charge state data.

[0049] Figure 4 It is a flowchart based on the example method.

[0050] Figure 5 It is a flowchart based on the example method.

[0051] Figure 6 It is a flowchart based on the example method.

[0052] Figure 7A The example illustrates the correlation between the temperature detected inside the casing of the power supply device and the temperature detected at the external battery of the power supply device.

[0053] Figure 7B The example illustrates the correlation between the temperature detected at the normal mode bypass circuit of the power supply device and the temperature detected at the external battery of the power supply device.

[0054] Figure 7C The example illustrates the correlation between the temperature detected at the battery mode bypass circuit of the power supply device and the temperature detected at the external battery of the power supply device.

[0055] Figure 7D The example illustrates the correlation between the temperature detected at the battery charger circuit of the power supply device and the temperature detected at the external battery of the power supply device.

[0056] Figure 8A The example illustrates the correlation between the temperature detected at different locations within the main body of the power supply device and the temperature detected at the external battery of the power supply device.

[0057] Figure 8B The example illustrates the correlation between the temperature detected at different locations within the main body of the power supply device and the temperature detected at the external battery of the power supply device.

[0058] Figure 9A The example illustrates the correlation between battery voltage and battery state of charge.

[0059] Figure 9B The example illustrates the correlation between battery current and battery state of charge.

[0060] Figure 9C The example illustrates the correlation between battery temperature and battery state of charge.

[0061] Figure 10A The example illustrates the correlation between battery voltage, current, and temperature and the battery's state of charge.

[0062] Figure 10B The example illustrates the correlation between battery voltage, current, and temperature and the battery's state of charge. Detailed Implementation

[0063] This disclosure provides a method for accurately estimating the temperature or state of charge of an external battery of a power supply device using information readily available during operation of the power supply device, and includes a method for training a first computational model to infer the temperature of the external battery of the power supply device based on one or more temperatures of the body (e.g., casing) of the power supply device. The method includes accessing a first dataset, each of which includes one or more first temperatures of a first body of the first power supply device and a second temperature of a first battery of the first power supply device. Typically, numerous experiments are conducted, in which the actual temperature of the battery is monitored over time, while one or more temperatures of the corresponding body are also monitored. For example, a first temperature sensor positioned to directly contact the inner or outer surface of the casing of the external battery can be used to detect the temperature of the battery over time, and one or more second temperature sensors can be mounted at different locations inside or outside the body of the power supply device. These datasets can be used to identify statistical correlations between the temperatures detected at the body and the temperatures detected at the battery. The method also includes using the first dataset to train a first computational model to perform a function during runtime. The function includes (e.g., during runtime operation of the second power supply device) receiving one or more third temperatures of a second body of the second power supply device and generating (e.g., inferring) a fourth temperature of a second (e.g., external) battery of the second power supply device based on the one or more third temperatures. As described below, during operation, the inferred temperature of the battery can be used to further infer the battery's state of charge.

[0064] The computing device can be further trained to infer the state of charge (SOC) of a battery, at least in part, based on the battery's inferred temperature. The SOC can be the ratio of the battery's currently stored charge to its nominal or actual full storage capacity. For example, the method includes accessing datasets, each comprising a first voltage, a first current flowing through the first battery, a first SOC of the first battery, and a first temperature of the first battery across a first power supply device. Typically, numerous experiments are conducted in which the actual output voltage, output current, SOC, and temperature of the battery are monitored over time during various operating conditions, such as charging or providing power to a load. These datasets can be used to identify statistical correlations between the battery's SOC and its voltage, current, and temperature. The method also includes using the datasets to train a computational model to perform a function. This function includes receiving a second voltage, a second current flowing through the second battery, and a second temperature of the second battery across a second external battery of the second power supply device during its runtime operation, and generating the SOC of the second battery based on the second voltage, second current, and second temperature. The foregoing method facilitates the estimation of battery temperature and battery SOC using data more readily available during the runtime operation of the power supply device.

[0065] The disclosed examples will now be described more fully below with reference to the accompanying drawings, which illustrate some, but not all, of the disclosed examples. In fact, several different examples may be described, and should not be construed as limited to the examples set forth herein. Rather, these examples are described so that this disclosure will be thorough and complete, and will fully convey the scope of this disclosure to those skilled in the art.

[0066] Figure 1 This is a block diagram of a computing device 12A, a power supply device 14A, and a battery 16A that can implement aspects of this disclosure. The computing device 12A and the power supply device 14A each include a communication interface 108, one or more processors 110, a computer-readable medium 112, and a user interface 116. The user interface 116, the communication interface 108, the processor 110, and the computer-readable medium 112 can be interconnected via a system bus, a network, or other connection mechanism 114.

[0067] The communication interface 108 can take various forms and is configured to allow the computing device 12A and the power supply device 14A to communicate with one or more other devices according to any number of protocols. In some examples, the communication interface 108 can take the form of a wired interface such as an Ethernet interface. Additionally or alternatively, the communication interface 108 can take the form of a wireless interface such as a cellular interface, a Bluetooth interface, or a Wi-Fi interface.

[0068] Processor 110 may include general-purpose processors (such as microprocessors or microcontrollers) and / or special-purpose processors (such as digital signal processors (DSPs)).

[0069] Computer-readable medium 112 may include one or more volatile, non-volatile, removable, and / or non-removable storage components (such as magnetic, optical, or flash storage devices) and may be integrated integrally or partially with processor 110. Furthermore, computer-readable medium 112 may store program instructions (e.g., compiled or uncompiled program logic and / or machine code) thereon, which, when executed by processor 110, cause computing device 12A or power supply device 14A to perform one or more functions, such as those described in this disclosure. In some examples, computer-readable medium 112 includes electrically erasable programmable read-only memory (EEPROM), non-volatile memory, and random access memory (RAM).

[0070] In some examples, computer-readable medium 112 stores instructions representing computational model 19A and computational model 19B. Computational model 19A and computational model 19B may take the form of, for example, artificial neural networks, machine learning algorithms, and / or multinomial regression algorithms.

[0071] User interface 116 is configured to facilitate interaction between computing device 12A or power supply device 14A and the user, for example by receiving input from the user and providing output to the user. Therefore, user interface 116 may include input components such as a keyboard, touchscreen, or buttons. Furthermore, user interface 116 may include output components such as a display screen, speaker, or other audio output mechanism.

[0072] The computing device 12A can take the form of a desktop computer, tablet computer, smartphone, or server. Other examples are possible.

[0073] The power supply unit 14A can take the form of an industrial power supply unit or a UPS system. For example, the power supply unit 14A can take the form of a SolaHD SDN10 24 100D, SDN20 24 100D or SDN40 24 100D power supply unit or a SolaHD uninterruptible power supply system SDU10 24B or SDU20 24B.

[0074] The power supply device 14A includes a normal mode bypass circuit 20A, a battery mode bypass circuit 20B, and a battery charger circuit 20C. The normal mode bypass circuit 20A is configured to receive a primary power input (e.g., from a utility company or power plant) and generate or provide regulated and / or controlled power output to the load. The battery mode bypass circuit 20B is configured to receive an auxiliary power input from the battery 16A (e.g., via power cable 15A) and generate regulated and / or controlled power output to the load. The battery charger circuit 20C is configured to receive the primary power input and charge the battery 16A via power cable 15A.

[0075] Temperature sensor 18B is positioned or mounted near or to the normal mode bypass circuit 20A. Temperature sensor 18C is positioned or mounted near or to the battery mode bypass circuit 20B. Temperature sensor 18D is positioned or mounted near or to the battery charger circuit 20C. Temperature sensor 18B may be mounted on the inner surface of the housing of body 13A near the normal mode bypass circuit 20A. Temperature sensor 18C may be mounted on the inner surface of the housing of body 13A near the battery mode bypass circuit 20B. Temperature sensor 18D may be mounted on the inner surface of the housing of body 13A near the battery charger circuit 20C. Other examples are possible.

[0076] Battery 16A typically takes the form of one or more electrochemical cells that store electrical charge. Battery 16A is connected to power supply device 14A via power cable 15A. Battery 16A is typically mounted externally to power supply device 14A or at another standardized location shortly from power supply device 14A. Temperature sensor 18A is positioned adjacent to or mounted to battery 16A.

[0077] Figure 1 The components shown can be used to train a computational model 19A to infer the temperature of the battery 16A based on one or more temperatures detected at the body 13A (e.g., housing or control module) of the power supply device 14A. Figure 1 The components shown can also be used to train computational model 19B to infer the state of charge of battery 16A based on the voltage, current and temperature of battery 16A.

[0078] In this disclosure, computational models 19A and 19B are generally described as being stored on computing device 12A; however, computational models 19A and 19B may also be stored on and executed on power supply device 14A. Therefore, actions described as being performed by computing device 12A may also generally be performed by power supply device 14A. Other examples are possible.

[0079] Figure 2 This is a block diagram of a computing device 12B, a power supply device 14B, and a battery 16B that can implement aspects of this disclosure. The computing device 12B, power supply device 14B, and battery 16B generally have all the structures and functions described above in the structure and functionality of the computing device 12A, power supply device 14A, and battery 16A. The battery 16B is connected to the power supply device 14B via a power cable 15B.

[0080] Figure 2 The components shown can be used in conjunction with calculation model 19A to infer the temperature of battery 16B based on one or more temperatures detected at the body 13B of power supply device 14B. Figure 2 The components shown can also be used in conjunction with the calculation model 19B to infer the state of charge of battery 16B based on the voltage, current and temperature of battery 16B.

[0081] Figure 3Datasets 202A, 202B, 202N, 204A, 204B, and 204N are shown. Although only six datasets are shown, typically N in datasets 202A-N and 204A-N can be greater than 100, greater than 1000, or even more, and therefore more datasets can be collected to train computational models 19A and 19B.

[0082] Dataset 202A includes time t A1 The collected temperature data, dataset 202B includes data at time t B1 The collected temperature data, and the dataset 202N includes data at time t. N1 Collected temperature data, etc. Dataset 202A includes temperature T detected by temperature sensor 18A. AA The temperature T detected by temperature sensor 18B AB The temperature T detected by temperature sensor 18C AC and the temperature T detected by temperature sensor 18D AD Dataset 202B includes temperatures T detected by temperature sensor 18A. BA The temperature T detected by temperature sensor 18B BB The temperature T detected by temperature sensor 18C BC and the temperature T detected by temperature sensor 18D BD Data set 202N includes temperatures T detected by temperature sensor 18A. NA The temperature T detected by temperature sensor 18B NB The temperature T detected by temperature sensor 18C NC and the temperature T detected by temperature sensor 18D ND Data sets 202A-N typically include data collected during various operating states of the power supply device 14A, such as when the power supply device 14A operates with the normal mode bypass circuit 20A enabled, when the power supply device 14A operates with the battery mode bypass circuit 20B enabled, or when the power supply device 14A operates with the battery charger circuit 20C enabled.

[0083] To train computational model 19A, computing device 12A accesses dataset 202A-N. As described above, in some examples, dataset 202A-N may include thousands of temperature data points collected by temperature sensors 18A-D during experiments. For example, the temperature may be a numerical value in degrees Celsius or Fahrenheit. Typically, dataset 202A indicates the temperature T. AA Temperature T AB Temperature T AC and temperature T ADAll in time t A1 The detected dataset 202B indicates temperature T. BA Temperature T BB Temperature T BC and temperature T BD All in time t B1 Being detected, etc.

[0084] Computing device 12A trains computational model 19A based on dataset 202A-N by adjusting the weights assigned to each node of computational model 19A. After analyzing each of datasets 202A-N, the weights are incrementally adjusted so that computational model 19A more accurately assigns a given set of temperatures detected by temperature sensors 18B-D to temperatures detected by temperature sensor 18A. In one example, computational model 19A is a multinomial regression algorithm. Therefore, computing device 12A can adjust the parameters of the following equation to achieve an improved or optimal fit to dataset 202A-N: Therefore, the trained computational model 19A is configured to receive the temperature detected by the temperature sensor 18B-D during the operation of the power supply device 14B as input, and to generate or infer the temperature of the battery 16B without using a temperature sensor that directly monitors the battery 16B. This is achieved by training the computational model 19B to use the temperature T as a set. AB Temperature T AC and temperature T AD Mapped to temperature T AA Temperature T will be used as a group BB Temperature T BC and temperature T BD Mapped to temperature T BA Wait until it's done.

[0085] Dataset 204A includes time t A2 The collected data, dataset 204B, includes data at time t. B2 The collected data, and dataset 204N includes data at time t. N2 The collected data, etc. Data set 204A includes the voltage V of a 16A battery. A The current I of the 16A battery A The state of charge Q of battery 16A A and the temperature T detected by temperature sensor 18A A Data sets 204A-N each indicate the time period during which their data were collected. Data set 204B includes the voltage V of battery 16A. B The current I of the 16A battery B The state of charge Q of battery 16A B and the temperature T detected by temperature sensor 18A BDataset 204N includes the voltage V of a 16A battery. N The current I of the 16A battery N The state of charge Q of battery 16A N and the temperature T detected by temperature sensor 18A N For example, voltage V can be a value in volts, current I can be a positive or negative value in amperes, state of charge Q can be a percentage or value in ampere-hours, and temperature can be a value in degrees Celsius or degrees Fahrenheit. By convention, current I will be positive when battery 16A is supplying power to a load and negative when it is being charged. Data sets 204A-N typically include data collected during various operating states of power supply device 14A, such as when power supply device 14A operates with normal mode bypass circuit 20A enabled, power supply device 14A operates with battery mode bypass circuit 20B enabled, or power supply device 14A operates with battery charger circuit 20C enabled. During training of computational model 19B, the state of charge Q of battery 16A can be determined using any known method.

[0086] To train the computational model 19B, the computing device 12A accesses voltage V. A Voltage V B Voltage V N Current I A Current I B Current I N Charge state Q A Charge state Q B Charge state Q C Temperature T A Temperature TB and temperature T N The dataset 204A-N. As mentioned above, in some examples, dataset 204A-N may include thousands of datasets collected during experiments by temperature sensors 18A-D, ammeters, or voltmeters.

[0087] Computing device 12A trains computing model 19B based on dataset 204A-N by adjusting the weights assigned to each node of computing model 19B. The weights are incrementally adjusted after each analysis of dataset 204A-N, allowing computing model 19B to more accurately assign a given set of battery voltage, battery current, and battery temperature to the state of charge. In one example, computing model 19B is a multinomial regression algorithm. Therefore, computing device 12A can adjust the parameters of the following equation to achieve an improved or optimal fit to dataset 204A-N: Therefore, the trained computational model 19B is configured to receive as input the voltage V of battery 16B detected by a voltmeter, the current I of battery 16B detected by an ammeter, and the temperature T of battery 16B inferred by the computational model 19A during the operation of the power supply device 14B, and to generate or infer the state of charge Q of battery 16B without using a sensor that directly monitors the temperature of battery 16B. This is achieved by training the computational model 19B to take the battery voltage V as a set... A Battery current I A and temperature T A Mapping to charge state Q A The battery voltage V will be used as a group. B Battery current I B and temperature T B Mapping to charge state Q B Wait until it's done.

[0088] Therefore, during runtime operation, computing device 12B receives the corresponding temperatures detected by temperature sensors 18B, 18C, and 18D as input, and uses computing model 19A to generate or infer the temperature of battery 16B as output. This output can be generated by user interface 116 and / or stored on computer-readable medium 112. This process can be repeated periodically and / or continuously.

[0089] Next, the computing device 12B receives the voltage V across the battery 16B detected by a voltmeter, the current I flowing through the battery 16B detected by an ammeter, and the temperature T generated as output by the computing model 19A as inputs, and generates or infers the state of charge Q of the battery 16B based on the voltage V, current I, and temperature T. This output can be generated by the user interface 116 and / or stored on the computer-readable medium 112.

[0090] Figure 4 , Figure 5 and Figure 6 These are block diagrams of methods 300, 400, and 500, respectively. In some examples, the methods are performed by computing device 12A and / or computing device 12B. Figure 4-6 As shown, methods 300, 400, and 500 include one or more operations, functions, or actions as illustrated by boxes 302, 304, 402, 404, 502, and 504. Although the boxes are illustrated in a sequential order, these boxes may also be executed in parallel, and / or in an order different from those described herein. Furthermore, depending on the desired implementation, the individual boxes may be combined into fewer boxes, divided into additional boxes, and / or removed.

[0091] At box 302, method 300 includes computing device 12A accessing datasets 202A-N, each of which includes a temperature T of the body 13A of power supply device 14A detected by temperature sensors 18B-D, and each includes a temperature of battery 16A detected by temperature sensor 18A. Battery 16A is located outside the body 13A. (See above reference) Figure 1-3 The functionality associated with box 302 is described.

[0092] At box 304, method 300 includes computing device 12A using dataset 202A-N to train computing model 19A to perform a function. The function includes receiving a temperature T of the body 13B of power supply device 14B detected by temperature sensors 18B-D, and generating or inferring a temperature of battery 16B of power supply device 14B based on the temperature T detected by temperature sensors 18B-D. Battery 16B is located outside the body 13B. (See above reference...) Figure 1-3 The functionality associated with box 304 is described.

[0093] At box 402, method 400 includes computing device 12A accessing datasets 204A-N, each of which includes the voltage V of battery 16A across power supply device 14A detected by a voltmeter, the current I flowing through battery 16A detected by an ammeter, the (e.g., known) state of charge Q of battery 16A, and the temperature T of battery 16A detected by temperature sensor 18A. (See above reference...) Figure 1-3 The functionality associated with box 402 is described.

[0094] At box 404, method 400 includes computing device 12A using dataset 204A-N to train computing model 19B to perform a function. The function includes receiving a voltage V across battery 16B of power supply device 14B detected by a voltmeter, a current I flowing through battery 16B detected by an ammeter, and a temperature T of battery 16B generated or inferred by computing model 19A. The function also includes generating or inferring a state of charge Q of battery 16B based on the voltage V across battery 16B, the current I flowing through battery 16B, and the temperature T of battery 16B generated or inferred by computing model 19A. (See above reference) Figure 1-3 The functionality associated with box 404 is described.

[0095] At box 502, method 500 includes computing device 12B receiving temperature T detected by temperature sensor 18B, temperature T detected by temperature sensor 18C, and temperature T detected by temperature sensor 18D of power supply device 14B from power supply device 14B. (Refer to above) Figure 1-3 The functionality associated with box 502 is described.

[0096] At box 504, method 500 includes computing device 12B using computing model 19A, using temperature T detected by temperature sensor 18B, temperature T detected by temperature sensor 18C, and temperature T detected by temperature sensor 18D of power supply device 14B, to generate or infer the temperature T of battery 16B of power supply device 14B. (See above reference) Figure 1-3 The functionality associated with box 502 is described.

[0097] Figure 7A The correlation between the temperature of the power supply device 14A detected by temperature sensor 18E and the temperature of the battery 16A detected by temperature sensor 18A is shown.

[0098] Figure 7B The correlation between the temperature of the power supply device 14A detected by temperature sensor 18B and the temperature of the battery 16A detected by temperature sensor 18A is shown.

[0099] Figure 7C The correlation between the temperature of the power supply device 14A detected by temperature sensor 18C and the temperature of the battery 16A detected by temperature sensor 18A is shown.

[0100] Figure 7D The correlation between the temperature of the power supply device 14A detected by temperature sensor 1DE and the temperature of the battery 16A detected by temperature sensor 18A is shown.

[0101] Figure 8A The correlation between the temperature detected at different locations within the main body 13A of the power supply device 14A and the temperature detected at the battery 16A is shown.

[0102] Figure 8B The correlation between the temperature detected at different locations within the main body 13A of the power supply device 14A and the temperature detected at the battery 16A is shown.

[0103] Figure 9A The correlation between the voltage of battery 16A and the state of charge of battery 16A is shown.

[0104] Figure 9B The correlation between the current of battery 16A and the state of charge of battery 16A is shown.

[0105] Figure 9C The correlation between the temperature of battery 16A and the state of charge of battery 16A is shown.

[0106] Figure 10AThe correlation between the voltage, current, and temperature of battery 16A and the state of charge of battery 16A is shown.

[0107] Figure 10B The correlation between the voltage, current, and temperature of battery 16A and the state of charge of battery 16A is shown.

[0108] Descriptions of various advantageous arrangements have been given for purposes of illustration and description and are not intended to be exhaustive or limited to examples taking the disclosed form. Many modifications and variations will be apparent to those skilled in the art. Furthermore, different advantageous examples may describe different advantages compared to other advantageous examples. One or more examples have been selected and described to explain the principles of the examples, their practical applications, and to enable others skilled in the art to understand the disclosure of various examples with various modifications, as well as to suit the particular intended use.

Claims

1. A method comprising: Access a first dataset, each of the first datasets including one or more first temperatures of a first body of a first power supply device and a second temperature of a first battery of the first power supply device, wherein the first battery is outside the first body; as well as The first dataset is used to train a first computational model to perform a function, which includes: One or more third temperatures are received from the second body of the second power supply device; as well as A fourth temperature is generated for the second battery of the second power supply device based on the one or more third temperatures, wherein the second battery is located outside the second body.

2. The method according to claim 1, wherein, The first main body includes the control module of the first power supply device.

3. The method according to claim 2, wherein, The control module detects one or more first temperatures.

4. The method according to claim 1, wherein, The one or more first temperatures include the temperature of the inner surface of the outer casing of the first body.

5. The method according to claim 1, wherein, The one or more first temperatures include the temperature of a normal mode bypass circuit, which is configured to receive a main power input and generate a power output.

6. The method according to claim 1, wherein, The one or more first temperatures include the temperature of a battery mode bypass circuit configured to receive auxiliary power input from the first battery and generate power output.

7. The method according to claim 1, wherein, The one or more first temperatures include the temperature of a charger circuit configured to charge the first battery.

8. The method according to claim 1, wherein, The first battery is connected to the first body via a power cable.

9. The method according to claim 1, wherein, Using the first dataset to train the first computational model includes training the first computational model as a multinomial regression algorithm.

10. The method of claim 1, further comprising: Access the second dataset, each of the second datasets including a first voltage across the first battery, a first current flowing through the first battery, a first state of charge of the first battery, and a second temperature of the first battery; as well as The second dataset is used to train a second computational model to perform a function, which includes: Receive a second voltage across the second battery, a second current flowing through the second battery, and the fourth temperature; as well as The second charge state of the second battery is generated based on the second voltage, the second current, and the fourth temperature.

Images