Method, apparatus, device and medium for estimating remaining available energy of power battery

Abstract

This application relates to a method, apparatus, computer device, and computer-readable storage medium for estimating the remaining usable energy of a power battery. The method includes: determining the mapping relationship between the power battery's output power spectrum and battery operating parameters based on historical battery operating data; determining the depth of discharge of the power battery under current driving habits based on current battery operating data and the mapping relationship; determining the remaining unusable energy corresponding to the current driving habits based on the depth of discharge; and determining the remaining usable energy of the power battery based on the current real-time state of the power battery and the remaining unusable energy. This method improves the accuracy of estimating the depth of discharge for different battery driving conditions, solves the problem of distortion in remaining usable energy caused by inaccurate charge state inference obtained from a two-dimensional table of charge state and remaining usable energy at low temperatures, and improves the accuracy of calculating the remaining usable energy of the power battery.

Description

Technical Field

[0001] This application relates to the field of battery energy calculation technology, and in particular to a method, apparatus, computer equipment, and storage medium for estimating the remaining usable energy of a power battery. Background Technology

[0002] With the development of new energy vehicles, range anxiety has become a major problem for drivers. As the mileage of pure electric vehicles on the market is generally misrepresented, the accuracy of mileage estimation is not very accurate. The main source of mileage estimation lies in the estimation of the remaining usable energy of the lithium battery.

[0003] Currently, the estimation of remaining usable energy in lithium batteries generally falls into two categories: The first category uses the battery's capacity as a benchmark, releasing different amounts of energy at different discharge rates as a reference. Under experimental conditions, a general mapping relationship between charge state and remaining usable energy is established to infer the remaining usable energy. However, this method is prone to distortion of the remaining usable energy at low temperatures or when the charge state estimation is inaccurate, potentially leading to vehicle breakdowns. The second category is based on a battery model, estimating the battery's remaining usable energy by estimating the depth of discharge. This method is superior to the first, as it can adapt to low-temperature environments. However, the actual depth of discharge is related to the vehicle's operating conditions, making it difficult to estimate the depth of discharge without prior knowledge of these conditions.

[0004] Therefore, the fundamental problem of the difficulty in predicting the remaining usable energy of a battery pack is that the battery's operating conditions are uncertain in complex or low-temperature environments. Consequently, the state of charge at the discharge cutoff of the battery pack also becomes uncertain and will change with operating conditions, temperature, and driving habits. Summary of the Invention

[0005] Based on this, it is necessary to provide a method, apparatus, computer device, computer-readable storage medium, and computer program product for estimating the remaining usable energy of a power battery by predicting driving conditions through machine learning and accurately obtaining the remaining usable energy of the battery based on the predicted driving conditions and the battery state at the current temperature.

[0006] In a first aspect, this application provides a method for estimating the remaining usable energy of a power battery, the method comprising:

[0007] Based on historical battery operation data, the mapping relationship between the output power spectrum of the power battery and the battery operation parameters is determined;

[0008] Based on the current battery operating data and the mapping relationship, the depth of discharge of the power battery under the current driving behavior is determined;

[0009] Based on the depth of discharge, determine the remaining unusable energy corresponding to the current driving habits;

[0010] The remaining usable energy of the power battery is determined based on the current real-time status of the power battery and the remaining unavailable energy.

[0011] In one embodiment, determining the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on historical battery operating data includes:

[0012] A machine learning model is trained based on the historical battery operating data, which includes temperature, voltage, current, and state of charge.

[0013] Based on the machine learning model, the mapping relationship between the output power spectrum of the power battery and the battery operating parameters is determined.

[0014] In one embodiment, training the machine learning model based on the historical battery operation data includes:

[0015] Based on the first time window, the corresponding data is obtained from the historical battery operation data as a training set;

[0016] The machine learning model is trained by using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs.

[0017] In one embodiment, training the machine learning model based on the historical battery operation data further includes:

[0018] Based on the second time window, the corresponding data is obtained from the historical battery operation data as a verification set;

[0019] Using the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs, the machine learning model is validated to obtain validation results;

[0020] If the verification result does not meet the set conditions, the first time window and the second time window are slid to obtain a new training set and a verification set, and the model training and model verification steps are repeated until the new verification result meets the set conditions.

[0021] In one embodiment, determining the depth of discharge of the power battery under the current driving behavior based on the current battery operating data and the mapping relationship includes:

[0022] The output power spectrum is determined based on the current battery operating data and the mapping relationship;

[0023] Based on the output power spectrum, the depth of discharge under the current driving behavior is determined.

[0024] In one embodiment, the battery operating parameters include at least one of the following: temperature, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage at the time of operation.

[0025] In one embodiment, determining the remaining unavailable energy corresponding to the current driving habit based on the depth of discharge includes:

[0026] A two-dimensional mathematical model of the charge state and battery energy of the battery operating parameters at different temperatures is established.

[0027] Based on the two-dimensional mathematical model, the remaining unusable energy corresponding to the discharge depth is determined.

[0028] Secondly, this application also provides a device for estimating the remaining usable energy of a power battery, the device comprising:

[0029] The first determining module is used to determine the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on historical battery operating data.

[0030] The second determining module is used to determine the depth of discharge of the power battery under the current driving behavior based on the current battery operating data and the mapping relationship.

[0031] The third determining module is used to determine the remaining unusable energy corresponding to the current driving habit based on the discharge depth;

[0032] The energy calculation module is used to determine the remaining usable energy of the power battery based on the current real-time status of the power battery and the remaining unavailable energy.

[0033] Thirdly, this application also provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the content of the first aspect above.

[0034] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the contents of the first aspect described above.

[0035] The aforementioned method, apparatus, computer equipment, and storage medium for estimating the remaining usable energy of a power battery determine the mapping relationship between the power battery's output power spectrum and battery operating parameters based on historical battery operating data; determine the depth of discharge of the power battery under the current driving habits based on current battery operating data and the mapping relationship; determine the remaining unusable energy corresponding to the current driving habits based on the depth of discharge; and determine the remaining usable energy of the power battery based on the current real-time state of the power battery and the remaining unusable energy. This improves the estimation accuracy of different battery discharge depths corresponding to different battery driving conditions, solves the problem of distortion of the remaining usable energy caused by inaccurate charge state inference obtained from a two-dimensional table of charge state and remaining usable energy at low temperatures, and improves the accuracy of calculating the remaining usable energy of the power battery. Attached Figure Description

[0036] Figure 1 This is an application environment diagram of a method for estimating the remaining usable energy of a power battery in one embodiment;

[0037] Figure 2 This is a flowchart illustrating a method for estimating the remaining usable energy of a power battery in one embodiment.

[0038] Figure 3 This is a mapping process for the mapping relationship between the output power spectrum of a power battery and the battery operating parameters in one embodiment;

[0039] Figure 4 This is a schematic diagram of the process of training a machine learning model in one embodiment;

[0040] Figure 5 This is a flowchart illustrating the process of validating a machine learning model in one embodiment;

[0041] Figure 6 This illustrates the relationship between the training set of the first time window and the validation set of the second time window of the power battery in an example embodiment, along the time axis.

[0042] Figure 7 This is a mapping calculation process for the remaining usable energy of a power battery in an example embodiment;

[0043] Figure 8 This is a structural block diagram of a device for estimating the remaining usable energy of a power battery in one embodiment;

[0044] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] The method for estimating the remaining usable energy of a power battery provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. Data storage system 106 can store the data that server 104 needs to process. Data storage system 106 can be integrated on server 104 or placed on the cloud or other network servers. On server 104, based on historical battery operation data, the mapping relationship between the output power spectrum of the power battery and the battery operation parameters is determined; based on current battery operation data and the mapping relationship, the depth of discharge of the power battery under the current driving behavior is determined, and based on the depth of discharge, the remaining unavailable energy under the current driving habit is determined. Finally, based on the current real-time status of the power battery and the remaining unavailable energy, the remaining available energy of the power battery is determined and displayed on terminal 102. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, smart TVs, smart vehicle devices, and portable wearable devices. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0047] In one embodiment, such as Figure 2 As shown, a method for estimating the remaining usable energy of a power battery is provided, and this method is applied to... Figure 1 Taking the application environment in [the document] as an example, the following steps are included:

[0048] S202, based on historical battery operation data, determines the mapping relationship between the output power spectrum of the power battery and the battery operation parameters.

[0049] The battery operating parameters include the temperature, state of charge, current, total voltage, battery pack power, maximum single-cell voltage, and minimum single-cell voltage at the operating time. The battery pack power is the product of the current and the total voltage.

[0050] Optionally, a time series window is established based on historical battery operation data, and a machine learning model is trained. The correlation between historical battery operation data is established through machine learning, and the mapping relationship between the power battery output power spectrum and battery operation parameters under this correlation is output.

[0051] The mapping process between the output power spectrum of the power battery and the battery operating parameters is as follows: Figure 3 As shown, after machine learning of attributes such as positive correlation, negative correlation, and no correlation, the battery operating data yielded the mapping relationship between battery operating parameters.

[0052] S204 determines the depth of discharge of the power battery under the current driving behavior based on the current battery operation data and mapping relationship.

[0053] Optionally, the current battery operating data can be used as input to the machine learning model to obtain the current power battery output power spectrum. Using the functional relationship between the output power spectrum and the state of charge and temperature at the current operating time, the minimum state of charge value that satisfies the output power, i.e., the depth of discharge, can be calculated.

[0054] S206, based on the depth of discharge, determines the remaining unavailable energy corresponding to the current driving habits.

[0055] Optionally, a two-dimensional mathematical model of the relationship between the state of charge and battery energy of battery operating parameters at different temperatures can be established to obtain the remaining unusable energy corresponding to the depth of discharge under the current driving habits.

[0056] S208 determines the remaining usable energy of the power battery based on the current real-time status of the power battery and the remaining unusable energy.

[0057] Optionally, the current real-time charge state of the power battery can be used as the input to a two-dimensional mathematical model of charge state and battery energy to obtain the current real-time energy of the power battery. The difference between the real-time energy and the remaining unusable energy is the remaining usable energy of the power battery.

[0058] In the above embodiments, the mapping relationship between the output power spectrum of the power battery and the battery operating parameters is determined based on historical battery operating data. Based on the current battery operating data and the mapping relationship, the depth of discharge of the power battery under the current driving behavior is determined. Based on the depth of discharge, the remaining unusable energy corresponding to the current driving behavior is determined. Finally, based on the current real-time state of the power battery and the remaining unusable energy, the remaining usable energy of the power battery is determined. This improves the estimation accuracy of different battery discharge depths corresponding to different battery driving conditions, solves the problem of distortion of the remaining usable energy of the battery caused by the inaccuracy of the charge state inferred from the two-dimensional table of charge state and remaining usable energy at low temperatures, and improves the accuracy of the calculation of the remaining usable energy of the power battery.

[0059] In one embodiment, such as Figure 4 As shown, the machine learning models trained based on historical battery operation data include:

[0060] S402, based on the first time window, obtains the corresponding data from historical battery operation data as the training set.

[0061] Optionally, a first time window is established with a window length of L. A specific operating moment is selected from historical battery operating data as the starting point, and a dataset with a backward time of L is used as the training set. The data obtained from the historical battery operating data in the first time window includes temperature, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage.

[0062] S404, using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs, trains the machine learning model.

[0063] Optionally, the machine learning model is trained by taking the temperature, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage of the training set as inputs and the state of charge as output.

[0064] In one embodiment, such as Figure 5 As shown, the machine learning model trained based on historical battery operation data also includes:

[0065] S502, based on the second time window, obtains corresponding data from historical battery operation data as a validation set.

[0066] Optionally, a second time window is established with a window length of S. A specific operating moment is selected from historical battery operating data as the starting point, and a dataset with an extension time of S is used as the validation set. The data obtained from the historical battery operating data within the second time window includes temperature, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage.

[0067] S504 uses the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs to validate the machine learning model and obtain the validation results.

[0068] Optionally, the temperature, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage of the validation set are used as inputs, and the state of charge is used as the output to validate the machine learning model and obtain the validation results.

[0069] S506, if the verification result does not meet the set conditions, slide the first time window and the second time window to obtain a new training set and a verification set, and repeat the model training and model verification steps until the new verification result meets the set conditions.

[0070] Optionally, if the verification result does not meet the set conditions, the first time window and the second time window are equidistantly slid to establish a new training set and a verification set. The system is then retrained on the new training set and verified on the new verification set. In this way, the battery operating parameters are continuously adjusted and the machine learning model is iterated until the new verification result meets the set conditions, thereby obtaining the mapping relationship between the output power spectrum of the power battery and the battery operating parameters.

[0071] Based on the above Figure 4 , Figure 5 In two embodiments, alternatively, historical battery operation data is cleaned to remove abnormal data.

[0072] In the two embodiments described above, by establishing a time window and using historical battery operation data as the training and validation sets, a machine learning model of the charge state of the power battery is obtained, which estimates the vehicle's driving habits. As the window slides and iterates during the machine learning process, the optimized model tracks and corrects driving habits in real time, thereby obtaining more accurate remaining available energy.

[0073] In one example embodiment, a method for estimating the remaining usable energy of a power battery in an electric vehicle is provided, comprising the following steps:

[0074] S601 collects historical battery operation data of electric vehicles from the BMS battery management system, including temperature T, state of charge (SOC), current I, total voltage V, battery pack power P, highest single-cell voltage VMAX, and lowest single-cell voltage VMIN corresponding to the operation time t.

[0075] S602, arrange historical battery operation data from oldest to most recent, select time i and backtrack for a time length of L to establish a first time window, where the first time window includes temperature T. i-L ~T i State of charge (SOC) i-L ~SOC i Current I i-L ~I i Total voltage V i-L ~V i Battery pack power P i-L ~P i The highest single-cell voltage VMAX i-L ~VMAX i Minimum single-cell voltage VMIN i-L ~VMIN i We obtain the historical battery operation data corresponding to the current time window as the training set.

[0076] S603 performs data cleaning and removes outliers from the historical battery operation data in the training set, followed by max-min standardization. An LSTM temporal neural network is then used to train the standardized training set, with charge state as the network output and temperature, total voltage, highest single-cell voltage, lowest single-cell voltage, and current as the network input. This establishes the correlation between various battery operation parameters within the first time window and outputs the mapping relationship between the power battery's output power spectrum and the battery operation parameters.

[0077] The maximum and minimum standardized formulas are as follows: x represents the value of the sequence, min represents the minimum value in the sequence data, and max represents the maximum value in the sequence data.

[0078] S604, Select time i and extend it by a time length of S to establish a second time window, where the second time window includes temperature T. i ~T i+S State of charge (SOC) i ~SOC i+S Current I i ~I i+S Total voltage V i ~V i+S Battery pack power P i ~P i+S The highest single-cell voltage VMAX i ~VMAX i+S Minimum single-cell voltage VMIN i ~VMIN i+S We obtain the historical battery operation data corresponding to the current second time window as the validation set.

[0079] The relationship between the training set in the first time window and the validation set in the second time window on the time axis is as follows: Figure 6 As shown.

[0080] S605. After cleaning and removing outliers from the historical battery operation data in the validation set, the data is subjected to maximum and minimum standardization. The state of charge is used as the network output, and temperature, total voltage, highest single-cell voltage, lowest single-cell voltage, and current are used as the network input. The correlation obtained from the training set in S603 is verified, and the mean square error of MSE is used to record and evaluate the verification error of the mapping relationship between the output power spectrum of the power battery and the battery operation parameters.

[0081] The formula for calculating the mean square error of MSE is as follows: m represents the size of the training set. Let y represent the predicted value of the i-th time series data in the training set, and let y represent the true value of the i-th time series data.

[0082] S606 involves sliding the first and second time windows equidistantly along the time axis to establish new training and validation sets. The system is then retrained, and the input and output parameters are continuously adjusted based on the validation error using the backpropagation algorithm inherent in the LSTM temporal neural network. This process iteratively optimizes the LSTM temporal neural network until it reaches its optimal state. The mapping relationship between the output power spectrum of the power battery and its operating parameters under optimal conditions is then obtained.

[0083] S607, based on power battery charging and discharging experiments, establishes a two-dimensional mathematical relationship model between battery energy SOE and charge state SOC at different temperatures.

[0084] S608 acquires the current battery operating data of the electric vehicle and uses an LSTM time-series neural network to predict the current depth of discharge (SOC). k Based on the two-dimensional mathematical relationship model between battery energy SOE and state of charge SOC at different temperatures, the current depth of discharge SOC is determined. k Remaining unavailable energy (SOE) under the corresponding current driving habits k .

[0085] S608, based on the current real-time state of the power battery (SOE0) and the remaining unavailable energy (SOE)... k The remaining usable energy of the power battery is SOE0-SOE k .

[0086] The mapping calculation process for the remaining usable energy of the power battery is as follows: Figure 7 As shown. The depth of discharge (SOC) is predicted based on the LSTM temporal neural network. k By mapping the functional relationships in the two-dimensional mathematical model, the corresponding remaining unusable energy (SOE) is obtained. k The current battery charge state SOC0 is mapped to the corresponding current real-time state energy SOE0 through the functional relationship in the two-dimensional mathematical model. SOE0 and SOE k The energy difference between them is the remaining usable energy under the current operating conditions.

[0087] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0088] Based on the same inventive concept, this application also provides a device for estimating the remaining usable energy of a power battery to implement the above-described method for estimating the remaining usable energy of a power battery. The solution provided by this device is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the device for estimating the remaining usable energy of a power battery provided below can be found in the limitations of the method for estimating the remaining usable energy of a power battery described above, and will not be repeated here.

[0089] In one embodiment, such as Figure 8 As shown, a device for estimating the remaining usable energy of a power battery is provided, comprising: a first determining module 82, a second determining module 84, a third determining module 86, and an energy calculation module 88, wherein:

[0090] The first determining module 82 is used to determine the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on historical battery operating data.

[0091] The second determining module 84 is used to determine the depth of discharge of the power battery under the current driving behavior based on the current battery operating data and mapping relationship;

[0092] The third determining module 86 is used to determine the remaining unavailable energy corresponding to the current driving habits based on the depth of discharge;

[0093] The energy calculation module 88 is used to determine the remaining usable energy of the power battery based on the current real-time status of the power battery and the remaining unavailable energy.

[0094] In one embodiment, the first determining module 82 includes:

[0095] A machine learning model is trained based on the historical battery operation data, which includes temperature, voltage, current, and state of charge.

[0096] Based on the machine learning model, the mapping relationship between the output power spectrum of the power battery and the battery operating parameters is determined.

[0097] In one embodiment, the first determining module 82 includes:

[0098] Based on the first time window, the corresponding data is obtained from historical battery operation data as the training set;

[0099] The machine learning model is trained by using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs.

[0100] In one embodiment, the first determining module 82 further includes:

[0101] Based on the second time window, corresponding data is obtained from historical battery operation data as a validation set;

[0102] Using the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs, the machine learning model is validated to obtain validation results.

[0103] If the validation results do not meet the set conditions, slide the first time window and the second time window to obtain a new training set and validation set, and repeat the model training and model validation steps until the new validation results meet the set conditions.

[0104] In one embodiment, the second determining module 84 includes:

[0105] The output power spectrum is determined based on the current battery operating data and mapping relationships;

[0106] Based on the output power spectrum, the depth of discharge under the current driving behavior is determined.

[0107] In one embodiment, the battery operating parameters include at least one of the following: temperature at the time of operation, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage.

[0108] In one embodiment, the third determining module 86 includes:

[0109] A two-dimensional mathematical model of the state of charge and battery energy under different temperatures was established.

[0110] Based on a two-dimensional mathematical model, the remaining unusable energy corresponding to the discharge depth is determined.

[0111] Each module in the aforementioned device for estimating the remaining energy of the power battery can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0112] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for estimating the remaining usable energy of a power battery. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0113] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0114] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0115] Step 1: Based on historical battery operation data, determine the mapping relationship between the output power spectrum of the power battery and the battery operation parameters;

[0116] Step 2: Based on the current battery operating data and mapping relationship, determine the depth of discharge of the power battery under the current driving behavior;

[0117] Step 3: Based on the depth of discharge, determine the remaining unavailable energy corresponding to the current driving habits;

[0118] Step 4: Determine the remaining usable energy of the power battery based on its current real-time status and remaining unusable energy.

[0119] In one embodiment, the processor further performs the following steps when executing the computer program:

[0120] A machine learning model is trained based on the historical battery operation data, which includes temperature, voltage, current, and state of charge.

[0121] Based on the machine learning model, the mapping relationship between the output power spectrum of the power battery and the battery operating parameters is determined.

[0122] In one embodiment, the processor further performs the following steps when executing the computer program:

[0123] Based on the first time window, the corresponding data is obtained from historical battery operation data as the training set;

[0124] The machine learning model is trained by using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs.

[0125] In one embodiment, the processor further performs the following steps when executing the computer program:

[0126] Based on the second time window, corresponding data is obtained from historical battery operation data as a validation set;

[0127] Using the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs, the machine learning model is validated to obtain validation results.

[0128] If the validation results do not meet the set conditions, slide the first time window and the second time window to obtain a new training set and validation set, and repeat the model training and model validation steps until the new validation results meet the set conditions.

[0129] In one embodiment, the processor further performs the following steps when executing the computer program:

[0130] The output power spectrum is determined based on the current battery operating data and mapping relationships;

[0131] Based on the output power spectrum, the depth of discharge under the current driving behavior is determined.

[0132] In one embodiment, the battery operating parameters include at least one of the following: temperature at the time of operation, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage.

[0133] In one embodiment, the processor further performs the following steps when executing the computer program:

[0134] A two-dimensional mathematical model of the state of charge and battery energy under different temperatures was established.

[0135] Based on a two-dimensional mathematical model, the remaining unusable energy corresponding to the discharge depth is determined.

[0136] Furthermore, in conjunction with the estimation methods for the remaining usable energy of the power battery provided in the above embodiments, this embodiment can also provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the estimation methods for the remaining usable energy of the power battery in the above embodiments.

[0137] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0138] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0139] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0140] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of estimating the remaining available energy of a power cell, characterized in that, The method includes: Determining the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on historical battery operating data includes: training a machine learning model based on the historical battery operating data, which includes temperature, voltage, current, and state of charge; and determining the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on the machine learning model. Based on the current battery operating data and the mapping relationship, the discharge depth of the power battery under the current driving behavior is determined, and the discharge depth is the minimum state of charge value that satisfies the output power. Based on the depth of discharge, determine the remaining unusable energy corresponding to the current driving habits; The remaining usable energy of the power battery is determined based on the current real-time status of the power battery and the remaining unavailable energy. The step of training a machine learning model based on the historical battery operation data includes: obtaining corresponding data from the historical battery operation data as a training set based on a first time window; and training the machine learning model by using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs. The process of training a machine learning model based on the historical battery operation data further includes: obtaining corresponding data from the historical battery operation data as a validation set based on a second time window; using the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs to validate the machine learning model and obtain a validation result; if the validation result does not meet the set conditions, sliding the first time window and the second time window to obtain a new training set and a validation set, and repeating the model training and model validation steps until the new validation result meets the set conditions.

2. The method of estimating the remaining available energy of a power cell according to claim 1, characterized in that, The process of determining the depth of discharge of the power battery under the current driving behavior based on the current battery operating data and the mapping relationship includes: The output power spectrum is determined based on the current battery operating data and the mapping relationship; Based on the output power spectrum, the depth of discharge under the current driving behavior is determined.

3. The method of claim 1, wherein, The battery operating parameters include at least one of the following at the time of operation: temperature, state of charge, current, total voltage, battery pack power, highest single-cell voltage, and lowest single-cell voltage.

4. The method of claim 1, wherein, Based on the depth of discharge, the remaining unavailable energy corresponding to the current driving habit is determined to include: A two-dimensional mathematical model of the charge state and battery energy of the battery operating parameters at different temperatures is established. Based on the two-dimensional mathematical model, the remaining unusable energy corresponding to the discharge depth is determined.

5. An apparatus for estimating the remaining available energy of a power cell, characterized in that The device includes: The first determining module is used to determine the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on historical battery operating data. This includes: training a machine learning model based on the historical battery operating data, which includes temperature, voltage, current, and state of charge; and determining the mapping relationship between the output power spectrum of the power battery and the battery operating parameters based on the machine learning model. The second determining module is used to determine the depth of discharge of the power battery under the current driving behavior based on the current battery operation data and the mapping relationship. The depth of discharge is the minimum state of charge value that satisfies the output power of the power battery. The third determining module is used to determine the remaining unusable energy corresponding to the current driving habit based on the discharge depth; An energy calculation module is used to determine the remaining usable energy of the power battery based on the current real-time status of the power battery and the remaining unusable energy. The step of training a machine learning model based on the historical battery operation data includes: obtaining corresponding data from the historical battery operation data as a training set based on a first time window; and training the machine learning model by using the temperature, voltage, and current of the training set as inputs and the charge state of the training set as outputs. The process of training a machine learning model based on the historical battery operation data further includes: obtaining corresponding data from the historical battery operation data as a validation set based on a second time window; using the temperature, voltage, and current of the validation set as inputs and the charge state of the validation set as outputs to validate the machine learning model and obtain a validation result; if the validation result does not meet the set conditions, sliding the first time window and the second time window to obtain a new training set and a validation set, and repeating the model training and model validation steps until the new validation result meets the set conditions. 6.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-5 when the computer program is executed by the processor. When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images