WO2021218423A1 - 用于充电管控的方法和装置 - Google Patents

用于充电管控的方法和装置 Download PDF

Info

Publication number
WO2021218423A1
WO2021218423A1 PCT/CN2021/080028 CN2021080028W WO2021218423A1 WO 2021218423 A1 WO2021218423 A1 WO 2021218423A1 CN 2021080028 W CN2021080028 W CN 2021080028W WO 2021218423 A1 WO2021218423 A1 WO 2021218423A1
Authority
WO
WIPO (PCT)
Prior art keywords
charging
samples
basic prediction
charging data
predicted
Prior art date
Application number
PCT/CN2021/080028
Other languages
English (en)
French (fr)
Inventor
郭文静
王妍
苗磊
卢信先
吴都明
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2021218423A1 publication Critical patent/WO2021218423A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/007—Regulation of charging or discharging current or voltage
    • H02J7/0071—Regulation of charging or discharging current or voltage with a programmable schedule
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J7/00—Circuit arrangements for charging or depolarising batteries or for supplying loads from batteries
    • H02J7/007—Regulation of charging or discharging current or voltage
    • H02J7/00712—Regulation of charging or discharging current or voltage the cycle being controlled or terminated in response to electric parameters
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02E60/10—Energy storage using batteries

Definitions

  • This application relates to the charging field, and more specifically to a method and device for charging management and control in the charging field.
  • Different users have different charging habits when charging terminal devices. Different charging habits will affect the life of the battery. For example, overcharging will reduce the life of the battery and reduce the endurance, thereby affecting the user experience.
  • the embodiments of the present application provide a method and device for charging management and control, which can manage and control the charging of a terminal device according to a predicted duration, which helps to extend the life of the battery and improve the battery life.
  • a method for charge management and control is provided.
  • the method can be executed by an electronic device, and the electronic device may be a device that can support the electronic device to implement the functions required by the method, such as a chip system.
  • the method includes: acquiring first charging data; inputting at least part of the charging data in the first charging data into a plurality of basic prediction models, and determining the first predicted charging duration corresponding to each basic prediction model; At least part of the charging data in the data is input into the weighting model to obtain multiple weighting coefficients; determining the second predicted charging time length according to the multiple weighting coefficients and the first predicted charging time length corresponding to each basic prediction model, the The second predicted charging duration is used to control the charging of the electronic device.
  • the electronic device obtains the first predicted charging time corresponding to each basic prediction model according to a plurality of basic prediction models, and the electronic device obtains a plurality of weight coefficients according to the weight model.
  • the electronic device determines the second predicted duration according to multiple weighting coefficients and the first predicted charging duration corresponding to each basic prediction model.
  • the electronic device controls the charging of the electronic device according to the second predicted charging duration, thereby prolonging battery life and improving The battery life can help improve the user experience.
  • At least two of the multiple weight coefficients are the same, or at least two of the multiple weight coefficients are different.
  • each basic prediction model may be different, or the charging data input to at least two basic prediction models may be the same.
  • charging data input to the weight model and the charging data input to multiple basic prediction models may be different.
  • different basic prediction models in the multiple basic prediction models correspond to different application scenarios.
  • determining the second predicted charging time length according to the multiple weighting coefficients and the first predicted charging time length corresponding to each basic prediction model includes: comparing the multiple weighting coefficients with the first predicted charging time length. Weighted calculation is performed on the first predicted charging duration corresponding to each basic prediction model to obtain the second predicted charging duration.
  • the method further includes: obtaining a first actual charging duration of the electronic device corresponding to the first charging data; combining the first charging data, the first actual charging duration, and the The second predicted charging duration is added as a sample to the first sample set; and the weight model is updated according to the samples in the first sample set.
  • the samples in the first sample set can be updated in real time, thereby ensuring the accuracy of the weight model.
  • the updating the weight model according to the samples in the first sample set includes: determining a first pass rate of the samples in the first sample set; if the first pass If the number of qualified samples in the first sample set is greater than the preset value of the first sample quantity, the number of qualified samples in the first sample set is determined according to the preset value of the first qualified rate. Part of the qualified samples amend the weight model.
  • the electronic device determines whether the samples in the first sample set do not meet the requirements of the first pass rate, it determines whether the qualified sample data in the first sample set meets the requirements of the preset value of the first sample quantity, if If the requirements are met, the weight model is modified according to some samples in the qualified samples, so that the accuracy of the modified weight model can be guaranteed.
  • the number of qualified samples used to modify the weight model described above needs to meet the preset number, so as to ensure the robustness requirements of the modified weight model.
  • the correcting the weight model according to a part of the qualified samples includes: training according to the partial sample to obtain a first correction parameter, and correcting the weight model according to the first correction parameter.
  • the weight model, the revised weight model is obtained.
  • the electronic device can determine the first correction parameter used to correct the first weight model, so as to ensure the accuracy of the weight model.
  • the correcting the weight model according to a part of the qualified samples includes: training according to the partial sample to obtain the first correction parameter, and adding the first correction parameter Send to the cloud;
  • the weight model is corrected according to the second correction parameter to obtain a corrected weight model.
  • each electronic device sends the first correction parameter obtained by itself to the cloud.
  • the first correction parameter is the correction parameter of the weight model.
  • There is no user information of the electronic device. Can also protect the privacy of users and help improve security.
  • the second correction parameter determined by the cloud using big data can meet the robustness requirements.
  • the method further includes: testing the stability of the modified weight model according to the remaining part of the qualified samples.
  • the inputting at least part of the charging data in the first charging data into multiple basic prediction models to obtain the first predicted charging duration corresponding to each basic prediction model includes:
  • the adjustment parameters of each basic prediction model are used to adjust the third predicted charging duration corresponding to each basic prediction model to obtain the first predicted charging duration corresponding to each basic prediction model.
  • the method further includes:
  • the adjustment parameter corresponding to the first basic prediction model is determined according to the samples in the second sample set.
  • the determining the adjustment parameter corresponding to the first basic prediction model according to the samples in the second sample set includes:
  • the adjustment parameters corresponding to the first basic prediction model are determined according to the qualified samples.
  • the adjustment parameter corresponding to the first basic prediction model is the regression coefficient of the linear relationship And constants.
  • part of the qualified samples can be used for non-linear training to obtain the coefficients of the non-linear relationship. It is the adjustment parameter of the first basic prediction model.
  • the first charging data and the second charging data include at least one of the following: the type of charger used to charge the electronic device, the manufacturer of the battery of the electronic device, The nominal capacity of the battery, the cell type of the battery, the type of charging cable used to charge the electronic device, the number of cycles that the battery has been charged, the nominal cycle that the battery can be charged The number of times, the average internal resistance of the battery, the maximum internal resistance of the battery, the historical plug-in time and historical unplug time of the charger, the charge cut-off time of the battery cell, the battery cell Starting power and ending power, the actual charging time of the preset time period of each day within the preset number of days.
  • a method for charging management and control including: acquiring first charging data; inputting at least part of the charging data in the first charging data into a first basic prediction model to obtain a third prediction duration; The third predicted charging duration is adjusted according to the adjustment parameter corresponding to the first basic prediction to obtain a first predicted charging duration, and the first predicted charging duration is used to control the charging of the electronic device.
  • the electronic device can use the adjustment parameters corresponding to the first basic prediction model to adjust the third predicted charging time length obtained by the first basic prediction model to obtain the first predicted charging time length for charging management and control.
  • the third predicted charging time length obtained by the first basic prediction model is not accurate, and the adjustment parameters can be used to adjust, so that the first predicted charging time length that may be accurate can be obtained, and the accuracy of charging control can also be improved.
  • the method further includes: acquiring second charging data of the electronic device and a second actual charging duration corresponding to the second charging data; and inputting the second charging data to the In the first basic prediction model, the fourth predicted charging time length is obtained; the fourth predicted charging time length, the second charging data, and the second actual charging time length are added as samples to the second sample set; The samples in the second sample set determine the adjustment parameters corresponding to the first basic prediction model.
  • the adjustment parameters corresponding to the first basic prediction model in the above solution are obtained based on multiple actual samples, and therefore, can meet actual adjustment requirements.
  • the determining the adjustment parameter corresponding to the first basic prediction model according to the samples in the second sample set includes:
  • the adjustment parameters corresponding to the first basic prediction model are determined according to the qualified samples.
  • the adjustment parameter corresponding to the first basic prediction model is the regression coefficient of the linear relationship And constants.
  • part of the qualified samples can be used for non-linear training to obtain the coefficients of the non-linear relationship. It is the adjustment parameter of the first basic prediction model.
  • the first charging data and the second charging data include at least one of the following: the type of charger used to charge the electronic device, the manufacturer of the battery of the electronic device, The nominal capacity of the battery, the cell type of the battery, the type of charging cable used to charge the electronic device, the number of cycles that the battery has been charged, the nominal cycle that the battery can be charged The number of times, the average internal resistance of the battery, the maximum internal resistance of the battery, the historical plug-in time and historical unplug time of the charger, the charge cut-off time of the battery cell, the battery cell Starting power and ending power, the actual charging time of the preset time period of each day within the preset number of days.
  • a device for charging management and control is provided, and the device is used to execute the foregoing first aspect or the method in any possible implementation manner of the first aspect.
  • the apparatus may include a module for executing the first aspect or the method in any possible implementation manner of the first aspect.
  • a device for charging management and control is provided, and the device is used to execute the above-mentioned second aspect or any possible implementation of the second aspect.
  • the device may include a module for executing the second aspect or the method in any possible implementation manner of the second aspect.
  • a device for charging management and control includes a processor, the processor is coupled with a memory, the memory is used to store computer programs or instructions, and the processor is used to execute the computer programs or instructions stored in the memory, so that the first The method in one aspect is executed.
  • the processor is configured to execute a computer program or instruction stored in the memory, so that the apparatus executes the method in the first aspect.
  • the device includes one or more processors.
  • the device may further include a memory coupled with the processor.
  • the device may include one or more memories.
  • the memory can be integrated with the processor or provided separately.
  • the device may also include a transceiver.
  • a device for charging management and control includes a processor, the processor is coupled with a memory, the memory is used to store computer programs or instructions, and the processor is used to execute the computer programs or instructions stored in the memory, so that the first The method in the two aspects is executed.
  • the processor is used to execute a computer program or instruction stored in the memory, so that the device executes the method in the second aspect.
  • the device includes one or more processors.
  • the device may further include a memory coupled with the processor.
  • the device may include one or more memories.
  • the memory can be integrated with the processor or provided separately.
  • the device may also include a transceiver.
  • a computer-readable storage medium on which a computer program (also referred to as an instruction or code) for implementing the method in the first aspect is stored.
  • the computer when the computer program is executed by a computer, the computer can execute the method in the first aspect.
  • a computer-readable storage medium is provided, and a computer program (also referred to as an instruction or code) for implementing the method in the first aspect or the second aspect is stored thereon.
  • the computer when the computer program is executed by a computer, the computer can execute the method in the second aspect.
  • the present application provides a chip including a processor.
  • the processor is used to read and execute the computer program stored in the memory to execute the method in the first aspect and any possible implementation manners thereof.
  • the chip further includes a memory, and the memory and the processor are connected to the memory through a circuit or a wire.
  • the chip further includes a communication interface.
  • the present application provides a chip system including a processor.
  • the processor is used to read and execute the computer program stored in the memory to execute the method in the second aspect and any possible implementation manners thereof.
  • the chip further includes a memory, and the memory and the processor are connected to the memory through a circuit or a wire.
  • the chip further includes a communication interface.
  • the present application provides a computer program product.
  • the computer program product includes a computer program (also referred to as an instruction or code).
  • the computer program When the computer program is executed by a computer, the computer realizes the method.
  • the present application provides a computer program product.
  • the computer program product includes a computer program (also referred to as an instruction or code).
  • the computer program When the computer program is executed by a computer, the computer realizes the method.
  • Fig. 1 is a schematic diagram of a management and control strategy provided by an embodiment of the present application.
  • Fig. 2 is a schematic diagram of a system architecture provided by an embodiment of the present application.
  • Fig. 3 is a schematic diagram of a system architecture provided by an embodiment of the present application.
  • Fig. 4 is a schematic diagram of training a basic prediction model provided by an embodiment of the present application.
  • Fig. 5 is a schematic diagram of a training weight model provided by an embodiment of the present application.
  • FIG. 6 is a schematic diagram of a method for charging management and control provided by an embodiment of the present application.
  • Fig. 7 is a schematic diagram of a modified weight model provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram of another modified weight model provided by an embodiment of the present application.
  • FIG. 9 is a schematic diagram of another method for charging management and control provided by an embodiment of the present application.
  • FIG. 10 is a schematic diagram of a method for obtaining adjustment parameters corresponding to a first basic prediction model provided by an embodiment of the present application.
  • FIG. 11 is a schematic diagram of effects provided by an embodiment of the present application.
  • FIG. 12 is a schematic diagram of a management and control strategy provided by an embodiment of the present application.
  • FIG. 13 is a schematic block diagram of a device for charging management and control provided by an embodiment of the present application.
  • FIG. 14 is a schematic block diagram of another device for charging management and control provided by an embodiment of the present application.
  • FIG. 15 is a schematic block diagram of another device for charging management and control provided by an embodiment of the present application.
  • Different users have different charging habits when charging electronic devices. Different charging habits will affect the battery life of electronic devices. For example, some users like to plug in the charger the night before and start charging until the charger is unplugged the next morning, which will cause the battery to be overcharged. For example, some users like to recharge during working hours during the day and unplug it when fully charged, so that the battery life will not be affected. For example, some users like to charge the same electronic device with different power chargers. Specifically, some users like to use low-power chargers to charge electronic devices that require high-power. This will make the charging time too long. Of users like to use high-power chargers to charge electronic devices that require low-power. This will make the battery quickly reach a saturated state. If the charger is not unplugged in time, it will cause overcharging. For example, some users like to charge while using electronic devices. The battery needs to be charged and discharged continuously, which shortens the service life of the battery.
  • t3 in Figure 1 is the predicted charging time according to user habits.
  • the electronic device can use t3 to determine that the control strategy in Figure 1 can be adopted: charging normally during the period from 0 to t1, and reaching 70% of the time.
  • the time period from t1 to t2 enters the protection state, and the electronic device is not charged, or the electronic device is charged with a small current, and the charging continues after t2 until t3 reaches 100%.
  • Different control strategies t1, t2, and t3 The value of is different.
  • the management and control strategy in Fig. 1 is only an exemplary description, and should not cause any limitation on this application.
  • the use of the predicted charging duration to manage and control the charging of the electronic device in the embodiment of the present application may also be another management and control strategy, which is not limited in the embodiment of the present application.
  • the method for charging management and control provided in the embodiments of the present application can be used to manage and control any electronic device that needs to be charged.
  • the charging of a terminal device can be managed and controlled.
  • the terminal equipment mentioned in the embodiments of this application is user equipment (UE), mobile station (MS), mobile terminal (MT), access terminal, user unit, user station, mobile station, Mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.
  • UE user equipment
  • MS mobile station
  • MT mobile terminal
  • access terminal user unit, user station, mobile station, Mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.
  • the terminal device may be a device that provides voice/data connectivity to the user, for example, a handheld device with a wireless connection function, a vehicle-mounted device, and so on.
  • a handheld device with a wireless connection function for example, a vehicle-mounted device, and so on.
  • some examples of terminals are: mobile phones (mobile phones), tablet computers, notebook computers, handheld computers, mobile internet devices (MID), wearable devices, virtual reality (VR) devices, and augmented reality.
  • Wireless terminals in transportation safety transportation safety
  • wireless terminals in smart city smart city
  • wireless terminals in smart home smart home
  • cellular phones cordless phones
  • session initiation protocol SIP
  • wireless local loop wireless local loop
  • WLL wireless local loop
  • PDA personal digital assistant
  • handheld device with wireless communication function computing device or other processing device connected to wireless modem
  • vehicle Devices wearable devices
  • terminal devices in a 5G network or terminal devices in a public land mobile network (PLMN) that will evolve in the future, etc., which are not limited in the embodiment of the present application.
  • PLMN public land mobile network
  • the terminal device may also be a wearable device.
  • Wearable devices can also be called wearable smart devices. It is a general term for using wearable technology to intelligently design everyday wear and develop wearable devices, such as glasses, gloves, watches, clothing and shoes.
  • a wearable device is a portable device that is directly worn on the body or integrated into the user's clothes or accessories. Wearable devices are not only a kind of hardware device, but also realize powerful functions through software support, data interaction, and cloud interaction.
  • wearable smart devices include full-featured, large-sized, complete or partial functions that can be achieved without relying on smart phones, such as smart watches or smart glasses, and only focus on a certain type of application function, and need to cooperate with other devices such as smart phones.
  • the terminal device may also be a terminal device in the Internet of Things (IoT) system.
  • IoT Internet of Things
  • the terminal device of the present application may also be a vehicle-mounted module, vehicle-mounted module, vehicle-mounted component, vehicle-mounted chip, or vehicle-mounted unit built into a vehicle as one or more components or units. The vehicle passes through the built-in vehicle-mounted module, vehicle-mounted module, An on-board component, on-board chip, or on-board unit can implement the method of the present application.
  • V2X vehicle to everything
  • LTE-V long term evolution-vehicle
  • V2V vehicle-to-vehicle
  • V2V vehicle-to-vehicle
  • the following describes a schematic diagram of a system architecture provided by an embodiment of the present application in conjunction with FIG. 2.
  • multiple electronic devices are charged at the end-side.
  • the charging data of multiple charges is used for artificial intelligence (AI) model prediction, and the predicted data is used for intelligent management and control of the battery.
  • the charging time and the real charging time are updated on the end-side model.
  • the modified parameters of the model obtained by multiple end-sides can be sent to the cloud, and the cloud is aggregated and then sent back to each end-side, and each end-side uses the aggregated correction parameters to modify the model.
  • the modified parameters sent by multiple end-sides do not contain any user information, and the privacy of the user can also be protected, which is beneficial to improve security.
  • the cloud uses big data to fit and modify the parameters, which can meet the robustness requirements.
  • the embodiment of the present application includes: determining the model, and using the determined model to obtain the predicted charging time length of the electronic device.
  • the predicted charging time length module in FIG. 3 may be the AI model prediction model in FIG.
  • the charging management and control module of 3 can be the battery intelligent management and control module of FIG.
  • FIG. 3 may not include a correction process, where the determination model includes a determination weight model and multiple basic prediction models.
  • the weight model may be preset or trained or indicated by the cloud.
  • multiple basic prediction models may also be preset or trained or instructed by the cloud, which is not limited in the embodiment of the present application.
  • the electronic device and the cloud can train according to the steps shown in Figure 4 to obtain multiple basic prediction models.
  • the training shown in Figure 4 The process can be executed by an electronic device or in the cloud.
  • the following describes only a basic prediction model (first basic prediction model) obtained by training as an example.
  • the method 300 includes:
  • an initial basic prediction model may be preset, and the model parameters of the initial basic prediction model are preset initial values.
  • the samples in the training sample set include: multiple sets of charging data, the actual charging duration corresponding to each set of charging data, and multiple predicted charging durations obtained by inputting each group of charging data into the initial basic prediction model.
  • a set of charging data can be a piece of historical charging data of an electronic device. Inputting a specific piece of historical charging data into the initial basic prediction model can get a predicted charging time, and a piece of specific historical charging data corresponds to the actual situation of the electronic device. Charging time.
  • one sample in the training sample set is a set of charging data, the predicted charging duration obtained by inputting the charging data into the initial basic prediction model, and the actual charging duration corresponding to the set of charging data.
  • One piece of historical charging data may include at least one of the following charging parameters: the type of charger used to charge the electronic device, the manufacturer of the battery that produces the electronic device, the nominal capacity of the battery, the type of battery cell, and the type of battery used to charge the electronic device.
  • the type of charging cable used to charge the electronic device the number of times the battery has been charged, the nominal number of times the battery can be charged, the average internal resistance of the battery, the maximum internal resistance of the battery, the historical plug-in time and historical unplug time of the charger, The charge cut-off time of the battery cell, the historical starting power and the historical ending power of the battery cell, the actual charging time in a specific time period of each day within a preset number of days, and so on.
  • a set of charging data may include other charging parameters, which are not limited in the embodiment of the present application.
  • the samples in the training sample set may be samples marked with a positive sample label. Specifically, if the absolute value of the difference between a set of charging data input in the initial basic prediction model and the output of the predicted charging duration and the actual charging duration under the set of charging data is less than the preset value, then the sample is a positive sample and can be used as training The sample in the sample set.
  • the predicted charging time corresponding to the samples in the training sample set is greater than the preset value, so as to prevent the samples from being generated by the user's charging habit of unplugging immediately after charging.
  • the samples generated from the charging data affect the accuracy of the basic prediction model training.
  • S320 Determine whether the number of samples in the training sample set reaches the preset number of training samples. If the number of samples in the training sample set reaches the preset number of training samples, execute S330; otherwise, execute S310 to continue adding samples to the training sample set.
  • the preset number of training samples may be a fixed value specified in the agreement or a variable value, which is not limited in the embodiment of the present application.
  • S330 includes: the initial basic prediction model can be input according to the samples in the sample set, the initial basic prediction model outputs a predicted charging time, the actual charging time is compared with the predicted charging time, and the coefficients of the initial basic prediction model are repeatedly corrected according to the comparison result.
  • the revised basic prediction model is the first basic prediction model.
  • Different basic prediction models can be for different charging mode application scenarios. For example, some basic prediction models are for night charging mode, some are for daytime charging mode, and some are for workday charging mode.
  • the basic predictive model is for the charging mode on rest days.
  • the input of the basic prediction model may be different or the same, which is not limited in the embodiment of the present application.
  • the basic prediction model can be a support vector machine (SVM) model, a decision tree (decision tree) model, a neural network (neural network) model, a bagging model or a boosting model.
  • SVM support vector machine
  • the weight model may also be other models, which are not limited in the embodiment of the present application.
  • the method 400 shown in the specific map 5 includes:
  • an initial weight model may be preset, and the model parameters of the initial weight model are preset initial values.
  • the samples in the training sample set include: multiple sets of charging data, the actual charging duration corresponding to each set of charging data, and multiple predicted charging durations obtained by inputting each set of charging data into the initial weight model and multiple basic prediction models (see Method 500 for details) description of).
  • a set of charging data can be a piece of historical charging data of an electronic device. Input a specific historical charging data into the initial weight model to obtain multiple initial weight coefficients, and input a specific historical charging data into multiple basic prediction models to obtain multiple initial weight coefficients. Multiple predicted charging time lengths are weighted by multiple initial weight coefficients and multiple predicted charging time lengths to obtain a final predicted charging time length.
  • the electronic device corresponds to an actual charging time length under the condition of a specific historical charging data.
  • a sample in the training sample set is a set of charging data, a final predicted charging duration obtained by inputting the charging data to the initial weight model and multiple basic prediction models, and the actual charging duration corresponding to the set of charging data.
  • the samples in the training sample set may be samples marked with a positive sample label. Specifically, if the absolute value of the difference between a predicted charging duration obtained from a set of charging data input into the initial weight model and multiple basic prediction models and the actual charging duration under the set of charging data is less than the preset value, the sample It is a positive sample and can be used as a sample in the training sample set.
  • the predicted charging time corresponding to the samples in the training sample set is greater than the preset value, so as to prevent the samples from being generated by the user's charging habit of unplugging immediately after charging.
  • the samples generated from the charging data affect the accuracy of the basic prediction model training.
  • S420 Determine whether the number of samples in the training sample set reaches the preset number of training samples. If the number of samples in the training sample set reaches the preset number of training samples, execute S430; otherwise, execute S410 to continue adding samples to the training sample set.
  • S430 includes: inputting a set of charging data corresponding to the samples in the training sample set into an initial weighting model, the initial weighting model outputting multiple weighting coefficients, and inputting the set of charging data into multiple training methods 300 Based on the multiple predicted charging durations output by the basic prediction model, the final predicted charging duration obtained by weighting multiple weighting coefficients and multiple predicted charging durations is compared with the actual charging duration, and the coefficients of the initial weighting model are repeatedly corrected according to the comparison results until A plurality of weighting coefficients are weighted with the predicted charging duration output by the basic prediction model trained in the method 300 to obtain the final predicted charging duration and the actual charging duration.
  • the accuracy is less than the preset value for a long time and lasts for a certain period of time, and then the weighting model is obtained.
  • the weight model obtained in the above training process may be a support vector machine (SVM) model, a decision tree (decision tree) model, a neural network model, a bagging model or a boosting model.
  • SVM support vector machine
  • decision tree decision tree
  • neural network model e.g., a bagging model
  • bagging model e.g., a bagging model
  • boosting model e.g., a bagging model
  • each set of charging data may be charging data before preprocessing, or charging data after preprocessing. If each set of charging data is the charging data before preprocessing, in the process of S330 and S430 training model, the charging data needs to be preprocessed.
  • the preprocessing includes denoising points, quantization processing, normalization processing and other operations. At least one standardized treatment.
  • the charging data may not be preprocessed, and it is possible that the charging data itself meets the requirements of the training model, which is not limited in the embodiment of the present application.
  • the predicted charging time corresponding to each set of charging data in the training sample set in the method 300 and the method 400 needs to be greater than the preset time.
  • the effective samples in the training sample set are at least sustained by the user
  • the charging behavior with a preset duration prevents the samples in the training sample set from being samples generated by the charging habit that the user unplugs immediately after charging and powering on, which affects the accuracy of the trained model.
  • the number of samples in the training sample set in the method 300 and the method 400 may be the same or different, which is not limited in the embodiment of the present application.
  • the preset number of training samples in the method 300 and the method 400 may be the same or different, which is not limited in the embodiment of the present application.
  • the charging parameters included in the samples in the training sample set in the method 300 and the method 400 may be the same or different, which is not limited in the embodiment of the present application.
  • the basic prediction model required by the method 400 for training the weight model may be obtained according to the method 300, or may be multiple preset basic predictions.
  • the model may also be multiple basic prediction models indicated by the cloud, which is not limited in the embodiment of the present application.
  • the following describes the multiple basic prediction models obtained by the method 300 and the weight model obtained by the method 400 with reference to FIG. 6 to describe the method 500 for charging management and control, and the method 500 includes:
  • S510 The electronic device obtains first charging data.
  • the first charging data may be a set of charging data in the method 300 and the method 400.
  • the first charging data can be discussed in the following situations.
  • Case 1 The electronic device collects original charging data, and the original charging data meets the input of the weight model of S520 and the input of multiple basic prediction models of S530.
  • the first charging data is the collected original charging data.
  • the electronic device saves the collected raw charging data.
  • Case 2 The electronic device collects the original charging data, and the original charging data does not meet the input of the weight model of S520 and the input of multiple basic prediction models of S530.
  • the first charging data is the pre-processing of the original charging data. Charging data after processing.
  • the electronic device saves the pre-processed charging data.
  • the electronic device collects the original charging data.
  • the original charging data does not meet the input of the weight model of S520 and the input of multiple basic predictive models of S530.
  • the electronic device can save the collected original charging data.
  • the device can preprocess the collected original charging data, and the preprocessed charging data obtained is the first charging data.
  • the electronic device may save the first charging data.
  • the electronic device can only save the collected raw charging data, and when necessary, preprocess the collected raw charging data to obtain the first charging data, or the electronic device can save both the collected raw data and the first charging data.
  • the first charging data obtained by preprocessing can be saved.
  • the preprocessing of the original charging data includes at least one standardization process among operations such as denoising points, quantization processing, and normalization processing.
  • the first charging data includes the historical plug-in time and the historical plug-in time of the charger, the difference between the historical plug-in time and the historical plug-in time of the charger is greater than the preset duration, in other words,
  • the first charging data is the charging data generated by the user's actual charging behavior that lasts for at least a preset period of time, which prevents the first charging data from being the charging data generated by the user's charging habit of unplugging immediately after charging and powering on, which affects the accuracy of the prediction.
  • S520 The electronic device inputs at least part of the charging data in the first charging data into a plurality of basic prediction models, and determines a first predicted charging duration corresponding to each basic prediction model.
  • S530 includes: inputting at least part of the charging data in the first charging data into multiple basic prediction models to obtain multiple third predicted charging durations, and each basic prediction model can output a third predicted charging duration There is a corresponding adjustment parameter for each basic prediction model, and the third prediction charging time length output by each basic prediction model is adjusted using the adjustment parameters corresponding to each basic prediction model, so as to obtain the first prediction charging time length.
  • the third predicted charging duration obtained by each basic prediction model may be the same or different.
  • the first predicted charging time length output by each basic prediction model is the predicted charging time length adjusted according to the adjustment parameters, which can make the prediction more accurate.
  • S530 The electronic device inputs at least part of the charging data in the first charging data into a weighting model to obtain multiple weighting coefficients.
  • At least two of the multiple weighting coefficients are the same, or at least two of the multiple weighting coefficients are different.
  • the method 900 will describe how to obtain the corresponding adjustment parameters in the multiple basic prediction models. In order to avoid repetition, it will not be described in detail here.
  • At least part of the charging data input to the weight model in S530 may be the same or different from at least part of the charging data input to the multiple basic prediction models in S230, which is not limited in the embodiment of the present application.
  • each basic prediction model in S520 may be the same or different, which is not limited in the embodiment of the present application.
  • S520 and S530 there is no restriction on the order of S520 and S530, and S520 can be performed before or after S530 or at the same time, which is not limited in the embodiment of the present application.
  • the electronic device determines a second predicted charging duration according to the multiple weight coefficients and the first predicted charging duration corresponding to each basic prediction model, and the second predicted charging duration is used to control the charging of the electronic device.
  • the second predicted charging duration may be t3-0 as shown in FIG. 1, so that the electronic device can be controlled according to the management and control strategy as shown in FIG. 1.
  • S540 includes: weighting the multiple weight coefficients with the first predicted charging duration corresponding to each basic prediction model to obtain the second predicted charging duration.
  • the L weight coefficients output by S530 are respectively ⁇ 1 , ⁇ 2 , ..., ⁇ L
  • the L first predicted charging durations output by the L basic prediction models in S520 are K 1 , K 2 , ..., K L
  • the second predicted charging duration is ⁇ 1 K 1 + ⁇ 2 K 2 +...+ ⁇ L K L , where L is the preset value.
  • the electronic device obtains multiple weight coefficients according to the weight model.
  • the electronic device obtains the first predicted charging duration corresponding to each basic prediction model according to multiple basic prediction models, and determines the second predicted duration according to the multiple weight coefficients and the first predicted charging duration corresponding to each basic prediction model.
  • the second predicted charging duration controls the charging of the electronic device, thereby prolonging the life of the battery and improving the endurance of the battery, which helps to improve the user experience.
  • the weight model in the above method 500 may be preset.
  • the weight model may be preset in the processor of the electronic device when the electronic device is shipped from the factory.
  • the weight model in the above method 500 may be obtained by training according to the method 400. Regardless of whether the above-mentioned weight model is obtained through the training of the electronic device or preset, the weight model can be modified in the process of predicting the charging time.
  • the method 500 may be executed multiple times, input at least part of the multiple sets of charging data into the multiple basic prediction models obtained by the method 300 and the weight model obtained by the method 400, and output multiple second predicted charging durations. Multiple samples, each sample includes a set of charging data, the second predicted charging duration corresponding to the set of charging data, and the actual charging duration corresponding to the set of charging. The following describes the process of modifying the weight model in two cases.
  • Case 1 The electronic device modifies the weight model according to the samples in the first sample set. As shown in Figure 7, Case 1 specifically includes:
  • S610 The electronic device determines a qualified sample among the multiple samples output by the method 500.
  • the absolute value of the difference between the first actual charging duration and the second predicted charging duration in a sample is less than the preset duration. For example, if the preset duration is 60 minutes, the sample is a qualified sample, also called Positive samples, otherwise unqualified samples, also called negative samples.
  • the second predicted charging duration of the qualified samples must be greater than the preset duration,
  • the qualified samples are the actual charging behavior of the user that lasts for at least the preset duration, and avoid some samples that are generated by the user's charging habit of unplugging immediately after charging and powering on, which affects the accuracy of the modified weight model.
  • S620 The electronic device determines whether the first pass rate under the current weight model of the sample is greater than or equal to the first pass rate preset value, for example, the first pass rate preset value is 95%, if it is greater than or equal to the first pass rate preset value Then execute S630, otherwise, execute S640.
  • the first qualified rate is the number of qualified samples/the total number of samples.
  • determining whether the qualified rate of the sample reaches the first qualified rate preset value can be replaced with determining whether the non-qualified rate of the sample all reaches the preset failure rate value, for example, the preset failure rate is 5 %, if the preset value of the failure rate is reached in S640, if the preset value of the failure rate is not reached, S630 is executed.
  • S640 The electronic device determines whether the number of positive samples is greater than the preset value of the first sample quantity, for example, the preset value of the first sample quantity is 500, if it is greater than the preset value of the first sample quantity, execute S640, if it is not less than or If it is equal to the preset value of the first sample quantity, return to S610, and continue to input samples to S610 through the method 500.
  • S650 The electronic device uses part of the qualified samples to train the first correction parameter of the weight model, and uses the first correction parameter to correct the weight model to obtain the corrected weight model.
  • 300 qualified samples can be used to train the first correction parameter of the weight model.
  • S650 includes: the process of the method 500 can be executed for each set of charging data in the qualified samples to obtain the second predicted charging duration corresponding to each set of charging data and the actual charging duration under each set of charging data, and compare the second predicted The charging time and the actual charging time are repeatedly determined according to the comparison result of the first correction parameter of the weight model.
  • the electronic device uses part of the qualified samples to train the first correction parameters of the weighting model
  • the actual charging duration under each set of charging data can be compared with the first predicted charging duration output by each basic prediction model. Therefore, among the multiple weight coefficients output by the first modified parameter to modify the weight model, among the first predicted charging time length output by the basic prediction model, the closer the actual charging time length is, the higher the weight coefficient of the corresponding basic prediction model.
  • the actual charging time under charging data 1 is 10 hours
  • the first predicted charging time output by the basic prediction model 1 is 9.5 hours
  • the first predicted charging time output by the basic prediction model 2 is 12 hours
  • the weight coefficient multiplied by the basic prediction model 1 output by the weight model is higher than the weight coefficient multiplied by the basic prediction model 2.
  • the weight coefficient multiplied by the basic prediction model 1 is 0.8.
  • the weight coefficient of prediction model 2 is 0.2.
  • S660 The electronic device uses the remaining qualified samples in the qualified samples to test the pass rate of the samples generated by the modified weight model, and then judge the pass rate according to S620. If the qualified rate meets the preset value of the qualified rate, the weight model is stable, and the weight model in the method 500 can be modified to the weight model obtained according to FIG. 6.
  • the number of qualified samples is 500, 300 qualified samples can be used to train the first correction parameter of the weight model, and the remaining 200 qualified samples are used to test the stability of the weight model corrected according to the first correction parameter.
  • the electronic device can determine the first correction parameter for correcting the first weight model, so as to ensure the accuracy of the weight model.
  • Case 2 The electronic device obtains the first correction parameter according to the samples in the first sample set, and sends the first correction parameter to the cloud, and the cloud fits the second correction parameter according to the first correction parameters of the multiple electronic devices. The specific steps are shown in Figure 8.
  • each electronic device of the plurality of electronic devices executes S710-S750, it sends the first modified parameter obtained by each electronic device to the cloud.
  • the cloud will receive multiple first correction parameters sent by multiple electronic devices. At least some of the multiple first correction parameters may be the same or different.
  • the modified parameter fitting generates a second modified parameter, and the second modified parameter generated by the fitting is sent to the electronic device.
  • S770 The electronic device modifies the weight model according to the second modified parameter to obtain a modified weight model.
  • the electronic device uses the remaining positive samples in the positive samples to test the pass rate of the samples generated by the modified weight model, and then judge the pass rate according to S720. If the qualification rate meets the preset value of the qualification rate, the weight model is stable.
  • the weight model in the method 500 may be modified to the weight model obtained according to FIG. 7.
  • the above method of correcting the weight model in FIG. 8 can simplify the complexity of the electronic device correction model.
  • Each electronic device sends the first correction parameter obtained by itself to the cloud.
  • the first correction parameter is the correction parameter of the weight model.
  • the user’s information of the electronic device can also protect the privacy of the user, which is conducive to improving security.
  • the second correction parameter determined by the cloud using big data can meet the robustness requirements.
  • Figure 4 above is the process of training multiple basic prediction models
  • Figure 5 is the process of training the weight model
  • Figure 6 is the process of using the model trained in Figures 4 and 5 to predict the charging time
  • Figure 7 and Figure 8 are the modified training of Figure 5 The process of weighting the model.
  • the prediction duration of the output of multiple basic prediction models obtained in Figure 4 can also be modified.
  • Figure 9 can be an independent implementation. For example, it may also be an embodiment combining the foregoing method, which is not limited in the embodiment of the present application.
  • FIG. 9 shows a method 800 for charging management and control according to an embodiment of the present application, including:
  • S810 The electronic device obtains first charging data.
  • the first charging data may be the charging parameters included in the first charging data of the method 500, and the value of the charging parameters included in the first charging data in the method 800 may be the same as the charging parameters included in the first charging data in the method 500.
  • the values are the same or different, in order to avoid repetitive descriptions, they are not described in detail here.
  • S820 The electronic device inputs at least part of the charging data in the first charging data into the first basic prediction model to obtain a third predicted charging duration.
  • the first basic prediction model may be preset and stored in the electronic device, or trained in the cloud according to method 300 and instructed to the electronic device, or trained by the electronic device itself according to method 300, which is implemented in this application
  • the example does not limit this.
  • S830 The electronic device adjusts the third predicted charging time length according to the adjustment parameter corresponding to the first basic prediction model to obtain the first predicted charging time length.
  • the method 800 is an independent embodiment, and the first predicted charging duration obtained by the method 800 can be used to control the charging of the electronic device, because the third predicted charging output from the first basic prediction model The duration may be quite different from the actual charging duration. Therefore, the adjustment parameters corresponding to the first basic prediction model can be used to adjust the third predicted charging duration, and the adjusted first predicted charging duration can be used to control the charging of electronic devices, thereby improving the prediction. Accuracy.
  • the first charging data in the method 800 may be the same as or different from the first charging data in the foregoing embodiment.
  • the first predicted charging duration in the method 800 may be the same as the first predicted charging time in the foregoing embodiment. The duration is the same or different, and the embodiment of the application does not limit it.
  • the embodiments of the method 800 and the method 500 can be combined, and the method 800 can be used to obtain a first predicted charging duration output by the first basic prediction model, and the method 800 can output multiple basic prediction models. A first predicted charging duration corresponding to each basic prediction model is then used as the first predicted charging duration of S530 in method 500.
  • the electronic device needs to obtain the adjustment parameters of each of the multiple basic prediction models.
  • the following describes how to obtain the adjustment parameters corresponding to the first basic prediction model in detail with reference to FIG. 10. Only obtaining the adjustment parameters corresponding to the first basic prediction model is described as an example. The adjustment parameters corresponding to other basic prediction models are similar to obtaining the adjustment parameters corresponding to the first basic prediction model. One enumerate.
  • the electronic device obtains the second charging data and the actual charging duration corresponding to the second charging data.
  • the electronic device inputs the second charging data into the first basic prediction model and outputs the fourth predicted charging duration.
  • the first basic prediction model can It was trained according to Method 300.
  • the electronic device adds the second charging data, the fourth predicted charging duration, and the second actual charging duration as a sample to the second sample set, and so on to determine the samples in the second sample set.
  • S910 The electronic device determines qualified samples in the second sample set.
  • the absolute value of the difference between the second actual charging duration and the fourth predicted charging duration in a sample is less than the preset duration. For example, if the preset duration is 60 min, the sample is a qualified sample, also called Positive samples, otherwise unqualified samples, also called negative samples.
  • the fourth predicted charging duration of the qualified samples needs to be greater than the preset value,
  • the qualified samples are the actual charging behavior of the user that lasts for at least the preset duration, and avoid some samples that are caused by the user's charging habit of unplugging immediately after charging and powering on, which affects accuracy.
  • S920 The electronic device determines whether the second pass rate of the sample is greater than or equal to the second preset value of pass rate, and if the second pass rate is greater than or equal to the second preset value of pass rate, execute S930; otherwise, execute S940.
  • the second pass rate is the number of qualified samples/the total number of samples.
  • determining whether the qualified rate of the sample reaches the second qualified rate preset value can be replaced with determining whether the non-qualified rate of the sample reaches the preset failure rate value, for example, the preset failure rate is 5% If the preset value of the failure rate is reached in S940, if the preset value of the failure rate is not reached, go to S930.
  • the initial adjustment parameter may be a preset value or a value obtained based on historical experience data, which is not limited in the embodiment of the present application.
  • S940 The electronic device determines whether the number of qualified samples is greater than the second preset value of the number of samples, and if it is greater than the preset value of the second sample number, execute S950; otherwise, execute S910 to determine whether there are new qualified samples.
  • the electronic device trains the adjustment parameters of the first basic prediction model by using part of the qualified samples.
  • the partially qualified samples are used to perform linear fitting to obtain the regression coefficients and constants of the linear relationship, namely It is the adjustment parameter of the first basic prediction model.
  • the second actual charging time length is y
  • the partially qualified samples are used for non-linear training to obtain the coefficients of the non-linear relationship. It is the adjustment parameter of the first basic prediction model.
  • S960 The electronic device updates the adjustment parameters of the first basic prediction model.
  • the electronic device uses the remaining qualified samples to test the stability of the updated adjustment parameters.
  • multiple weight coefficients and the first predicted charging duration corresponding to each basic prediction model can be used to determine the second predicted charging duration.
  • FIG. 11 a certain A schematic diagram of the comparison between the actual charging time length of n (n is a positive integer) times of the electronic device and the second predicted charging time length of each prediction.
  • Figure 12 when the charging start time of the electronic device is 22:00 in the evening, if the charging of the battery of the electronic device is not controlled, in the upper column of the diagram in Figure 12, when it is 23:00, the battery of the electronic device About to be fully charged, the battery will enter an expanded state at 00:00 until the end of charging at 08:00 the next morning.
  • the battery can be charged and controlled from 23:00 to 23:00. It enters the protection state at 06:00 and does not charge the electronic device. Continue to charge after 06:00 until it reaches 100% at 08:00, thereby reducing the time the battery is in the swelling state and improving the service life of the battery.
  • step x is greater than A
  • step y is less than or equal to A
  • the embodiment of this application is not limited.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered as going beyond the scope of protection of this application.
  • the embodiments of the present application may divide the electronic device into functional modules according to the foregoing method examples.
  • each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software functional modules. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other feasible division methods in actual implementation. The following is an example of dividing each function module corresponding to each function.
  • FIG. 13 is a schematic block diagram of an apparatus 1000 for charging management and control provided by an embodiment of the application.
  • the device 1000 includes an acquisition unit 1010 and a processing unit 1020.
  • the acquisition unit 1010 can communicate with the outside.
  • the obtaining unit 1010 may also be referred to as a communication interface or a communication unit, and the obtaining unit 1010 is configured to perform the operations related to obtaining or receiving and sending on the electronic device side in the embodiments of FIGS. 6-8 or 10 above.
  • the processing unit 1020 is used to perform data processing, and the processing unit 1020 is used to perform processing-related operations on the electronic device side in the embodiments of FIGS. 6-8 or 10 above.
  • the obtaining unit 1010 is used to obtain first charging data
  • the processing unit 1020 is configured to input at least part of the charging data in the first charging data into multiple basic prediction models, and determine the first predicted charging duration corresponding to each basic prediction model;
  • the processing unit 1020 is further configured to input at least part of the charging data in the first charging data into a weighting model to obtain multiple weighting coefficients;
  • the processing unit 1020 is further configured to determine a second predicted charging duration according to the multiple weighting coefficients and the first predicted charging duration corresponding to each basic prediction model, and the second predicted charging duration is used for monitoring the electronic device. Charge control.
  • the obtaining unit 1010 is further configured to: obtain the first actual charging duration of the electronic device corresponding to the first charging data;
  • the processing unit 1020 is further configured to:
  • the weight model is updated according to the samples in the first sample set.
  • processing unit 1020 is specifically configured to:
  • first qualified rate is less than the preset value of the first qualified rate, determine the number of qualified samples in the first sample set
  • the weight model is modified according to some samples in the qualified samples.
  • processing unit 1020 is specifically configured to:
  • the first modified parameter is obtained by training according to the partial samples, and the weight model is modified according to the first modified parameter to obtain the modified weight model; or including:
  • the device 1000 further includes:
  • a transceiver unit configured to send the first correction parameter to the cloud, and receive a second correction parameter determined by the cloud according to the first correction parameter;
  • the processing unit 1020 is further configured to correct the weight model according to the second correction parameter to obtain a corrected weight model.
  • processing unit 1020 is further configured to: according to the remaining samples in the qualified samples
  • the remaining samples test the stability of the modified weight model.
  • the processing unit 1020 is specifically configured to: input at least part of the charging data in the first charging data into the multiple basic prediction models to obtain the third corresponding to each basic prediction model. Forecast charging time;
  • the adjustment parameters of each basic prediction model are used to adjust the third predicted charging duration corresponding to each basic prediction model to obtain the first predicted charging duration corresponding to each basic prediction model.
  • the obtaining unit 1010 is further configured to: obtain the second charging data of the electronic device and the second actual charging duration corresponding to the second charging data;
  • the processing unit 1020 is further configured to:
  • the adjustment parameter corresponding to the first basic prediction model is determined according to the samples in the second sample set.
  • processing unit 1020 is specifically configured to:
  • the adjustment parameters corresponding to the first basic prediction model are determined according to the qualified samples.
  • the adjustment parameter corresponding to the first basic prediction model is the regression coefficient and the linear relationship constant.
  • the first charging data and the second charging data include at least one of the following: the type of charger used to charge the electronic device, the manufacturer that produces the battery of the electronic device, and the The nominal capacity of the battery, the cell type of the battery, the type of charging cable used to charge the electronic device, the number of cycles that the battery has been charged, and the nominal number of cycles that the battery can be charged , The average internal resistance of the battery, the maximum internal resistance of the battery, the historical plug-in time and historical unplug time of the charger, the charge cut-off time of the battery cell, the start of the battery cell Starting power and ending power, the actual charging time for the preset time period of each day within the preset number of days.
  • FIG. 14 is a schematic block diagram of another device 1100 for charging management and control provided by an embodiment of the application.
  • the device 1100 includes an obtaining unit 1110 and a processing unit 1120.
  • the obtaining unit 1110 can communicate with the outside.
  • the obtaining unit 1110 may also be referred to as a communication interface or a communication unit, and the obtaining unit 1110 is configured to perform operations related to obtaining or receiving and sending on the electronic device side in the embodiment of FIG. 9 or FIG. 10 above.
  • the processing unit 1120 is used to perform data processing, and the processing unit 1120 is used to perform processing-related operations on the electronic device side in the embodiment of FIG. 9 or FIG. 10 above.
  • the obtaining unit 1110 is used to obtain the first charging data.
  • the processing unit 1120 is configured to: input at least part of the charging data in the first charging data into a first basic prediction model to obtain a third prediction duration; and adjust the third prediction according to the adjustment parameters corresponding to the first basic prediction
  • the charging time length obtains the first predicted charging time length, and the first predicted charging time length is used to manage and control the charging of the electronic device.
  • the acquiring unit 1110 is further configured to acquire the second charging data of the device and the second actual charging duration corresponding to the second charging data.
  • the processing unit 1120 is further configured to: input the second charging data into the first basic prediction model to obtain a fourth predicted charging duration; combine the fourth predicted charging duration, the second charging data, and the The second actual charging duration is added as a sample to a second sample set; and the adjustment parameter corresponding to the first basic prediction model is determined according to the samples in the second sample set.
  • processing unit 1120 is specifically configured to:
  • the adjustment parameters corresponding to the first basic prediction model are determined according to the qualified samples.
  • the adjustment parameter corresponding to the first basic prediction model is the regression coefficient and the linear relationship constant.
  • the first charging data and the second charging data include at least one of the following: the type of charger used to charge the device, the manufacturer of the battery of the device, and the battery The nominal capacity of the battery, the cell type of the battery, the type of charging cable used to charge the device, the number of cycles that the battery has been charged, the nominal number of cycles that the battery can be charged, the The average internal resistance of the battery, the maximum internal resistance of the battery, the historical insertion time and historical unplugging time of the charger, the charging cut-off time of the battery cell, the initial capacity of the battery cell and The actual charging time during the preset period of time each day within the preset number of days when the battery is terminated.
  • FIG. 15 is a schematic structural diagram of an apparatus 1200 for charging management and control provided by an embodiment of the present application.
  • the communication device 1200 includes: a processor 1210, a memory 1220, a communication interface 1230, and a bus 1240.
  • the processor 1210 in the apparatus 1200 shown in FIG. 15 may correspond to the processing unit 1020 in the apparatus 1000 in FIG. 13.
  • the communication interface 1230 in the device 1200 shown in FIG. 15 may correspond to the acquiring unit 1010 in the device 1000 in FIG. 13.
  • the processor 1210 in the apparatus 1200 shown in FIG. 15 may correspond to the processing unit 1120 in the apparatus 1100 in FIG. 14.
  • the communication interface 1230 in the device 1200 shown in FIG. 15 may correspond to the acquiring unit 1110 in the device 1100 in FIG. 14.
  • the processor 1210 may be connected to the memory 1220.
  • the memory 1220 can be used to store the program code and data. Therefore, the memory 1220 may be a storage unit inside the processor 1210, or an external storage unit independent of the processor 1210, or may include a storage unit inside the processor 1210 and an external storage unit independent of the processor 1210. part.
  • the apparatus 1200 may further include a bus 1240.
  • the memory 1220 and the communication interface 1230 may be connected to the processor 1210 through a bus 1240.
  • the bus 1240 may be a peripheral component interconnect standard (PCI) bus or an extended industry standard architecture (EISA) bus, etc.
  • the bus 1240 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one line is used in FIG. 15, but it does not mean that there is only one bus or one type of bus.
  • the processor 1210 may adopt a central processing unit (CPU).
  • the processor can also be other general-purpose processors, digital signal processors (digital signal processors, DSP), application specific integrated circuits (ASICs), ready-made programmable gate arrays (field programmable gate arrays, FPGAs) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the processor 1210 adopts one or more integrated circuits to execute related programs to implement the technical solutions provided in the embodiments of the present application.
  • the memory 1220 may include a read-only memory and a random access memory, and provides instructions and data to the processor 810.
  • a part of the processor 810 may also include a non-volatile random access memory.
  • the processor 810 may also store device type information.
  • the processor 1210 executes the computer-executable instructions in the memory 1220 to execute the operation steps of the foregoing method through the apparatus 1200.
  • the device 1200 may correspond to the device 1000 and the device 1100 in the embodiment of the present application, and the above-mentioned and other operations and/or functions of each unit in the device 1000 and the device 1100 are used to implement the method. For the sake of brevity, the corresponding process will not be repeated here.
  • the embodiments of the present application also provide a computer-readable medium, the computer-readable medium storing program code, when the computer program code is run on the computer, the computer executes The methods in the above aspects.
  • the embodiments of the present application further provide a computer program product
  • the computer program product includes: computer program code, when the computer program code runs on a computer, the computer executes the above Methods in all aspects.
  • the terminal device or the network device includes a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system layer.
  • the hardware layer may include hardware such as a central processing unit (CPU), a memory management unit (MMU), and memory (also referred to as main memory).
  • the operating system at the operating system layer can be any one or more computer operating systems that implement business processing through processes, such as Linux operating systems, Unix operating systems, Android operating systems, iOS operating systems, or windows operating systems.
  • the application layer can include applications such as browsers, address books, word processing software, and instant messaging software.
  • the embodiment of this application does not specifically limit the specific structure of the execution subject of the method provided in the embodiment of this application, as long as it can run a program that records the code of the method provided in the embodiment of this application, according to the method provided in the embodiment of this application.
  • the execution subject of the method provided in the embodiments of the present application may be a terminal device or a network device, or a functional module in the terminal device or the network device that can call and execute the program.
  • Computer-readable media may include, but are not limited to: magnetic storage devices (for example, hard disks, floppy disks, or tapes, etc.), optical disks (for example, compact discs (CD), digital versatile discs (digital versatile disc, DVD), etc.), etc. ), smart cards and flash memory devices (for example, erasable programmable read-only memory (EPROM), cards, sticks or key drives, etc.).
  • magnetic storage devices for example, hard disks, floppy disks, or tapes, etc.
  • optical disks for example, compact discs (CD), digital versatile discs (digital versatile disc, DVD), etc.
  • smart cards and flash memory devices for example, erasable programmable read-only memory (EPROM), cards, sticks or key drives, etc.
  • the various storage media described herein may represent one or more devices and/or other machine-readable media for storing information.
  • the term "machine-readable medium” may include, but is not limited to, wireless channels and various other media capable of storing, containing, and/or carrying instructions and/or data.
  • processors mentioned in the embodiments of this application may be a central processing unit (central processing unit, CPU), or other general-purpose processors, digital signal processors (digital signal processors, DSP), and application-specific integrated circuits ( application specific integrated circuit (ASIC), ready-made programmable gate array (field programmable gate array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc.
  • CPU central processing unit
  • DSP digital signal processors
  • ASIC application specific integrated circuit
  • FPGA field programmable gate array
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the memory mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), and electrically available Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • the volatile memory may be random access memory (RAM).
  • RAM can be used as an external cache.
  • RAM may include the following various forms: static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM) , Double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (synchlink DRAM, SLDRAM) and Direct RAM Bus RAM (DR RAM).
  • static random access memory static random access memory
  • dynamic RAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM synchronous DRAM
  • Double data rate synchronous dynamic random access memory double data rate SDRAM, DDR SDRAM
  • enhanced SDRAM enhanced synchronous dynamic random access memory
  • SLDRAM Direct RAM Bus RAM
  • the processor is a general-purpose processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component
  • the memory storage module
  • memories described herein are intended to include, but are not limited to, these and any other suitable types of memories.
  • the disclosed system, device, and method can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or It can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present application may be integrated into one unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the essence of the technical solution of this application, or the part that contributes to the existing technology, or the part of the technical solution, can be embodied in the form of a computer software product, and the computer software product is stored in a storage
  • the computer software product includes several instructions, which are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media may include but are not limited to: U disk, mobile hard disk, read-only memory (read-only memory, ROM), random access memory (random access memory, RAM), magnetic disks or optical disks, etc., which can store programs The medium of the code.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)
  • Secondary Cells (AREA)

Abstract

提供了一种用于充电管控的方法和装置。电子设备获得第一充电数据（S510）；电子设备将第一充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长（S520）；电子设备将第一充电数据输入到权重模型得到多个权重系数（S530）；电子设备根据多个权重系数和每个基础预测模型对应的第一预测充电时长确定第二预测时长，电子设备根据第二预测充电时长对电子设备的充电进行管控（S540）。本方法和装置可以延长电池的寿命并提高电池的续航能力。

Description

用于充电管控的方法和装置
本申请要求于2020年12月10日提交中国专利局、申请号为202011463325.5、申请名称为“用于充电管控的方法和装置”，以及，要求于2020年4月30日提交中国专利局、申请号为202010361658.0、申请名称为“用于充电管控的方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及充电领域，并且更具体地涉及充电领域中的用于充电管控的方法和装置。
背景技术
不同用户在对终端设备进行充电时的充电习惯不同，不同的充电习惯会影响电池的寿命，例如过度充电会将降低电池寿命并且降低续航能力，从而影响用户体验。
为了解决上述问题，有必要预测用户的充电时长，并根据预测的充电时长对终端设备的充电进行管控，从而延长电池的寿命并且提高续航能力，因此，如何预测充电时长是亟需解决的问题。
发明内容
本申请实施例提供了一种用于充电管控的方法和装置，能够根据预测的时长对终端设备的充电进行管控，有助于延长电池的寿命并且提高电池的续航能力。
第一方面，提供了一种用于充电管控的方法，所述方法可由电子设备执行，电子设备可以是能够支持电子设备实现该方法所需的功能的装置，例如芯片系统。包括：获取第一充电数据；将所述第一充电数据中的至少部分充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长；将所述第一充电数据中的至少部分充电数据输入到权重模型中，得到多个权重系数；根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，所述第二预测充电时长用于对电子设备的充电进行管控。
在上述方案中，电子设备根据多个基础预测模型得到每个基础预测模型对应的第一预测充电时长，电子设备根据权重模型得到多个权重系数。电子设备根据多个权重系数和每个基础预测模型对应的第一预测充电时长确定第二预测时长，电子设备根据第二预测充电时长对电子设备的充电进行管控，从而可以延长电池的寿命并提高电池的续航能力，有助于提高用户体验。
其中，多个权重系数中至少存在两个权重系数相同，或者多个权重系数中至少存在两个权重系数不同。
可以理解的是，输入每个基础预测模型中的充电数据可以不同，或者输入至少两个基础预测模型的充电数据相同。
也可以理解的是，输入权重模型的充电数据和输入多个基础预测模型的充电数据 可以不同。
在一些可能的实现方式中，多个基础预测模型中不同的基础预测模型对应不同的应用场景。
在一些可能的实现方式中，根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，包括：将所述多个权重系数与所述每个基础预测模型对应的第一预测充电时长进行加权计算，得到第二预测充电时长。
在一些可能的实现方式中，所述方法还包括：获取所述第一充电数据对应的所述电子设备的第一实际充电时长；将所述第一充电数据、所述第一实际充电时长和所述第二预测充电时长作为样本添加到第一样本集；根据所述第一样本集中的样本更新所述权重模型。
这样，可以实时的更新第一样本集中的样本，从而可以保证权重模型的准确性。
在一些可能的实现方式中，所述根据所述第一样本集中的样本更新所述权重模型，包括：确定所述第一样本集中的样本的第一合格率；若所述第一合格率小于第一合格率预设值，确定所述第一样本集中的合格的样本的数量；若所述第一样本集中的合格的样本的数量大于第一样本数量预设值，根据所述合格的样本中部分样本修正所述权重模型。
在上述方案中，电子设备确定第一样本集中的样本不满足第一合格率的要求时，确定第一样本集中的合格的样本数据是否满足第一样本数量预设值的要求，如果满足要求，则根据合格的样本中的部分样本修正权重模型，从而可以保证修正后的权重模型的准确性。
上述的用于修正权重模型的合格样本的数量需要满足预设数量，从而才能保证修正的权重模型的鲁棒性要求。
在一些可能的实现方式中，所述根据所述合格的样本中的部分样本修正所述权重模型，包括：根据所述部分样本训练得到第一修正参数，根据所述第一修正参数修正所述权重模型，得到修正后的权重模型。
上述方案中，电子设备可以确定用于修正第一权重模型的第一修正参数，从而能够保证权重模型的准确性。
在一些可能的实现方式中，所述根据所述合格的样本中的部分样本修正所述权重模型，包括：根据所述部分样本训练得到所述第一修正参数，并将所述第一修正参数发送给云端；
接收云端根据所述第一修正参数确定的第二修正参数；
根据所述第二修正参数修正所述权重模型，得到修正后的权重模型。
上述方案中，可以简化电子设备修正模型的复杂度，每个电子设备将自身得到的第一修正参数发送给云端，第一修正参数为权重模型的修正参数，没有任何的电子设备的用户的信息，也能保护用户的私密性，有利于提高安全性。同时云端利用大数据确定的第二修正参数，能满足鲁棒性的要求。
在一些可能的实现方式中，所述方法还包括：根据所述合格的样本中的剩余部分样本测试所述修正后的权重模型的稳定性。
在一些可能的实现方式中，所述将所述第一充电数据中的至少部分充电数据输入 到多个基础预测模型中，得到每个基础预测模型对应的第一预测充电时长，包括：
将所述第一充电数据中的至少部分充电数据输入到所述多个基础预测模型中，得到每个基础预测模型对应的第三预测充电时长；
利用所述每个基础预测模型的调整参数调整所述每个基础预测模型对应的第三预测充电时长，得到所述每个基础预测模型对应第一预测充电时长。
在一些可能的实现方式中，所述方法还包括：
获取所述电子设备的第二充电数据和所述第二充电数据对应的第二实际充电时长；
将所述第二充电数据输入到所述多个基础预测模型的第一基础预测模型中，得到第四预测充电时长；
将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；
根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
在一些可能的实现方式中，所述根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数，包括：
确定所述第二样本集中的样本的第二合格率；
若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本的数量；
若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
在一些可能的实现方式中，若所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
在一些可能的实现方式中，如果所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足非线性关系，则可以利用部分合格的样本进行非线性训练，得到非线性关系的系数即为第一基础预测模型的调整参数。
在一些可能的实现方式中，所述第一充电数据和第二充电数据包括以下至少一项：用于对所述电子设备进行充电的充电器的类型、生产所述电子设备的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述电子设备进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
第二方面，提供了一种用于充电管控的方法，包括：获取第一充电数据；将所述第一充电数据中的至少部分充电数据输入第一基础预测模型中，得到第三预测时长；根据所述第一基础预测对应的调整参数调整所述第三预测充电时长，得到第一预测充电时长，所述第一预测充电时长用于对电子设备的充电进行管控。
在上述方案中，电子设备能够利用第一基础预测模型对应的调整参数调整第一基础预测模型得到的第三预测充电时长，得到用于充电管控的第一预测充电时长，换句话说，即使利用第一基础预测模型得到的第三预测充电时长不准确，则可以利用调整 参数调整，从而可以得到可能准确的第一预测充电时长，也能提高充电管控的准确性。
在一些可能的实现方式中，所述方法还包括：获取所述电子设备的第二充电数据和所述第二充电数据对应的第二实际充电时长；将所述第二充电数据输入到所述第一基础预测模型中，得到第四预测充电时长；将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
上述方案中的第一基础预测模型对应的调整参数是根据多个实际样本得到的，因此，能够满足实际的调整需求。
在一些可能的实现方式中，所述根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数，包括：
确定所述第二样本集中的样本的第二合格率；
若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本的数量；
若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
在一些可能的实现方式中，若所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
在一些可能的实现方式中，如果所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足非线性关系，则可以利用部分合格的样本进行非线性训练，得到非线性关系的系数即为第一基础预测模型的调整参数。
在一些可能的实现方式中，所述第一充电数据和第二充电数据包括以下至少一项：用于对所述电子设备进行充电的充电器的类型、生产所述电子设备的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述电子设备进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
第三方面，提供一种用于充电管控的装置，所述装置用于执行上述第一方面或第一方面的任一可能的实现方式中的方法。具体地，所述装置可以包括用于执行第一方面或第一方面的任一可能的实现方式中的方法的模块。
第四方面，提供一种用于充电管控的装置，所述装置用于执行上述第二方面或第二方面的任一可能的实现方式中的方法。具体地，所述装置可以包括用于执行第二方面或第二方面的任一可能的实现方式中的方法的模块。
第五方面，提供一种用于充电管控的装置，所述装置包括处理器，处理器与存储器耦合，存储器用于存储计算机程序或指令，处理器用于执行存储器存储的计算机程序或指令，使得第一方面中的方法被执行。
例如，处理器用于执行存储器存储的计算机程序或指令，使得该装置执行第一方面中的方法。
可选地，该装置包括的处理器为一个或多个。
可选地，该装置中还可以包括与处理器耦合的存储器。
可选地，该装置包括的存储器可以为一个或多个。
可选地，该存储器可以与该处理器集成在一起，或者分离设置。
可选地，该装置中还可以包括收发器。
第六方面，提供一种用于充电管控的装置，所述装置包括处理器，处理器与存储器耦合，存储器用于存储计算机程序或指令，处理器用于执行存储器存储的计算机程序或指令，使得第二方面中的方法被执行。
例如，处理器用于执行存储器存储的计算机程序或指令，使得该装置执行第二方面中的方法。
可选地，该装置包括的处理器为一个或多个。
可选地，该装置中还可以包括与处理器耦合的存储器。
可选地，该装置包括的存储器可以为一个或多个。
可选地，该存储器可以与该处理器集成在一起，或者分离设置。
可选地，该装置中还可以包括收发器。
第七方面，提供一种计算机可读存储介质，其上存储有用于实现第一方面中的方法的计算机程序(也可称为指令或代码)。
例如，该计算机程序被计算机执行时，使得该计算机可以执行第一方面中的方法。
第八方面，提供一种计算机可读存储介质，其上存储有用于实现第一方面或者第二方面中的方法的计算机程序(也可称为指令或代码)。
例如，该计算机程序被计算机执行时，使得该计算机可以执行第二方面中的方法。
第九方面，本申请提供一种芯片，包括处理器。处理器用于读取并执行存储器中存储的计算机程序，以执行第一方面及其任意可能的实现方式中的方法。
可选地，所述芯片还包括存储器，存储器与处理器通过电路或电线与存储器连接。
进一步可选地，所述芯片还包括通信接口。
第十方面，本申请提供一种芯片系统，包括处理器。处理器用于读取并执行存储器中存储的计算机程序，以执行第二方面及其任意可能的实现方式中的方法。
可选地，所述芯片还包括存储器，存储器与处理器通过电路或电线与存储器连接。
进一步可选地，所述芯片还包括通信接口。
第十一方面，本申请提供一种计算机程序产品，所述计算机程序产品包括计算机程序(也可称为指令或代码)，所述计算机程序被计算机执行时使得所述计算机实现第一方面中的方法。
第十二方面，本申请提供一种计算机程序产品，所述计算机程序产品包括计算机程序(也可称为指令或代码)，所述计算机程序被计算机执行时使得所述计算机实现第二方面中的方法。
附图说明
图1是本申请实施例提供的管控策略的示意图。
图2是本申请实施例提供的系统架构示意图。
图3是本申请实施例提供的系统架构示意图。
图4是本申请实施例提供的训练基础预测模型的示意图。
图5是本申请实施例提供的训练权重模型的示意图。
图6是本申请实施例提供的用于充电管控的方法示意图。
图7是本申请实施例提供的修正权重模型的示意图。
图8是本申请实施例提供的另一修正权重模型的示意图。
图9是本申请实施例提供的另一用于充电管控的方法示意图。
图10是本申请实施例提供的获取第一基础预测模型对应的调整参数的方法示意图。
图11是本申请实施例提供的效果示意图。
图12是本申请实施例提供的管控策略示意图。
图13是本申请实施例提供的用于充电管控的装置的示意性框图。
图14是本申请实施例提供的另一用于充电管控的装置的示意性框图。
图15是本申请实施例提供的又一用于充电管控的装置的示意性框图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行描述。
不同的用户在对电子设备进行充电时，会有不同的充电习惯。不同的充电习惯会影响电子设备的电池的寿命。例如，有的用户喜欢前一天晚上插入充电器开始充电直到第二天早晨拔掉充电器，这样会对电池造成被过度充电。例如，有的用户喜欢白天上班时间充电，充满就拔掉，这样不会影响电池的寿命。例如，有的用户喜欢给同一电子设备用不同功率的充电器进行充电，具体地，有的用户喜欢用小功率的充电器给需要大功率的电子设备充电，这样会使得充电时间过长，有的用户喜欢用大功率的充电器给需要小功率的电子设备充电，这样会使得电池很快达到饱和状态，如何不及时拔掉充电器，则会造成过度充电。例如，有的用户喜欢边用电子设备边充电，电池需要不停的充放电，缩短电池的使用寿命。
在上述的用户习惯中，如果电池过度被充电，则会使得电池长时间处于膨胀状态，导致电池寿命降低并且续航能力也降低，从而影响用户体验。
为了延长电池的寿命，可以针对不同的用户习惯预测充电时长，利用预测的充电时长对电子设备的充电进行管控。例如，对于过度充电来说，图1中的t3为根据用户习惯得到的预测充电时长，电子设备通过t3确定可以采用图1的管控策略：在0～t1时间段正常充电，在达到70％的充电量时，t1～t2时间段进入保护状态，不对电子设备进行充电，或者是利用很小的电流进行充电，在t2之后继续充电，直到t3达到100％，不同的管控策略t1、t2和t3的取值不同。图1的管控策略只是示例性的描述，不应该造成对本申请的任何限定。本申请实施例的利用预测的充电时长对电子设备的充电进行管控还可以是其他的管控策略，本申请实施例对此不作限定。
需要说明的是，可以利用本申请实施例提供的用于充电管控的方法对任何一个需要充电的电子设备进行管控，例如，可以对终端设的充电进行管控。
本申请实施例提到的终端设备为用户设备(user equipment，UE)、移动台(mobile  station，MS)、移动终端(mobile terminal，MT)、接入终端、用户单元、用户站、移动站、移动台、远方站、远程终端、移动设备、用户终端、终端、无线通信设备、用户代理或用户装置等。
终端设备可以是一种向用户提供语音/数据连通性的设备，例如，具有无线连接功能的手持式设备、车载设备等。目前，一些终端的举例为：手机(mobile phone)、平板电脑、笔记本电脑、掌上电脑、移动互联网设备(mobile internet device，MID)、可穿戴设备，虚拟现实(virtual reality，VR)设备、增强现实(augmented reality，AR)设备、工业控制(industrial control)中的无线终端、无人驾驶(self driving)中的无线终端、远程手术(remote medical surgery)中的无线终端、智能电网(smart grid)中的无线终端、运输安全(transportation safety)中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端、蜂窝电话、无绳电话、会话启动协议(session initiation protocol，SIP)电话、无线本地环路(wireless local loop，WLL)站、个人数字助理(personal digital assistant，PDA)、具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备、可穿戴设备，5G网络中的终端设备或者未来演进的公用陆地移动通信网络(public land mobile network，PLMN)中的终端设备等，本申请实施例对此并不限定。
作为示例而非限定，在本申请实施例中，该终端设备还可以是可穿戴设备。可穿戴设备也可以称为穿戴式智能设备，是应用穿戴式技术对日常穿戴进行智能化设计、开发出可以穿戴的设备的总称，如眼镜、手套、手表、服饰及鞋等。可穿戴设备即直接穿在身上，或是整合到用户的衣服或配件的一种便携式设备。可穿戴设备不仅仅是一种硬件设备，更是通过软件支持以及数据交互、云端交互来实现强大的功能。广义穿戴式智能设备包括功能全、尺寸大、可不依赖智能手机实现完整或者部分的功能，例如：智能手表或智能眼镜等，以及只专注于某一类应用功能，需要和其它设备如智能手机配合使用，如各类进行体征监测的智能手环、智能首饰等。
此外，在本申请实施例中，终端设备还可以是物联网(internet of things，IoT)系统中的终端设备，IoT是未来信息技术发展的重要组成部分，其主要技术特点是将物品通过通信技术与网络连接，从而实现人机互连，物物互连的智能化网络。本申请的终端设备还可以是作为一个或多个部件或者单元而内置于车辆的车载模块、车载模组、车载部件、车载芯片或者车载单元，车辆通过内置的所述车载模块、车载模组、车载部件、车载芯片或者车载单元可以实施本申请的方法。因此，本申请实施例可以应用于车联网，例如车辆外联(vehicle to everything，V2X)、车间通信长期演进技术(long term evolution-vehicle，LTE-V)、车到车(vehicle-to-vehicle，V2V)等。
下面结合图2描述本申请实施例提供的系统架构示意图。如图2所示，多个电子设备在端侧充电开始，利用多次充电的充电数据进行人工智能(artificial intelligence，AI)模型预测，利用预测的数据进行电池的智能管控，也可以根据预测的充电时长和真实的充电时长对端侧模型的进行更新。此外，多个端侧得到的模型的修正参数(例如权重模型的修正参数)可以发送到云端，云端聚合之后再发送回各个端侧，各个端侧利用聚合后的修正参数修正模型。这样，多个端侧发送的修正参数没有任何的用户的信息，也能保护用户的私密性，有利于提高安全性。同时云端利用大数据拟合修正 参数，能满足鲁棒性的要求。
下面结合图3描述本申请实施例的系统架构。如图3所示，在本申请实施例中，包括：确定模型，利用确定的模型得到电子设备的预测充电时长，其中，图3的预测充电时长模块可以为图2的AI模型预测模型，图3的充电管控模块可以为图2的电池智能管控模块，预测充电时长用于对电子设备的充电进行管控，也可以利用预测充电时长对模型进行修正。可选地，图3也可以不包括修正过程，其中，确定模型包括确定权重模型和多个基础预测模型。可选地，权重模型可以是预设的或者训练得到的或者云端指示的。可选地，多个基础预测模型也可以是预设的或者训练得到的或者云端指示的，本申请实施例对此不作限制。
如果上述多个基础预测模型是云端训练得到的并指示给电子设备的或者电子设备训练得到，电子设备和云端可以根据图4所示的步骤训练得到多个基础预测模型，图4所示的训练过程可以由电子设备执行或者云端执行，下面仅以训练得到一个基础预测模型(第一基础预测模型)为例描述，具体地，方法300包括：
S310，确定训练第一基础预测模型的训练样本集。
在S310之前，可以预设一个初始基础预测模型，初始基础预测模型的模型参数为预设的初始值。其中，训练样本集中的样本包括：多组充电数据、每组充电数据对应的实际充电时长以及每组充电数据输入初始基础预测模型得到的多个预测充电时长。其中，一组充电数据可以是电子设备的一条历史充电数据，将一条特定的历史充电数据的输入到初始基础预测模型可以得到一个预测充电时长，一条特定的历史充电数据条件下对应电子设备的实际充电时长。这样，训练样本集中的一个样本为一组充电数据、该组充电数据输入到初始基础预测模型中所得到的预测充电时长，以及该组充电数据对应的实际充电时长。
其中一条历史充电数据可以包括以下至少一项充电参数：用于对电子设备进行充电的充电器的类型、生产电子设备的电池的厂家、电池的标称容量、电池的电芯类型、用于对电子设备进行充电的充电线的类型、电池已经被充电的次数、电池能够充电的标称次数、电池的平均内阻、电池的最大内阻、充电器的一次历史插入时间和历史拔出时间、电池的电芯的充电截止时间、电池的电芯的一次历史起始电量和历史终止电量、在预设天数内在每天特定时间段的实际充电时长等。当然，一组充电数据可以包括其他的充电参数，本申请实施例对此不作限制。
可选地，训练样本集中的样本可以是标有正样本标签的样本。具体地，如果一组充电数据输入初始基础预测模型中输出的预测充电时长与该组充电数据下的实际充电时长的差值的绝对值小于预设值，则该样本为正样本，可以作为训练样本集中的样本。
可选地，训练样本集中的样本除了具有正样本标签的之外，训练样本集中的样本对应的预测充电时长大于预设值，这样可以避免样本为用户刚充上电就拔掉的充电习惯产生的充电数据而产生的样本，影响基础预测模型训练的准确性。
S320，判断训练样本集的中样本数量是否达到预设的训练样本数量，如果训练样本集中的样本数量达到预设的训练样本数量，则执行S330，否则执行S310继续向训练样本集中增加样本。
需要说明的是，预设的训练样本数量可以是协议规定固定的值，也可以是可变的 值，本申请实施例对此不作限定。
S330，根据训练样本集中的样本训练得到第一基础预测模型。
具体地，S330，包括：可以根据样本集中的样本输入初始基础预测模型中，初始基础预测模型输出一个预测充电时长，比较实际充电时长与预测充电时长，根据比较结果反复修正初始基础预测模型的系数，当修正后的基础预测模型输出的预测充电时长与实际充电时长的精度长时间小于预设值，并持续预设的时间段，则修正后的基础预测模型为第一基础预测模型。依次类推，可以得到其他的基础预测模型。
不同的基础预测模型，可以是针对不同的充电模式应用场景，例如，有的基础预测模型针对夜晚充电模式，有的基础预测模型针对白天充电模式，有的基础预测模型针对工作日充电模式，有的基础预测模型针对休息日充电模式。不同应用场景下，基础预测模型的输入可能不同，也可能相同，本申请实施例对此不作限制。
基础预测模型可以是支持向量机(support vector machine，SVM)模型、决策树(decision tree)模型、神经网络(neural network)模型、套袋(Bagging)模型或者提升(boosting)模型。当然，权重模型还可以是其他的模型，本申请实施例对此不作限定。
如果权重模型是电子设备训练得到的或者云端训练得到的并指示给电子设备的，电子设备和云端可以根据图5所示的步骤训练得到权重模型，具体地图5所示的方法400包括：
S410，确定训练权重模型的训练样本集。
在S410之前，可以预设一个初始权重模型，初始权重模型的模型参数为预设的初始值。训练样本集中的样本包括：多组充电数据、每组充电数据对应的实际充电时长以及每组充电数据中输入到初始权重模型和多个基础预测模型得到的多个预测充电时长(具体参见方法500的描述)。其中，一组充电数据可以是电子设备的一条历史充电数据，将一条特定的历史充电数据输入到初始权重模型得到多个初始权重系数，将一条特定的历史充电数据输入到多个基础预测模型得到多个预测充电时长，利用多个初始权重系数与多个预测充电时长加权得到一个最终的预测充电时长，一条特定的历史充电数据的条件下电子设备对应一个实际充电时长。这样，训练样本集中的一个样本为一组充电数据、该组充电数据输入到初始权重模型和多个基础预测模型得到的一个最终的预测充电时长以及该组充电数据对应的实际充电时长。
其中，一条历史充电数据的描述可以参考方法300的描述。
可选地，训练样本集中的样本可以是标有正样本标签的样本。具体地，如果一组充电数据中输入到初始权重模型和多个基础预测模型得到的一个预测充电时长与该组充电数据下的实际充电时长的差值的绝对值小于预设值，则该样本为正样本，可以作为训练样本集中的样本。
可选地，训练样本集中的样本除了具有正样本标签的之外，训练样本集中的样本对应的预测充电时长大于预设值，这样可以避免样本为用户刚充上电就拔掉的充电习惯产生的充电数据而产生的样本，影响基础预测模型训练的准确性。
S420，判断训练样本集的中样本数量是否达到预设的训练样本数量，如果训练样本集中的样本数量达到预设的训练样本数量，则执行S430，否则执行S410继续向训 练样本集中增加样本。
S430，根据训练样本集中的样本训练得到权重模型。
具体地，S430，包括：可以将训练样本集中的样本对应的一组充电数据输入一个初始权重模型中，初始权重模型输出多个权重系数，将该组充电数据输入方法300中训练得到的多个基础预测模型输出的多个预测充电时长，将多个权重系数与多个预测充电时长进行加权计算得到的最终预测充电时长与实际充电时长进行比较，根据比较结果反复修正初始权重模型的系数，直到多个权重系数与方法300中训练得到的基础预测模型输出的预测充电时长加权得到最终的预测充电时长与实际充电时长的精度长时间小于预设值，并持续一定的时间，则得到权重模型。
上述训练过程得到的权重模型可以为支持向量机(support vector machine，SVM)模型、决策树(decision tree)模型、神经网络(neural network)模型、套袋(Bagging)模型或者提升(boosting)模型。当然，权重模型还可以是其他的模型，本申请实施例对此不作限定。
在上述的方法300和方法400中，每组充电数据可以是预处理之前的充电数据，也可以是预处理之后的充电数据。如果每组充电数据是预处理之前的充电数据，则在S330和S430的训练模型的过程中，需要对充电数据进行预处理，预处理包括去噪声点、量化处理、归一化处理等操作中的至少一种标准化处理。当然，也可以不对充电数据进行预处理，有可能充电数据本身满足训练模型的要求，本申请实施例对此不作限制。
例如，以去噪声点为例，方法300和方法400中的训练样本集中的每组充电数据对应的预测充电时长需要大于预设时长，换句话说，训练样本集中有效的样本为用户的至少持续预设时长的充电行为，避免训练样本集中的样本为用户刚充上电就拔掉的充电习惯产生的样本，影响所训练的模型的准确性。
可选地，方法300和方法400中的训练样本集中的样本数量可以相同或者不同，本申请实施例对此不作限制。可选地方法300和方法400中的预设的训练样本数量可以相同或者不同，本申请实施例对此不作限制。可选地方法300和方法400中的训练样本集中的样本包括的充电参数可以相同或者不同，本申请实施例对此不作限制。
也需要理解的是，方法300和方法400是两个独立的实施例，方法400训练权重模型的过程所需的基础预测模型可以是根据方法300得到的，也可以是预设的多个基础预测模型，也可以是云端指示的多个基础预测模型，本申请实施例对此不作限制。
下面结合图6描述根据方法300得到的多个基础预测模型与方法400得到的权重模型描述用于充电管控的方法500，方法500包括：
S510，电子设备获取第一充电数据。
需要说明的是，第一充电数据可以是方法300和方法400中的一组充电数据。
可以分以下几种情况讨论第一充电数据。
情况一，电子设备收集了原始充电数据，原始充电数据满足S520的权重模型的输入和S530的多个基础预测模型的输入，在这种情况下，第一充电数据即为收集的原始充电数据。相应地，电子设备保存的是收集的原始充电数据。
情况二，电子设备收集了原始充电数据，原始充电数据不满足S520的权重模型的 输入和S530的多个基础预测模型的输入，在这种情况下，第一充电数据为对原始充电数据进行预处理之后的充电数据。相应地，电子设备保存的是预处理之后的充电数据。
情况三，电子设备收集了原始充电数据，原始充电数据不满足S520的权重模型的输入和S530的多个基础预测模型的输入，在这种情况下，电子设备可以保存收集的原始充电数据，电子设备可以对收集的原始充电数据进行预处理，得到预处理后的充电数据即为第一充电数据。可选地，电子设备可以保存第一充电数据。换句话说，在情况三中，电子设备可以只保存收集的原始充电数据，在需要时，对收集的原始充电数据进行预处理得到第一充电数据，或者电子设备可以既保存收集的原始数据也可以保存预处理得到的第一充电数据。
在上述的情况二和情况三中，对原始充电数据的预处理包括去噪声点、量化处理、归一化处理等操作中的至少一种标准化处理。
例如，以去噪声点为例，若第一充电数据包括充电器的历史插入时间和历史拔出时间，则充电器的历史拔出时间和历史插入时间之差大于预设时长，换句话说，第一充电数据为用户的至少持续预设时长的实际充电行为产生的充电数据，避免第一充电数据为用户刚充上电就拔掉的充电习惯产生的充电数据，影响预测的准确性。
S520，电子设备将第一充电数据中的至少部分充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长。
可选地，S530，包括：将第一充电数据中的至少部分充电数据输入到多个基础预测模型中，得到多个第三预测充电时长，每个基础预测模型能够输出一个第三预测充电时长，每个基础预测模型存在一个对应的调整参数，利用每个基础预测模型对应的调整参数调整每个基础预测模型输出的第三预测充电时长，从而得到第一预测充电时长。每个基础预测模型得到的第三预测充电时长可以相同或者不同。换句话说，在这个实施例中，每个基础预测模型输出的第一预测充电时长是根据调整参数调整之后的预测充电时长，这样，能够使得预测的准确性更高。
S530，电子设备将第一充电数据中的至少部分充电数据输入到权重模型中，得到多个权重系数。
可选地，多个权重系数中至少存在两个权重系数相同，或者多个权重系数中至少存在两个权重系数不同。
具体地，方法900中会描述如何得到多个基础预测模型中的对应的调整参数，为了避免赘述，在此不详细描述。
需要说明的是，S530中输入权重模型的至少部分充电数据与S230中输入多个基础预测模型的至少部分充电数据可以相同或者不同，本申请实施例对此不作限制。
也需要说明的是，S520中输入每个基础预测模型中的至少部分充电数据可以相同或者不同，本申请实施例对此不作限制。
可以理解的是，S520和S530的顺序没有任何限制，S520可以在S530之前或者之后或者同时进行，本申请实施例不予限制。
S540，电子设备根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，第二预测充电时长用于对电子设备的充电进行管控。
示例性地，第二预测充电时长可以是图1所示的t3-0，这样电子设备可以根据如 图1所示的管控策略进行管控。
具体地，S540包括：将多个权重系数分别与每个基础预测模型对应的第一预测充电时长进行加权计算，得到第二预测充电时长。示例性地，S530输出的L个权重系数分别为ω 1，ω 2，...，ω L,S520中L个基础预测模型输出的L个第一预测充电时长分别为K 1，K 2，...，K L，则第二预测充电时长为ω 1K 1+ω 2K 2+...+ω LK L，其中L为预设值。
在本申请实施例的预测充电时长的过程，电子设备根据权重模型得到多个权重系数。电子设备根据多个基础预测模型得到每个基础预测模型对应的第一预测充电时长，并根据多个权重系数和每个基础预测模型对应的第一预测充电时长确定第二预测时长，电子设备根据第二预测充电时长对电子设备的充电进行管控，从而可以延长电池的寿命并提高电池的续航能力，有助于提高用户体验。
可选地，上述方法500中的权重模型可以是预设的，例如，在电子设备出厂时电子设备的处理器中可以预设有权重模型。可选地，上述方法500中的权重模型可以是根据方法400训练得到。不管上述的权重模型是电子设备训练得到的还是预设的，在预测充电时长的过程中，都可以对权重模型进行修正。具体地，可以执行多次方法500，将多组充电数据中的至少部分数据输入方法300得到的多个基础预测模型和方法400得到的权重模型，输出多个第二预测充电时长，这样会产生多个样本，每个样本包括一组充电数据、该组充电数据对应的第二预测充电时长以及该组充电对应的实际充电时长，下面分两种情况描述修正权重模型的过程。
情况一，电子设备根据第一样本集中的样本修正权重模型。如图7所示，情况一具体包括：
S610，电子设备在方法500输出的多个样本中确定合格的样本。
具体地，一个样本中的第一实际充电时长与第二预测充电时长的差值的绝对值小于时长预设值，例如，时长预设值为60min，则该样本为合格的样本，也称为正样本，否则为不合格的样本，也称为负样本。
可选地，合格的样本除了需要满足第一实际充电时长与第二预测充电时长的差值的绝对值小于时长预设值之外，合格的样本的第二预测充电时长需要大于预设时长，换句话说，合格的样本为用户的至少持续预设时长的实际充电行为，避免有的样本为用户刚充上电就拔掉的充电习惯产生的，影响修正权重模型的准确性。
S620，电子设备确定样本当前权重模型下的第一合格率是否大于或等于第一合格率预设值，例如第一合格率预设值为95％，如果大于或等于第一合格率预设值则执行S630，否则执行到S640。
其中，第一合格率为合格的样本数量/总样本数量。
可以理解的是，S620中，确定样本的合格率是否达到第一合格率预设值可以替换为确定样本的不合格率是都达到不合格率预设值，例如不合格率预设值为5％，如果达到不合格率预设值在S640，如果没有达到不合格率预设值，则执行S630。
S630，电子设备不修正权重模型。
S640，电子设备确定正样本的数量是否大于第一样本数量预设值，例如第一样本数量预设值为500，如果大于第一样本数量预设值则执行S640，如果没有小于或等于第一样本数量预设值则返回S610，可以通过方法500继续向S610输入样本。
S650，电子设备利用合格的样本中的部分样本训练权重模型的第一修正参数，利用第一修正参数修正权重模型，得到修正后的权重模型。
示例性地，合格的样本数量为500个，则可以利用300个合格的样本训练权重模型的第一修正参数。
具体地，S650包括：可以将合格的样本中的每组充电数据执行方法500的过程，得到每组充电数据对应的第二预测充电时长和每组充电数据下的实际充电时长，比较第二预测充电时长和实际充电时长，根据比较结果反复确定权重模型的第一修正参数。
进一步地，电子设备利用合格的样本中的部分样本训练权重模型的第一修正参数时，可以将每组充电数据下的实际充电时长与每个基础预测模型输出的第一预测充电时长进行比较，使得第一修正参数修正权重模型后输出的多个权重系数中基础预测模型输出的第一预测充电时长中与实际充电时长越接近相应的基础预测模型的权重系数越高。举例来说，充电数据1下的实际充电时长为10小时，基础预测模型1输出的第一预测充电时长为9.5小时，基础预测模型2输出的第一预测充电时长为12小时，则第一修正参数修正权重模型后，权重模型输出的与基础预测模型1相乘的权重系数比与基础预测模型2相乘的权重系数为高，例如与基础预测模型1相乘的权重系数为0.8，与基础预测模型2相乘的权重系数为0.2。
S660，电子设备利用合格的样本中的剩余合格的样本测试修正后的权重模型的产生的样本的合格率，然后根据S620判断合格率。如果合格率满足合格率的预设值，则权重模型稳定，可以将方法500中的权重模型修正为根据图6得到的权重模型。
例如，合格的样本数量为500个，可以利用300个合格的样本训练权重模型的第一修正参数，利用剩余的200个合格的样本测试根据第一修正参数修正后的权重模型的稳定性。
上述图7的修正权重模型的方法，当不满足第一合格率的时，电子设备可以确定用于修正第一权重模型的第一修正参数，从而能够保证权重模型的准确性。
情况二，电子设备根据第一样本集中的样本得到第一修正参数，并将第一修正参数发送给云端，云端根据多个电子设备的第一修正参数拟合第二修正参数。具体步骤如图8所示。
S710-S750，同S610-S650。
多个电子设备中的每个电子设备执行完S710-S750之后，将每个电子设备得到的第一修正参数发送给云端。
S760，云端会接收到多个电子设备发送的多个第一修正参数，多个第一修正参数中的至少存在部分第一修正参数可以相同或者不同，远端根据多个电子设备发送的第一修正参数拟合生成第二修正参数，并将拟合生成的第二修正参数发送给电子设备。
S770，电子设备根据第二修正参数修正权重模型，得到修正后的权重模型。
S780，电子设备利用正样本中的剩余正样本测试修正后的权重模型的产生的样本的合格率，然后根据S720判断合格率。如果合格率满足合格率的预设值，则权重模型稳定，可选地，可以将方法500中的权重模型修正为根据图7得到的权重模型。
上述通过图8的修正权重模型的方法，可以简化电子设备修正模型的复杂度，每个电子设备将自身得到的第一修正参数发送给云端，第一修正参数为权重模型的修正 参数，没有任何的电子设备的用户的信息，也能保护用户的私密性，有利于提高安全性。同时云端利用大数据确定的第二修正参数，能满足鲁棒性的要求。
需要说明的是，图7和图8修正权重模型的方法中，不同的权重模型对应的第一修正参数不同，第一修正参数或第二修正参数用于调整权重模型的模型参数。
上述图4为训练多个基础预测模型的过程，图5为训练权重模型的过程，图6利用图4和图5训练的模型预测充电时长的过程，图7和图8为修正图5训练的权重模型的过程。类似地，也可以对图4得到的多个基础预测模型输出的预测时长进行修正，下面结合图9描述对多个基础预测模输出的预测时长进行修正的过程，图9可以是一个独立的实施例，也可以是结合前述方法的实施例，本申请实施例对此不作限定。
图9示出了本申请实施例的用于充电管控的方法800，包括：
S810，电子设备获取第一充电数据。
其中，第一充电数据可以为前述方法500的第一充电数据包括的充电参数，方法800中第一充电数据包括的充电参数的取值可以与方法500中的第一充电数据包括的充电参数的取值相同或者不同，为了避免赘述，在此不详细描述。
S820，电子设备将第一充电数据中的至少部分充电数据输入第一基础预测模型中，得到第三预测充电时长。
可选地，第一基础预测模型可以是预设的保存在电子设备中的，或者为云端根据方法300训练好指示给电子设备的，或者为电子设备自身根据方法300训练好的，本申请实施例对此不作限定。
S830，电子设备根据所述第一基础预测模型对应的调整参数调整第三预测充电时长，得到第一预测充电时长。
在一种可能的实现方式中，方法800是一个独立的实施例，则可以利用方法800得到的第一预测充电时长对电子设备的充电进行管控，由于第一基础预测模型输出的第三预测充电时长可能与实际充电时长相差较大，因此可以利用第一基础预测模型对应的调整参数调整第三预测充电时长，利用调整后的第一预测充电时长对电子设备的充电进行管控，从而可以提高预测的准确性。在这种情况下，方法800中的第一充电数据可以与前述实施例中的第一充电数据相同或不同，同样地，方法800第一预测充电时长可以与前述实施例中的第一预测充电时长相同或不同，本申请实施例不予限制。
在另一种可能的实现方式中，方法800与方法500的实施例可以结合，可以利用方法800得到第一基础预测模型输出的一个第一预测充电时长，可以通过方法800输出多个基础预测模型每个基础预测模型对应的一个第一预测充电时长，然后作为方法500中的S530的第一预测充电时长。
不管在上述的哪种实现方式下，电子设备都需要获取多个基础预测模型中每个基础预测模型的调整参数，下面具体结合图10描述如何获取第一基础预测模型对应的调整参数，图10仅以获取第一基础预测模型对应的调整参数为例描述，其他的基础预测模型对应的调整参数与获取第一基础预测模型对应的调整参数类似，为了避免赘述，本申请实施例对比不小一一列举。
电子设备获取第二充电数据和第二充电数据下对应的实际充电时长，电子设备将第二充电数据输入到第一基础预测模型中，输出第四预测充电时长，其中，第一基础 预测模型可以是根据方法300训练得到的。电子设备将第二充电数据、第四预测充电时长以及第二实际充电时长作为一个样本添加到第二样本集，以此类推确定第二样本集中的样本。
S910，电子设备确定第二样本集中的合格的样本。
具体地，一个样本中的第二实际充电时长与第四预测充电时长的差值的绝对值小于时长预设值，例如，时长预设值为60min，则该样本为合格的样本，也称为正样本，否则为不合格的样本，也称为负样本。
可选地，合格的样本除了需要满足第二实际充电时长与第四预测充电时长的差值的绝对值小于时长预设值之外，合格的样本的第四预测充电时长需要大于预设值，换句话说，合格的样本为用户的至少持续预设时长的实际充电行为，避免有的样本为用户刚充上电就拔掉的充电习惯产生的，影响准确性。
S920，电子设备确定样本的第二合格率是否大于或等于第二合格率预设值，如果第二合格率大于或等于第二合格率预设值，则执行S930，否则执行S940。
其中，第二合格率为合格的样本数量/总样本数量。
可以理解的是，S920中，确定样本的合格率是否达到第二合格率预设值可以替换为确定样本的不合格率是否达到不合格率预设值，例如不合格率预设值为5％，如果达到不合格率预设值在S940，如果没有达到不合格率预设值，则到S930。
S930，电子设备不更新第一基础预测模型对应的调整参数。
可以理解的是初始的调整参数可以是预设的值，或者根据历史经验数据得到的值，本申请实施例不予限制。
S940，电子设备确定合格的样本数量是否大于第二样本数量预设值，若大于，则执行S950，否则执行S910，确定是否有新的合格的样本。
S950，电子设备利用部分合格的样本训练第一基础预测模型的调整参数。
在一种可能的实现方式中，如果部分合格的样本中的第二实际充电时长和第四预测充电时长满足线性关系，则利用部分合格的样本进行线性拟合得到线性关系的回归系数和常数即为第一基础预测模型的调整参数。
例如，第二实际充电时长为y，第四预测充电时长为x，若x和y满足y＝kx+b，则S950得到的回归系数k的取值，常数b的取值。
在另一种可能的实现方式中，如果部分合格的样本中的第二实际充电时长和第四预测充电时长满足非线性关系，则利用部分合格的样本进行非线性训练，得到非线性关系的系数即为第一基础预测模型的调整参数。
S960，电子设备更新第一基础预测模型的调整参数。
S970，电子设备利用剩余部分合格样本测试更新后的调整参数的稳定性。
示例性地，如果部分合格的样本中的第二实际充电时长和第四预测充电时长满足线性关系，具体地，第二实际充电时长为y，第四预测充电时长为x，若x和y满足y＝kx+b，则上述k，b满足试验阈值[0<k<M且b2<N]后，表示调整参数k，b稳定，其中M和N为预设的值。
本申请实施例提供的用于充电管控的方法，可以利用多个权重系数和每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，如图11所示，示出了某一电 子设备的n(n为正整数)次的实际充电时长和每次预测的第二预测充电时长的对比示意图，可以看出，本申请实施例提供的预测充电时长接近电子设备真实充电时长。如图12所示，当电子设备充电开始时间为晚上22:00，如果不针对该电子设备的电池的充电进行管控，图12的上面一列示意图中，当在23:00时，电子设备的电池将要充满，在00:00点电池进入膨胀状态，直到第二天早晨08:00点充电结束。这样会减少电池的使用寿命。在图12的下面一列示意图中，通过本申请实施例预测得到电子设备的充电时长为10小时或者接近10小时之后，当电池时间到23:00时，可以对电池进行充电管控，23:00～06:00时间段进入保护状态，不对电子设备进行充电，在06:00之后继续充电，直到08:00达到100％，从而降低电池处于膨胀状态的时间，有利于提高电池的使用寿命。
需要说明的是，在本申请实施例中，“大于”可以替换为“大于或等于”，“小于或等于”可以替换为“小于”，或者，“大于或等于”可以替换为“大于”，“小于”可以替换为“小于或等于”。例如，大于A执行步骤x，小于或等于A执行步骤y，可以替换为：大于或等于A执行步骤x，小于A执行步骤y，换句话说，在等于A时，可以执行步骤x也可以执行步骤y，本申请实施例不予限制。
本文中描述的各个实施例可以为独立的方案，也可以根据内在逻辑进行组合，这些方案都落入本申请的保护范围中。
可以理解的是，上述各个方法实施例中由电子设备实现的方法和操作，也可以由可用于电子设备的部件(例如芯片或者电路)实现。
上文描述了本申请提供的方法实施例，下文将描述本申请提供的装置实施例。应理解，装置实施例的描述与方法实施例的描述相互对应，因此，未详细描述的内容可以参见上文方法实施例，为了简洁，这里不再赘述。
本领域技术人员应该可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的保护范围。
本申请实施例可以根据上述方法示例，对电子设备进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有其它可行的划分方式。下面以采用对应各个功能划分各个功能模块为例进行说明。
图13为本申请实施例提供的用于充电管控的装置1000的示意性框图。该装置1000包括获取单元1010和处理单元1020。获取单元1010可以与外部进行通信。获取单元1010还可以称为通信接口或通信单元，获取单元1010用于执行上文图6-图8或者图10的实施例中电子设备侧的获取或收发相关的操作。处理单元1020用于进行数据处理，处理单元1020用于执行上文图6-图8或者图10的实施例中电子设备侧的处理相关的操作。
获取单元1010，用于获取第一充电数据；
处理单元1020，用于将所述第一充电数据中的至少部分充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长；
所述处理单元1020还用于将所述第一充电数据中的至少部分充电数据输入到权重模型中，得到多个权重系数；
所述处理单元1020还用于根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，所述第二预测充电时长用于对电子设备的充电进行管控。
作为一个可选实施例，所述获取单元1010还用于：获取所述第一充电数据对应的所述电子设备的第一实际充电时长；
所述处理单元1020还用于：
将所述第一充电数据、所述第一实际充电时长和所述第二预测充电时长作为样本添加到第一样本集；
根据所述第一样本集中的样本更新所述权重模型。
作为一个可选实施例，所述处理单元1020具体用于：
确定所述第一样本集中的样本的第一合格率；
若所述第一合格率小于第一合格率预设值，确定所述第一样本集中的合格的样本的数量；
若所述第一样本集中的合格的样本的数量大于第一样本数量预设值，根据所述合格的样本中部分样本修正所述权重模型。
作为一个可选实施例，所述处理单元1020具体用于：
根据所述部分样本训练得到第一修正参数，根据所述第一修正参数修正所述权重模型，得到修正后的权重模型；或者包括：
根据所述部分样本训练得到所述第一修正参数；
所述装置1000还包括：
收发单元，用于将所述第一修正参数发送给云端，并且接收云端根据所述第一修正参数确定的第二修正参数；
所述处理单元1020还用于根据所述第二修正参数修正所述权重模型，得到修正后的权重模型。
作为一个可选实施例，所述处理单元1020还用于：根据所述合格的样本中的剩
余部分样本测试所述修正后的权重模型的稳定性。
作为一个可选实施例，所述处理单元1020具体用于：将所述第一充电数据中的至少部分充电数据输入到所述多个基础预测模型中，得到每个基础预测模型对应的第三预测充电时长；
利用所述每个基础预测模型的调整参数调整所述每个基础预测模型对应的第三预测充电时长，得到所述每个基础预测模型对应第一预测充电时长。
作为一个可选实施例，所述获取单元1010还用于：获取所述电子设备的第二充电数据和所述第二充电数据对应的第二实际充电时长；
所述处理单元1020还用于：
将所述第二充电数据输入到所述多个基础预测模型的第一基础预测模型中，得到第四预测充电时长；
将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；
根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
作为一个可选实施例，所述处理单元1020具体用于：
确定所述第二样本集中的样本的第二合格率；
若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本的数量；
若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
作为一个可选实施例，若所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
作为一个可选实施例，所述第一充电数据和第二充电数据包括以下至少一项：用于对所述电子设备进行充电的充电器的类型、生产所述电子设备的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述电子设备进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
图14为本申请实施例提供的另一用于充电管控的装置1100的示意性框图。该装置1100包括获取单元1110和处理单元1120。获取单元1110可以与外部进行通信。获取单元1110还可以称为通信接口或通信单元，获取单元1110用于执行上文图9或者图10的实施例中电子设备侧的获取或收发相关的操作。处理单元1120用于进行数据处理，处理单元1120用于执行上文图9或者图10的实施例中电子设备侧的处理相关的操作。
其中，获取单元1110用于获取第一充电数据。处理单元1120用于：将所述第一充电数据中的至少部分充电数据输入第一基础预测模型中，得到第三预测时长；根据所述第一基础预测对应的调整参数调整所述第三预测充电时长，得到第一预测充电时长，所述第一预测充电时长用于对电子设备的充电进行管控。
作为一个可选实施例，所述获取单元1110还用于：获取所述装置的第二充电数据和所述第二充电数据对应的第二实际充电时长。所述处理单元1120还用于：将所述第二充电数据输入到所述第一基础预测模型中，得到第四预测充电时长；将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
作为一个可选实施例，所述处理单元1120具体用于：
确定所述第二样本集中的样本的第二合格率；
若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本 的数量；
若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
作为一个可选实施例，若所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
作为一个可选实施例，所述第一充电数据和第二充电数据包括以下至少一项：用于对所述装置进行充电的充电器的类型、生产所述装置的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述装置进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
图15是本申请实施例提供的用于充电管控的装置1200的结构性示意性图。所述通信装置1200包括：处理器1210、存储器1220、通信接口1230、总线1240。
在一种可能的实现方式中，图15所示的装置1200中的处理器1210可以对应于图13中的装置1000中的处理单元1020。图15所示的装置1200中的通信接口1230可以对应于图13中的装置1000中的获取单元1010。
在一种可能的实现方式中，图15所示的装置1200中的处理器1210可以对应于图14中的装置1100中的处理单元1120。图15所示的装置1200中的通信接口1230可以对应于图14中的装置1100中的获取单元1110。
其中，该处理器1210可以与存储器1220连接。该存储器1220可以用于存储该程序代码和数据。因此，该存储器1220可以是处理器1210内部的存储单元，也可以是与处理器1210独立的外部存储单元，还可以是包括处理器1210内部的存储单元和与处理器1210独立的外部存储单元的部件。
可选的，装置1200还可以包括总线1240。其中，存储器1220、通信接口1230可以通过总线1240与处理器1210连接。总线1240可以是外设部件互连标准(peripheral component interconnect，PCI)总线或扩展工业标准结构(extended industry standard architecture，EISA)总线等。所述总线1240可以分为地址总线、数据总线、控制总线等。为便于表示，图15中仅用一条线表示，但并不表示仅有一根总线或一种类型的总线。
应理解，在本申请实施例中，该处理器1210可以采用中央处理单元(central processing unit，CPU)。该处理器还可以是其它通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate Array，FPGA)或者其它可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。或者该处理器1210采用一个或多个集成电路，用于执行相关程序，以实现本申请实施例所提供的技术方案。
该存储器1220可以包括只读存储器和随机存取存储器，并向处理器810提供指令 和数据。处理器810的一部分还可以包括非易失性随机存取存储器。例如，处理器810还可以存储设备类型的信息。
在装置1200运行时，所述处理器1210执行所述存储器1220中的计算机执行指令以通过所述装置1200执行上述方法的操作步骤。
应理解，根据本申请实施例的装置1200可对应于本申请实施例中的装置1000和装置1100，并且装置1000和装置1100中的各个单元的上述和其它操作和/或功能分别为了实现方法的相应流程，为了简洁，在此不再赘述。
可选地，在一些实施例中，本申请实施例还提供了一种计算机可读介质，所述计算机可读介质存储有程序代码，当所述计算机程序代码在计算机上运行时，使得计算机执行上述各方面中的方法。
可选地，在一些实施例中，本申请实施例还提供了一种计算机程序产品，所述计算机程序产品包括：计算机程序代码，当所述计算机程序代码在计算机上运行时，使得计算机执行上述各方面中的方法。
在本申请实施例中，终端设备或网络设备包括硬件层、运行在硬件层之上的操作系统层，以及运行在操作系统层上的应用层。其中，硬件层可以包括中央处理器(central processing unit，CPU)、内存管理单元(memory management unit，MMU)和内存(也称为主存)等硬件。操作系统层的操作系统可以是任意一种或多种通过进程(process)实现业务处理的计算机操作系统，例如，Linux操作系统、Unix操作系统、Android操作系统、iOS操作系统或windows操作系统等。应用层可以包含浏览器、通讯录、文字处理软件、即时通信软件等应用。
本申请实施例并未对本申请实施例提供的方法的执行主体的具体结构进行特别限定，只要能够通过运行记录有本申请实施例提供的方法的代码的程序，以根据本申请实施例提供的方法进行通信即可。例如，本申请实施例提供的方法的执行主体可以是终端设备或网络设备，或者，是终端设备或网络设备中能够调用程序并执行程序的功能模块。
本申请的各个方面或特征可以实现成方法、装置或使用标准编程和/或工程技术的制品。本文中使用的术语“制品”可以涵盖可从任何计算机可读器件、载体或介质访问的计算机程序。例如，计算机可读介质可以包括但不限于：磁存储器件(例如，硬盘、软盘或磁带等)，光盘(例如，压缩盘(compact disc，CD)、数字通用盘(digital versatile disc，DVD)等)，智能卡和闪存器件(例如，可擦写可编程只读存储器(erasable programmable read-only memory，EPROM)、卡、棒或钥匙驱动器等)。
本文描述的各种存储介质可代表用于存储信息的一个或多个设备和/或其它机器可读介质。术语“机器可读介质”可以包括但不限于：无线信道和能够存储、包含和/或承载指令和/或数据的各种其它介质。
应理解，本申请实施例中提及的处理器可以是中央处理单元(central processing unit，CPU)，还可以是其他通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以 是任何常规的处理器等。
还应理解，本申请实施例中提及的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)。例如，RAM可以用作外部高速缓存。作为示例而非限定，RAM可以包括如下多种形式：静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synchlink DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
需要说明的是，当处理器为通用处理器、DSP、ASIC、FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件时，存储器(存储模块)可以集成在处理器中。
还需要说明的是，本文描述的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的保护范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。此外，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上， 或者说对现有技术做出贡献的部分，或者该技术方案的部分，可以以计算机软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，该计算机软件产品包括若干指令，该指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。前述的存储介质可以包括但不限于：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中在本申请的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本申请。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (22)

  1. 一种用于充电管控的方法，其特征在于，包括：
    获取第一充电数据；
    将所述第一充电数据中的至少部分充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长；
    将所述第一充电数据中的至少部分充电数据输入到权重模型中，得到多个权重系数；
    根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，所述第二预测充电时长用于对电子设备的充电进行管控。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    获取所述第一充电数据对应的所述电子设备的第一实际充电时长；
    将所述第一充电数据、所述第一实际充电时长和所述第二预测充电时长作为样本添加到第一样本集；
    根据所述第一样本集中的样本更新所述权重模型。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述第一样本集中的样本更新所述权重模型，包括：
    确定所述第一样本集中的样本的第一合格率；
    若所述第一合格率小于第一合格率预设值，确定所述第一样本集中的合格的样本的数量；
    若所述第一样本集中的合格的样本的数量大于第一样本数量预设值，根据所述合格的样本中部分样本修正所述权重模型。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述合格的样本中的部分样本修正所述权重模型，包括：
    根据所述部分样本训练得到第一修正参数，根据所述第一修正参数修正所述权重模型，得到修正后的权重模型；或者包括：
    根据所述部分样本训练得到所述第一修正参数，并将所述第一修正参数发送给云端；
    接收云端根据所述第一修正参数确定的第二修正参数；
    根据所述第二修正参数修正所述权重模型，得到修正后的权重模型。
  5. 根据权利要求4所述的方法，其特征在于，所述方法还包括：
    根据所述合格的样本中的剩余部分样本测试所述修正后的权重模型的稳定性。
  6. 根据权利要求1至5中任一项所述的方法，其特征在于，所述将所述第一充电数据中的至少部分充电数据输入到多个基础预测模型中，得到每个基础预测模型对应的第一预测充电时长，包括：
    将所述第一充电数据中的至少部分充电数据输入到所述多个基础预测模型中，得到每个基础预测模型对应的第三预测充电时长；
    利用所述每个基础预测模型的调整参数调整所述每个基础预测模型对应的第三预测充电时长，得到所述每个基础预测模型对应第一预测充电时长。
  7. 根据权利要求6所述的方法，其特征在于，所述方法还包括：
    获取所述电子设备的第二充电数据和所述第二充电数据对应的第二实际充电时长；
    将所述第二充电数据输入到所述多个基础预测模型的第一基础预测模型中，得到第四预测充电时长；
    将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；
    根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
  8. 根据权利要求7所述的方法，其特征在于，所述根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数，包括：
    确定所述第二样本集中的样本的第二合格率；
    若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本的数量；
    若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
  9. 根据权利要求8所述的方法，其特征在于，若所述第二样本集中合格的样本对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
  10. 根据权利要求1至9中任一项所述的方法，其特征在于，所述第一充电数据和第二充电数据包括以下至少一项：
    用于对所述电子设备进行充电的充电器的类型、生产所述电子设备的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述电子设备进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
  11. 一种用于充电管控的装置，其特征在于，包括：
    获取单元，用于获取第一充电数据；
    处理单元，用于将所述第一充电数据中的至少部分充电数据输入到多个基础预测模型中，确定每个基础预测模型对应的第一预测充电时长；
    所述处理单元还用于将所述第一充电数据中的至少部分充电数据输入到权重模型中，得到多个权重系数；
    所述处理单元还用于根据所述多个权重系数和所述每个基础预测模型对应的第一预测充电时长确定第二预测充电时长，所述第二预测充电时长用于对所述装置的充电进行管控。
  12. 根据权利要求11所述的装置，其特征在于，所述获取单元还用于：
    获取所述第一充电数据对应的所述装置的第一实际充电时长；
    所述处理单元还用于：
    将所述第一充电数据、所述第一实际充电时长和所述第二预测充电时长作为样本添加到第一样本集；
    根据所述第一样本集中的样本更新所述权重模型。
  13. 根据权利要求12所述的装置，其特征在于，所述处理单元具体用于：
    确定所述第一样本集中的样本的第一合格率；
    若所述第一合格率小于第一合格率预设值，确定所述第一样本集中的合格的样本的数量；
    若所述第一样本集中的合格的样本的数量大于第一样本数量预设值，根据所述合格的样本中部分样本修正所述权重模型。
  14. 根据权利要求13所述的装置，其特征在于，所述处理单元具体用于：
    根据所述部分样本训练得到第一修正参数，根据所述第一修正参数修正所述权重模型，得到修正后的权重模型；或者包括：
    根据所述部分样本训练得到所述第一修正参数；
    所述装置还包括：
    收发单元，用于将所述第一修正参数发送给云端，并且接收云端根据所述第一修正参数确定的第二修正参数；
    所述处理单元还用于根据所述第二修正参数修正所述权重模型，得到修正后的权重模型。
  15. 根据权利要求14所述的装置，其特征在于，所述处理单元还用于：
    根据所述合格的样本中的剩余部分样本测试所述修正后的权重模型的稳定性。
  16. 根据权利要求11至15中任一项所述的装置，其特征在于，所述处理单元具体用于：
    将所述第一充电数据中的至少部分充电数据输入到所述多个基础预测模型中，得到每个基础预测模型对应的第三预测充电时长；
    利用所述每个基础预测模型的调整参数调整所述每个基础预测模型对应的第三预测充电时长，得到所述每个基础预测模型对应第一预测充电时长。
  17. 根据权利要求16所述的装置，其特征在于，所述获取单元还用于：
    获取所述装置的第二充电数据和所述第二充电数据对应的第二实际充电时长；
    所述处理单元还用于：
    将所述第二充电数据输入到所述多个基础预测模型的第一基础预测模型中，得到第四预测充电时长；
    将所述第四预测充电时长、所述第二充电数据和所述第二实际充电时长作为样本添加到第二样本集；
    根据所述第二样本集中的样本确定所述第一基础预测模型对应的调整参数。
  18. 根据权利要求17所述的装置，其特征在于，所述处理单元具体用于：
    确定所述第二样本集中的样本的第二合格率；
    若所述第二合格率小于第二合格率预设值，确定所述第二样本集中的合格的样本的数量；
    若所述第二样本集中的合格的样本数量大于第二样本数量预设值，根据合格的样本确定所述第一基础预测模型对应的调整参数。
  19. 根据权利要求18所述的装置，其特征在于，若所述第二样本集中合格的样本 对应的实际充电时长和预测充电时长满足线性关系，所述第一基础预测模型对应的调整参数为所述线性关系的回归系数和常数。
  20. 根据权利要求11至19中任一项所述的装置，其特征在于，所述第一充电数据和第二充电数据包括以下至少一项：
    用于对所述装置进行充电的充电器的类型、生产所述装置的电池的厂家、所述电池的标称容量、所述电池的电芯类型、用于对所述装置进行充电的充电线的类型、所述电池已经被充电的循环次数、所述电池能够充电的标称循环次数、所述电池的平均内阻、所述电池的最大内阻、所述充电器的历史插入时间和历史拔出时间、所述电池的电芯的充电截止时间、所述电池的电芯的起始电量和终止电量、在预设天数内每天的预设时间段的实际充电时长。
  21. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机程序，当所述计算机程序被运行时，实现如权利要求1至10中任一项所述的方法。
  22. 一种芯片，包括处理器，所述处理器与存储器相连，所述存储器用于存储计算机程序，所述处理器用于执行所述存储器中存储的计算机程序，以使得所述芯片执行如权利要求1至10中任一项所述的方法。
PCT/CN2021/080028 2020-04-30 2021-03-10 用于充电管控的方法和装置 WO2021218423A1 (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202010361658.0 2020-04-30
CN202010361658.0A CN111564877A (zh) 2020-04-30 2020-04-30 用于充电管控的方法和装置
CN202011463325.5 2020-12-10
CN202011463325.5A CN113595164B (zh) 2020-04-30 2020-12-10 用于充电管控的方法和装置

Publications (1)

Publication Number Publication Date
WO2021218423A1 true WO2021218423A1 (zh) 2021-11-04

Family

ID=72071708

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/080028 WO2021218423A1 (zh) 2020-04-30 2021-03-10 用于充电管控的方法和装置

Country Status (2)

Country Link
CN (2) CN111564877A (zh)
WO (1) WO2021218423A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116224091A (zh) * 2022-12-01 2023-06-06 伏瓦科技(苏州)有限公司 电池的电芯故障检测方法、装置、电子设备及存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111564877A (zh) * 2020-04-30 2020-08-21 华为技术有限公司 用于充电管控的方法和装置

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101142732A (zh) * 2005-03-16 2008-03-12 Lg化学株式会社 用于估算电池充电状态的系统和方法
CN104584376A (zh) * 2012-08-30 2015-04-29 德克萨斯仪器股份有限公司 针对预定义充电持续时间在积分退化全局最小化下对电池进行充电的方法和装置
CN104956233A (zh) * 2013-02-01 2015-09-30 三洋电机株式会社 电池状态推断装置
US20170054317A1 (en) * 2015-08-20 2017-02-23 International Business Machines Corporation Modeling a change in battery degradation
CN107677962A (zh) * 2016-08-01 2018-02-09 现代自动车株式会社 基于充电所需时间来管理电池的系统和方法
CN109361818A (zh) * 2018-10-26 2019-02-19 深圳壹账通智能科技有限公司 充电提醒方法、装置、存储介质和终端
CN110010987A (zh) * 2019-04-12 2019-07-12 苏州正力蔚来新能源科技有限公司 一种基于大数据的电动汽车剩余充电时间预测方法
CN110674569A (zh) * 2019-08-30 2020-01-10 蜂巢能源科技有限公司 基于决策树的充电剩余时间估算方法及系统
CN111564877A (zh) * 2020-04-30 2020-08-21 华为技术有限公司 用于充电管控的方法和装置

Family Cites Families (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000287378A (ja) * 1999-03-29 2000-10-13 Sony Corp 充電装置、充電方法および電子機器
JP2006238598A (ja) * 2005-02-24 2006-09-07 Seiko Epson Corp 充電時間算出装置
JP2012151946A (ja) * 2011-01-17 2012-08-09 Alpine Electronics Inc 充電装置
JPWO2013115244A1 (ja) * 2012-02-01 2015-05-11 日本電気株式会社 充電時間予測装置、充電時間予測方法、及びプログラム
US20130231894A1 (en) * 2012-03-01 2013-09-05 Nokia Corporation Method and apparatus for providing battery information
US20150123595A1 (en) * 2013-11-04 2015-05-07 Xiam Technologies Limited Intelligent context based battery charging
US20150349385A1 (en) * 2014-04-01 2015-12-03 Medtronic, Inc. Method and System for Predicting Useful Life of a Rechargeable Battery
US9210662B1 (en) * 2014-05-29 2015-12-08 Apple Inc. Adaptive battery life extension
CN106602158B (zh) * 2015-10-14 2020-01-07 上海汽车集团股份有限公司 电池充电时间预测方法及装置
WO2018012055A1 (ja) * 2016-07-13 2018-01-18 ソニーモバイルコミュニケーションズ株式会社 情報処理装置、情報処理システム、充電方法
JP2018068081A (ja) * 2016-10-21 2018-04-26 京セラ株式会社 二次電池の充電時間の予測装置および予測方法
US20180115170A1 (en) * 2016-10-24 2018-04-26 International Business Machines Corporation Scheduling optimized charging of multiple battery enabled devices available to a group of users
CN107169588A (zh) * 2017-04-12 2017-09-15 中国电力科学研究院 一种电动汽车充电站短时功率预测方法及系统
WO2018215864A1 (ja) * 2017-05-22 2018-11-29 株式会社半導体エネルギー研究所 充電制御システム、及び充電制御装置
JP6760605B2 (ja) * 2017-10-13 2020-09-23 Ｎｅｃプラットフォームズ株式会社 制御装置、電子機器、制御方法、及び制御プログラム
JP7159195B2 (ja) * 2017-11-16 2022-10-24 株式会社半導体エネルギー研究所 二次電池の寿命推定装置
US10958082B2 (en) * 2018-04-25 2021-03-23 Microsoft Technology Licensing, Llc Intelligent battery cycling for lifetime longevity
CN110460118A (zh) * 2018-05-07 2019-11-15 畅充科技（上海）有限公司 充电方法、装置及终端充电设备
DE102018005797A1 (de) * 2018-07-23 2019-03-07 Daimler Ag Verfahren zum Abschätzen der für eine Vollladung einer wiederaufladbaren, elektrochemischen Energiespeichervorrichtung erforderlichen Zeitdauer
CN110046787A (zh) * 2019-01-15 2019-07-23 重庆邮电大学 一种城市区域电动汽车充电需求时空预测方法
CN110429672B (zh) * 2019-07-15 2021-06-08 百富计算机技术（深圳）有限公司 电池充电管理方法、终端设备及存储介质
CN110751314B (zh) * 2019-09-16 2022-05-31 上海电力大学 考虑用户充电行为特征数据驱动的电动汽车负荷预测方法
CN111055722B (zh) * 2019-12-20 2023-11-17 华为技术有限公司 一种预估充电时间的方法，装置及存储介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101142732A (zh) * 2005-03-16 2008-03-12 Lg化学株式会社 用于估算电池充电状态的系统和方法
CN104584376A (zh) * 2012-08-30 2015-04-29 德克萨斯仪器股份有限公司 针对预定义充电持续时间在积分退化全局最小化下对电池进行充电的方法和装置
CN104956233A (zh) * 2013-02-01 2015-09-30 三洋电机株式会社 电池状态推断装置
US20170054317A1 (en) * 2015-08-20 2017-02-23 International Business Machines Corporation Modeling a change in battery degradation
CN107677962A (zh) * 2016-08-01 2018-02-09 现代自动车株式会社 基于充电所需时间来管理电池的系统和方法
CN109361818A (zh) * 2018-10-26 2019-02-19 深圳壹账通智能科技有限公司 充电提醒方法、装置、存储介质和终端
CN110010987A (zh) * 2019-04-12 2019-07-12 苏州正力蔚来新能源科技有限公司 一种基于大数据的电动汽车剩余充电时间预测方法
CN110674569A (zh) * 2019-08-30 2020-01-10 蜂巢能源科技有限公司 基于决策树的充电剩余时间估算方法及系统
CN111564877A (zh) * 2020-04-30 2020-08-21 华为技术有限公司 用于充电管控的方法和装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116224091A (zh) * 2022-12-01 2023-06-06 伏瓦科技(苏州)有限公司 电池的电芯故障检测方法、装置、电子设备及存储介质
CN116224091B (zh) * 2022-12-01 2024-02-02 伏瓦科技(苏州)有限公司 电池的电芯故障检测方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
CN113595164A (zh) 2021-11-02
CN111564877A (zh) 2020-08-21
CN113595164B (zh) 2023-05-16

Similar Documents

Publication Publication Date Title
US11165270B2 (en) Predictive management of battery operation
WO2021218423A1 (zh) 用于充电管控的方法和装置
US20220043067A1 (en) Charging method and device, charging system, electronic equipment and storage medium
CN109324902B (zh) 一种调整移动终端工作频率的方法、移动终端及存储介质
US20190356137A1 (en) Rule based smart charging
US11678362B2 (en) Method to dynamically change the minimum candidate resources ratio in mode 2 resource selection procedure of NR V2X
US11619679B2 (en) Method, apparatus and medium for estimating battery remaining life
CN111416412B (zh) 一种充电控制方法、装置及终端设备
EP4125028A1 (en) Method and system to generate pricing for charging electric vehicles
CN116388112B (zh) 异常供应端断电方法、装置、电子设备和计算机可读介质
CN108038050A (zh) 性能调整方法、装置、存储介质及电子设备
CN112463391A (zh) 内存控制方法、内存控制装置、存储介质与电子设备
CN116315185A (zh) 电子设备及电池管理方法
CN112654877B (zh) 一种充电检测方法、充电检测装置及终端设备
CN113806656A (zh) 确定用户的特征的方法、设备及计算机可读介质
CN115513545A (zh) 电量控制方法、装置、计算机可读存储介质及设备
CN111049219B (zh) 一种充电方法及充电设备
CN113784251A (zh) Tws耳机、充电盒、充电方法及系统
CN109149676B (zh) 一种控制充电电流的方法及装置
CN112035324A (zh) 批量作业执行情况监控方法及装置
US20220029446A1 (en) Method and apparatus for charging, device, and storage medium
CN112766698B (zh) 应用业务压力确定方法及装置
CN117093291A (zh) 电池健康度的界面更新方法、装置、电子设备及存储介质
CN115827650A (zh) 确定电池使用参数的方法、装置及计算机存储介质
CN118093097A (zh) 数据存储集群资源调度方法、装置、电子设备和介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21797350

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21797350

Country of ref document: EP

Kind code of ref document: A1