A fast charging control method and system

Abstract

The present application discloses a fast charging control method and system, which relates to the field of intelligent control. It first sets the initial threshold parameters according to the output power of the charger, and further introduces data processing technology based on deep learning to conduct a comprehensive analysis of the battery operating temperature and user preference settings of the charging device, optimizes and adjusts the initial threshold parameters, and then based on the adjusted threshold parameters, limits the current parameters corresponding to the gears whose power consumption exceeds the threshold parameters in all the charging power gear information of the charger, and forms new gear information to provide to the charged main device, so that it can select a suitable charging gear for charging, so as to realize intelligent fast charging control. The present application dynamically adjusts the charging power gear by real-time monitoring of the battery temperature of the charging device and the user's usage habits, which can effectively avoid battery damage caused by overheating, while improving charging efficiency and user experience, and ensuring that the charging process is both efficient and safe.

Description

Technical Field

[0001] The present application relates to the field of intelligent control, and more specifically, to a fast charging control method and system. Background Art

[0002] In today's era of rapid development of electronic devices, fast charging technology has become an indispensable part of people's lives. With the increasing power of portable electronic devices such as smartphones, tablets, and laptops, their demand for power is also increasing, and the importance of fast charging technology is becoming more and more prominent. Traditional charging technologies, such as USB charging technology, have relatively low charging efficiency and are difficult to meet users' urgent needs for rapid replenishment of power. PD fast charging technology came into being, which has improved the charging efficiency to a certain extent and has been widely used. At present, the conventional PD fast charging power covers a variety of specifications such as 15W, 27W, 45W, 65W and even 100W or more. When a device that supports fast charging is connected to a charger that also supports fast charging, in a conventional charging environment, that is, when the charger directly charges the charged device, a handshake protocol will be performed between the two. The charger informs the charged device of its power supply capacity, and the charged device judges and requests the power supply power from the charger based on this. After the two parties reach a handshake, fast charging is achieved.

[0003] However, when the charged device integrates PD charging function and other extended functions (implemented by another extended device), the charger may continue to increase the output power until it works at high load or even full load. In serious cases, it may exceed its output threshold and eventually fail to charge. In this regard, the invention patent with publication number CN116647001A proposes a fast charging control method, which sets a fixed voltage, current and power limit according to the maximum output power of the charger, so as to adjust the original charging gear information of the charger to ensure that the charging process is carried out within a safe range.

[0004] Although the existing technology has solved the charger overload problem to a certain extent, when adjusting the charging gear information, it is mainly based on the design of static threshold parameters, and fails to fully consider the user's usage habits and the working status of the device battery. There may be certain safety hazards, or the adjusted charging gear information may deviate from the user's actual needs and be unacceptable to the user, thereby affecting the charging efficiency and user experience.

[0005] Therefore, an optimized fast charging control method and system are needed to solve the above technical problems. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a fast charging control method and system, which first sets the initial threshold parameters according to the output power of the charger, and further introduces data processing technology based on deep learning to comprehensively analyze the battery operating temperature and user preference settings of the charging device, optimizes and adjusts the initial threshold parameters, and then based on the adjusted threshold parameters, adjusts the current parameters corresponding to the gears whose power consumption exceeds the threshold parameters in all the charging power gear information of the charger, and forms new gear information to provide to the charged main device, so that it can select a suitable charging gear for charging, so as to realize intelligent fast charging control. By dynamically adjusting the charging power gear by real-time monitoring of the battery temperature of the charging device and the user's usage habits, it can effectively avoid battery damage caused by overheating, while improving charging efficiency and user experience, ensuring that the charging process is both efficient and safe.

[0007] According to one aspect of the present application, a fast charging control method and system are provided, which include:

[0008] An initial threshold parameter is set according to the output power of the charger, and the initial threshold parameter is dynamically adjusted based on the working conditions and user preference settings to obtain an adjusted threshold parameter; the gear information of the charging power of the charger is obtained and recorded as the first gear information, the first gear information is compared with the adjusted threshold parameter in sequence, and the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter is modified to form the second gear information; the charged main device receives the second gear information, and sends a power request information according to the received second gear information; the power request information is parsed, the charging gear required by the charged main device is selected, and the selected charging gear information is returned to the charger; the charger receives the charging gear information and starts charging the charged main device.

[0009] Specifically, the working condition is a time queue of the battery temperature of the charging device.

[0010] Specifically, the initial threshold parameters are dynamically adjusted based on working conditions and user preference settings to obtain adjusted threshold parameters, including: extracting time series features of the working conditions to obtain a charging device battery temperature time series implicit correlation feature vector; semantically encoding the user preference settings to obtain a user preference setting semantic coding feature vector; performing two-way attention interaction analysis on the charging device battery temperature time series implicit correlation feature vector and the user preference setting semantic coding feature vector to obtain a working condition-user preference semantic collaborative coding vector; and optimizing and adjusting the initial threshold parameters based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameters.

[0011] Specifically, extracting the time series features of the working conditions to obtain the time series implicit correlation feature vector of the battery temperature of the charging device includes: inputting the working conditions into a sequence encoder based on an LSTM-RNN hybrid model to obtain the time series implicit correlation feature vector of the battery temperature of the charging device.

[0012] Specifically, semantically encoding the user preference settings to obtain a semantically encoded feature vector of the user preference settings includes: using a language model based on a Transformer architecture to semantically encode the user preference settings to obtain the semantically encoded feature vector of the user preference settings.

[0013] Specifically, a bidirectional attention interaction analysis is performed on the implicit correlation feature vector of the battery temperature timing of the charging device and the semantic coding feature vector of the user preference setting to obtain a working condition-user preference semantic collaborative coding vector, including: constructing a positive and negative bidirectional attention balance field between the implicit correlation feature vector of the battery temperature timing of the charging device and the semantic coding feature vector of the user preference setting to obtain a working condition-user preference positive and negative bidirectional attention balance field; based on the working condition-user preference positive and negative bidirectional attention balance field, a field mapping interaction analysis is performed on the implicit correlation feature vector of the battery temperature timing of the charging device and the semantic coding feature vector of the user preference setting to obtain the working condition-user preference semantic collaborative coding vector.

[0014] Specifically, a positive and negative bidirectional attention balance field between the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector is constructed to obtain a working condition-user preference positive and negative bidirectional attention balance field, including: performing homography projection transformation on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain a charging device battery temperature time series feature homography projection coding vector and a user preference setting semantic feature homography projection coding vector; calculating the forward attention score field of the charging device battery temperature time series feature homography projection coding vector relative to the user preference setting semantic feature homography projection coding vector to obtain a working condition-user preference forward attention score field; calculating the reverse attention score field of the user preference setting semantic feature homography projection coding vector relative to the charging device battery temperature time series feature homography projection coding vector to obtain a working condition-user preference reverse attention score field; based on the working condition-user preference forward attention score field and the working condition-user preference reverse attention score field, constructing the working condition-user preference positive and negative bidirectional attention balance field.

[0015] Specifically, based on the working condition-user preference positive and negative bidirectional attention balance field, the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector are subjected to field mapping interactive analysis to obtain the working condition-user preference semantic collaborative coding vector, including: mapping the charging device battery temperature time series feature homography projection coding vector and the user preference setting semantic feature homography projection coding vector to the working condition-user preference positive and negative bidirectional attention balance field respectively to obtain the charging device battery temperature homography projection attention modulation coding vector and the user preference setting semantic homography projection attention modulation coding vector; calculating the position point division between the charging device battery temperature homography projection attention modulation coding vector and the user preference setting semantic homography projection attention modulation coding vector to obtain the working condition-user preference semantic collaborative coding vector.

[0016] Specifically, the initial threshold parameter is optimized and adjusted based on the working condition-user preference semantic co-coding vector to obtain the adjusted threshold parameter, including: inputting the working condition-user preference semantic co-coding vector into a decoder-based threshold dynamic optimizer to obtain an adjustment factor; multiplying the adjustment factor by the initial threshold parameter to obtain the adjusted threshold parameter.

[0017] According to another aspect of the present application, a fast charging control system is provided, comprising:

[0018] A threshold parameter adjustment module is used to set the initial threshold parameter according to the output power of the charger, and dynamically adjust the initial threshold parameter based on the working conditions and user preference settings to obtain the adjusted threshold parameter; a gear information modification module is used to obtain the gear information of the charging power of the charger and record it as the first gear information, compare the first gear information with the adjusted threshold parameter in turn, and modify the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information; a power request information sending module is used for the charged main device to receive the second gear information and send the power request information according to the received second gear information; a charging gear information return module is used to parse the power request information, select the charging gear required by the charged main device, and return the selected charging gear information to the charger; a charging module is used for the charger to receive the charging gear information and start charging the charged main device.

[0019] This application has at least the following technical effects:

[0020] Compared with the prior art, the present application provides a fast charging control method and system, which first sets the initial threshold parameters according to the output power of the charger, and further introduces data processing technology based on deep learning to conduct a comprehensive analysis of the battery operating temperature and user preference settings of the charging device, optimizes and adjusts the initial threshold parameters, and then based on the adjusted threshold parameters, limits the current parameters corresponding to the gears whose power consumption exceeds the threshold parameters in all the charging power gear information of the charger, and forms new gear information to provide to the charged main device, so that it can select a suitable charging gear for charging, thereby realizing intelligent fast charging control. The present application dynamically adjusts the charging power gear by real-time monitoring of the battery temperature of the charging device and the user's usage habits, which can effectively avoid battery damage caused by overheating, while improving charging efficiency and user experience, and ensuring that the charging process is both efficient and safe. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0022] Figure 1 is a flow chart of a fast charging control method according to an embodiment of the present application;

[0023] Figure 2 A data flow diagram of a fast charging control method according to an embodiment of the present application;

[0024] Figure 3 It is a flowchart of sub-step S1 of the fast charging control method according to an embodiment of the present application;

[0025] Figure 4 It is a flowchart of sub-step S13 of the fast charging control method according to an embodiment of the present application;

[0026] Figure 5 is a block diagram of a fast charging control system according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0028] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0029] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0030] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0031] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0032] In the technical solution of the present application, a fast charging control method is proposed. Figure 1 Flow chart of a fast charging control method according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the fast charging control method according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the fast charging control method according to the embodiment of the present application includes the following steps: S1, setting an initial threshold parameter according to the output power of the charger, and dynamically adjusting the initial threshold parameter based on the working conditions and user preference settings to obtain an adjusted threshold parameter; S2, obtaining the gear information of the charging power of the charger, and recording it as the first gear information, comparing the first gear information with the adjusted threshold parameter in turn, and modifying the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information; S3, the charged main device receives the second gear information, and sends a power request information according to the received second gear information; S4, parsing the power request information, selecting the charging gear required by the charged main device, and returning the selected charging gear information to the charger; S5, the charger receives the charging gear information and starts charging the charged main device.

[0033] In particular, the S1 sets the initial threshold parameter according to the output power of the charger, and dynamically adjusts the initial threshold parameter based on the working conditions and user preference settings to obtain the adjusted threshold parameter. It should be understood that the output power of the charger is its basic performance indicator. Different chargers have different power carrying capacities. Setting the initial threshold parameter based on this can provide a benchmark for subsequent charging control. For example, if the output power of the charger is high, a relatively high initial threshold parameter can be set accordingly to adapt to its larger power output range; conversely, if the charger power is small, the initial threshold parameter should also be appropriately reduced. In a specific example of the present application, the initial threshold parameter includes an initial threshold voltage, an initial threshold current, and an initial threshold power. In addition, it is considered that the battery temperature has a significant impact on charging safety and battery life. At different operating temperatures, the battery has different suitable charging powers. If high power charging is still used at high temperatures, safety issues such as overheating, battery expansion, and even explosion may occur, and charging efficiency will also be affected at low temperatures. At the same time, different users have different preferences for charging speed and battery life. For example, some users may prefer to extend battery life and choose a slower but gentle charging method; while other users may pay more attention to charging speed. Therefore, in order to ensure the safety of the charging process and achieve a personalized charging experience, the present application further dynamically adjusts the initial threshold parameters based on the working conditions (i.e., the time queue of the battery temperature of the charging device) and the user's preference settings, so that the charging process is more in line with the actual needs of the user, thereby overcoming the defect that the traditional static threshold cannot adapt to diversified scenarios. In one example, the time queue of the battery temperature of the charging device can be collected by a temperature sensor built into the charging device. The temperature sensor can monitor the operating temperature of the battery in real time and transmit the data to the main control unit at a certain frequency. In addition, for user preference settings, the user's charging preferences, such as the desired charging speed, whether to prioritize battery life, etc., can be obtained through the user interface. In addition, the system can also automatically infer the user's preferences by learning the user's daily usage pattern.

[0034] In the embodiments of the present application, Figure 3 As shown, the S1 includes: S11, extracting the timing features of the working conditions to obtain the timing implicit correlation feature vector of the battery temperature of the charging device; S12, semantically encoding the user preference settings to obtain the semantic coding feature vector of the user preference settings; S13, performing two-way attention interaction analysis on the timing implicit correlation feature vector of the battery temperature of the charging device and the semantic coding feature vector of the user preference settings to obtain the working condition-user preference semantic collaborative coding vector; S14, optimizing and adjusting the initial threshold parameters based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameters.

[0035] Specifically, the S11 extracts the time series features of the working conditions to obtain the implicit correlation feature vector of the battery temperature of the charging device. Considering that the change of the battery temperature of the charging device over time is a complex dynamic process, it contains important information such as the charging and discharging state of the battery, the heat dissipation and the influence of the surrounding environment on it. Therefore, in order to accurately capture the time series dynamic changes of the battery temperature of the charging device, the present application adopts a time series analysis method based on the LSTM-RNN hybrid model, and performs time series encoding processing on the working conditions to utilize the advantages of the LSTM model and the RNN model in processing time series data, accurately capture the rising and falling trends of the temperature, the periodic change law, and the potential correlation between the temperatures at different times, and convert it into a fixed-dimensional implicit correlation feature vector of the battery temperature of the charging device, thereby providing efficient data representation for subsequent data analysis. In particular, the LSTM-RNN hybrid model combines the characteristics of the long short-term memory network and the recurrent neural network, and is intended to handle sequence data problems with long-term dependencies. Specifically, RNN is a neural network structure for processing sequence data, which allows information to be transferred between different time steps in a sequence. However, traditional RNNs suffer from the problem of vanishing or exploding gradients, which limits their ability to capture long-term dependencies. To solve this problem, LSTM was introduced as an improved RNN unit that alleviates these problems through specially designed memory units and is able to effectively learn long-term dependencies.

[0036] Specifically, the S12 semantically encodes the user preference settings to obtain a semantically encoded feature vector of the user preference settings. In the technical solution of the present application, a language model based on the Transformer architecture is used to semantically encode the user preference settings to obtain a semantically encoded feature vector of the user preference settings. Considering that user preference settings usually exist in the form of natural language descriptions or unstructured data, they cannot be directly used by computer models for optimizing the initial threshold parameters. Therefore, the present application further uses natural language processing technology to semantically encode the user preference settings, extract key semantic information in the user preference settings, and convert it into a semantically encoded feature vector of the user preference settings, so as to effectively fuse and interact with the implicitly associated feature vector of the battery temperature timing of the charging device, so that the model can more comprehensively understand the various influencing factors in the charging process, and then make more accurate charging power gear adjustments. In a specific example of the present application, a language model based on the Transformer architecture is used to semantically encode the user preference settings to obtain a semantically encoded feature vector of the user preference settings. In particular, Transformer is a deep learning model architecture based on the self-attention mechanism. Compared with traditional RNN and LSTM, Transformer can handle long-distance dependency problems more effectively and has significant advantages in parallel training. The core idea is to completely abandon the loop or convolution structure and use the self-attention mechanism to capture the relationship between different positions in the input data. Among them, the self-attention mechanism calculates the query, key, and value vectors for each position, and then assigns attention weights according to the matching degree of the query and key, and finally weights and sums the corresponding value vector as the output. This mechanism allows the model to take into account global information when processing each position, rather than relying solely on local context.

[0037] Specifically, S13 performs a two-way attention interaction analysis on the implicitly associated feature vector of the battery temperature timing of the charging device and the semantic coding feature vector of the user preference setting to obtain a working condition-user preference semantic collaborative coding vector. Considering that the implicitly associated feature vector of the battery temperature timing of the charging device reflects the actual working state of the charging device, and the semantic coding feature vector of the user preference setting reflects the personalized needs of the user, both are crucial for the optimization adjustment of the initial threshold parameters. Therefore, in order to capture the synergistic effects of the battery temperature state and the user preference setting on the initial threshold parameters, the present application further performs a two-way feature attention interaction on the implicitly associated feature vector of the battery temperature timing of the charging device and the semantic coding feature vector of the user preference setting to identify and capture the synergistic effect between the two, generate a working condition-user preference semantic collaborative coding vector, and thereby adjust the initial threshold parameters more accurately. Specifically, in a specific example of the present application, such as Figure 4 As shown, the S13 includes: S131, constructing a positive and negative bidirectional attention balance field between the charging device battery temperature timing implicit association feature vector and the user preference setting semantic coding feature vector to obtain a working condition-user preference positive and negative bidirectional attention balance field; S132, based on the working condition-user preference positive and negative bidirectional attention balance field, performing field mapping interactive analysis on the charging device battery temperature timing implicit association feature vector and the user preference setting semantic coding feature vector to obtain the working condition-user preference semantic collaborative coding vector.

[0038] More specifically, the S131 constructs a positive and negative bidirectional attention balance field between the implicitly associated feature vector of the battery temperature sequence of the charging device and the semantically encoded feature vector of the user preference setting to obtain a positive and negative bidirectional attention balance field of the working condition-user preference. Specifically, first, the implicitly associated feature vector of the battery temperature sequence of the charging device and the semantically encoded feature vector of the user preference setting are subjected to homography projection transformation to obtain a homography projection encoding vector of the battery temperature sequence feature of the charging device and a homography projection encoding vector of the semantic feature of the user preference setting. It should be understood that the implicitly associated feature vector of the battery temperature sequence of the charging device and the semantically encoded feature vector of the user preference setting may be in different feature spaces, with different feature dimensions and distributions, and direct interaction analysis may result in poor results due to the inconsistency of the features. Therefore, the present application first performs a homography projection transformation on the implicitly associated feature vector of the battery temperature sequence of the charging device and the semantically encoded feature vector of the user preference setting to map the two into a unified feature space for effective interaction and fusion, which helps to reduce information loss or misunderstanding caused by inconsistent feature spaces. In a specific example, the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector are homographically projected to obtain the charging device battery temperature time series feature homographic projection coding vector and the user preference setting semantic feature homographic projection coding vector using the following homographic projection transformation formula; wherein the homographic projection transformation formula is:

[0039]

[0040]

[0041] in, represents the implicit correlation feature vector of the battery temperature time series of the charging device, represents the semantic encoding feature vector of the user preference setting, and are different homography projection matrices, and They respectively represent the homography projection coding vector of the battery temperature temporal feature of the charging device and the homography projection coding vector of the semantic feature of the user preference setting.

[0042] Next, the positive attention score field of the homography projection coding vector of the battery temperature time series feature of the charging device relative to the homography projection coding vector of the semantic feature of the user preference setting is calculated to obtain the working condition-user preference positive attention score field. By calculating the positive attention score field, the most relevant and critical information part to the user preference is mined from the battery temperature feature, highlighting the battery temperature change characteristics that are closely related to user needs during the charging process; for example, when the user prefers fast charging, the temperature change characteristics of the high temperature zone will be paid special attention to, so that in the subsequent collaborative encoding process, more focus can be placed on the information that has an important impact on the adjustment of the charging power gear. In a specific example, the positive attention score calculation formula is used to calculate the positive attention score field of the homography projection coding vector of the battery temperature time series feature of the charging device relative to the homography projection coding vector of the semantic feature of the user preference setting to obtain the working condition-user preference positive attention score field; wherein, the positive attention score calculation formula is:

[0043]

[0044] in, represents the transpose of a vector, represents the matrix multiplication operation, is the attention score scaling factor, Represents the working condition - user preference positive attention score field.

[0045] Furthermore, the reverse attention score field of the homography projection coding vector of the semantic feature of the user preference setting relative to the homography projection coding vector of the battery temperature timing feature of the charging device is calculated to obtain the reverse attention score field of the working condition-user preference. By calculating the reverse attention score field, the part closely related to the battery temperature state is identified from the user preference feature. For example, under high temperature conditions, the low temperature charging mode in the user preference setting will be specially considered to ensure that the adjustment of the charging power under high temperature conditions is more cautious to prevent safety problems caused by overheating and ensure battery safety. In a specific example, the reverse attention score field of the homography projection coding vector of the semantic feature of the user preference setting relative to the homography projection coding vector of the battery temperature timing feature of the charging device is calculated using the following reverse attention score calculation formula to obtain the reverse attention score field of the working condition-user preference; wherein, the reverse attention score calculation formula is:

[0046]

[0047] in, Represents the working condition-user preference reverse attention score field.

[0048] Subsequently, based on the working condition-user preference positive attention score field and the working condition-user preference reverse attention score field, the working condition-user preference positive and negative bidirectional attention balance field is constructed. Here, in order to comprehensively consider the positive and reverse attention information, the working condition-user preference positive and negative bidirectional attention balance field is further constructed to ensure that in the final interactive response encoding, it will neither be overly biased towards one party nor ignore the contribution of any party, so as to better balance the relationship between the battery working conditions and user preferences, ensure that the threshold adjustment not only meets the safety and performance requirements of the battery, but also can meet the user's expectations to the greatest extent, and improve the intelligence and flexibility of charging control. In a specific example, based on the working condition-user preference positive attention score field and the working condition-user preference reverse attention score field, the working condition-user preference positive and negative bidirectional attention balance field is constructed with the following formula; wherein, the formula is:

[0049]

[0050] in, Indicates cascade, Represents working conditions - user preference positive and negative bidirectional attention balance field Representation 3 3 convolution operations.

[0051] More specifically, in S132, based on the working condition-user preference positive and negative bidirectional attention balance field, the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector are subjected to field mapping interactive analysis to obtain the working condition-user preference semantic collaborative coding vector. Specifically, first, the charging device battery temperature time series feature homography projection coding vector and the user preference setting semantic feature homography projection coding vector are respectively mapped to the working condition-user preference positive and negative bidirectional attention balance field to obtain the charging device battery temperature homography projection attention modulation coding vector and the user preference setting semantic homography projection attention modulation coding vector. Specifically, the implicitly associated feature vector of the battery temperature time series of the charging device and the semantically encoded feature vector of the user preference setting after homography projection transformation are respectively projected into the working condition-user preference positive and negative bidirectional attention balance field, so as to adjust the feature representations of the two according to the attention weight distribution of the working condition-user preference positive and negative bidirectional attention balance field, so that the feature parts related to important information are enhanced and the feature parts related to secondary information are suppressed, thereby obtaining the homography projection attention modulation coding vector of the battery temperature of the charging device and the homography projection attention modulation coding vector of the user preference setting semantics. In a specific example, the homography projection coding vector of the battery temperature time series feature of the charging device and the homography projection coding vector of the semantic feature of the user preference setting are respectively mapped to the working condition-user preference positive and negative bidirectional attention balance field to obtain the homography projection attention modulation coding vector of the battery temperature of the charging device and the homography projection attention modulation coding vector of the semantic homography projection attention modulation coding vector of the user preference setting; wherein, the mapping formula is:

[0052]

[0053]

[0054] in, and They respectively represent the homography projection attention modulation coding vector of the battery temperature of the charging device and the homography projection attention modulation coding vector of the semantics of the user preference setting.

[0055] Furthermore, the working condition-user preference semantic collaborative coding vector is obtained by calculating the position point division between the homography projection attention modulation coding vector of the battery temperature of the charging device and the semantic homography projection attention modulation coding vector of the user preference setting. It should be understood that in the above-mentioned attention balance field projection process, the expression of the other party is essentially adjusted based on the reaction of one party, so that the generated homography projection attention modulation coding vector of the battery temperature of the charging device and the semantic homography projection attention modulation coding vector of the user preference setting reflect the collaborative relationship between working conditions and user preferences to a certain extent. In the technical solution of the present application, the homography projection attention modulation coding vector of the battery temperature of the charging device and the semantic homography projection attention modulation coding vector of the user preference setting are further interactively fused element by element through the position point division operation to generate a working condition-user preference semantic collaborative coding vector, so as to more comprehensively reflect the comprehensive influence of the working state of the charging device and the user preference. In a specific example, the working condition-user preference semantic collaborative coding vector is obtained by dividing the position point between the homography projection attention modulation coding vector of the battery temperature of the charging device and the homography projection attention modulation coding vector of the user preference setting semantics by the following formula; wherein the formula is:

[0056]

[0057] in, Represents the working condition-user preference semantic co-encoding vector.

[0058] Specifically, the S14 optimizes and adjusts the initial threshold parameter based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameter. In the technical solution of the present application, the working condition-user preference semantic collaborative coding vector is input into a threshold dynamic optimizer based on a decoder to obtain an adjustment factor; specifically, the threshold dynamic optimizer based on the decoder performs deep feature extraction and pattern recognition on the input working condition-user preference semantic collaborative coding vector through a multi-layer neural network structure, thereby generating an adjustment factor that matches the current charging environment and user needs. It should be understood that the adjustment factor is a quantitative representation of the comprehensive effect of the current battery temperature and user demand preference. When the battery temperature is at a high level, if the user prefers to prioritize battery safety and life, the decoder will tend to generate a lower adjustment factor to reduce the charging power and avoid the risk of overheating; conversely, when the battery temperature is low and the user prefers fast charging, the decoder will generate a higher adjustment factor to increase the charging power and meet user needs. Finally, the adjustment factor is multiplied by the initial threshold parameter to obtain the adjusted threshold parameter.

[0059] Preferably, the working condition-user preference semantics collaborative coding vector is passed through a decoder-based threshold dynamic optimizer to obtain an adjustment factor, including:

[0060] Determine the median eigenvalue in the working condition-user preference semantic co-encoding vector , the maximum eigenvalue and the minimum eigenvalue , and divide the median eigenvalue by the difference between the maximum eigenvalue and the minimum eigenvalue to obtain the working condition-user preference semantic collaborative coding distribution probability value ,Right now ;

[0061] The working condition-user preference semantic collaborative coding vector is normalized to the maximum value to obtain the working condition-user preference semantic collaborative coding probability feature vector ;

[0062] The power function of each eigenvalue of the working condition-user preference semantic collaborative coding probability feature vector is calculated by subtracting the working condition-user preference semantic collaborative coding distribution probability value from 1 to obtain the first working condition-user preference semantic collaborative coding convergence feature vector. ;

[0063] The unit feature vector is calculated based on the working condition-user preference semantic collaborative coding distribution probability value as an exponent Co-encoding probability feature vector with the working condition-user preference semantics The power function of each eigenvalue of the difference eigenvector is used to obtain the second working condition - user preference semantic collaborative coding convergence eigenvector ;

[0064] The working condition-user preference semantic collaborative coding probability feature vector is multiplied by 1 minus the difference between the working condition-user preference semantic collaborative coding distribution probability value to obtain a first working condition-user preference semantic collaborative coding limited feature vector ;

[0065] The point difference feature vector is multiplied by the working condition-user preference semantic collaborative coding distribution probability value to obtain a second working condition-user preference semantic collaborative coding limited feature vector ;

[0066] After the first working condition-user preference semantic collaborative coding convergence feature vector is dot-multiplied with the second working condition-user preference semantic collaborative coding convergence feature vector, the optimized working condition-user preference semantic collaborative coding vector is obtained by dot-multiplying the first working condition-user preference semantic collaborative coding limit feature vector and the second working condition-user preference semantic collaborative coding limit feature vector. ;

[0067] The optimized working condition-user preference semantics co-coding vector is passed through a decoder-based threshold dynamic optimizer to obtain an adjustment factor.

[0068] Among them, considering that the implicit correlation feature vector of the charging device battery temperature time series and the semantic coding feature vector of the user preference setting respectively represent the correlation feature of the charging device battery temperature time series and the semantic coding feature of the user preference setting, when performing interactive responses based on the feature positive and negative attention fields, the data modality differences between the working conditions and the user preference settings will make the fine-grained deviations of the positive and negative attention fields significant, resulting in the working condition-user preference semantic collaborative coding vector having a probabilistic convergence and divergence based on different attention interactivities, thereby affecting the accuracy of the adjustment factor obtained by the decoder-based threshold dynamic optimizer.

[0069] Based on this, by using the cross-entropy power series of the distribution analysis probability of the working condition-user preference semantic co-coding vector as the probability distribution convergence limit, the regression probability convergence approximation is performed on the basis of the combination of the feature set distribution and the probability density distribution of the working condition-user preference semantic co-coding vector, so as to avoid the conservative convergence control of the fatal potential divergence of the working condition-user preference semantic co-coding vector, and further use the probability distribution cross entropy of the working condition-user preference semantic co-coding vector as the objective function to guide the limited recovery strategy, so as to achieve the common agility of the convergence of the probability convergent and divergent feature set of the working condition-user preference semantic co-coding vector to the probability density distribution space, and improve the accuracy of the adjustment factor of the working condition-user preference semantic co-coding vector obtained by the decoder-based threshold dynamic optimizer.

[0070] In particular, the S2 obtains the gear information of the charging power of the charger and records it as the first gear information, compares the first gear information with the adjusted threshold parameter in turn, and modifies the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information. It should be understood that the charging gear refers to the different output voltage and current combinations that the charger can provide. Each charging gear represents a specific charging speed or efficiency setting, and defines the power level that the charger can provide to the charged device. For example, in the PD (Power Delivery) fast charging protocol, the charger may support multiple gears, such as 5V / 3A, 9V / 2A, 15V / 3A, etc. Different voltage and current combinations correspond to different charging powers. Specifically, 5V / 3A means that the charger can provide a current of up to 3 amperes at a voltage of 5 volts, which is usually used for charging ordinary USB devices with a total power of 15 watts. 9V / 2A means providing a current of 2 amperes at a voltage of 9 volts, which is suitable for some mobile devices that support fast charging, with a total power of 18 watts. 15V / 3A provides 3 amps of current at 15 volts, which is suitable for devices with higher power requirements, with a total power of 45 watts.

[0071] Accordingly, if the power consumption value of a certain gear of the charger exceeds the adjusted threshold parameter, it is very easy to cause the problem of uneven power distribution, which in turn causes the charger to be in a high-load or even full-load working state, which may cause irreversible damage to the charger. At this time, it is necessary to further adjust the current value of the gear to ensure that the charging process is both efficient and safe. In the specific implementation process, the voltage value in the gear information is first carefully compared with the adjusted threshold voltage. When the voltage value is greater than or equal to the adjusted threshold voltage value, the power pre-withholding enabling mechanism is activated, but the power pre-withholding operation is not immediately performed at this stage. Then, the current value in the current gear information is further compared with the adjusted threshold current value. Only when the current value is greater than the adjusted threshold current value, the power pre-withholding program is officially started. Subsequently, the current value in the gear information is adjusted to ensure that the modified current value is less than the current value before the modification, so as to achieve effective pre-withholding of power. The above steps are repeated continuously until the voltage value and current value of each gear in the first gear information are fully compared with the adjusted threshold parameters, and the current value is accurately modified accordingly, and finally the second gear information is formed.

[0072] In particular, in S3, the charged main device receives the second gear information, and sends power request information according to the received second gear information. It should be understood that since the second gear information has fully considered the power limit of the charger and the possible power consumption of the extended device, the power request information generated by the charged main device based on this is more scientific and reasonable. In the actual operation process, the charged main device will comprehensively consider multiple factors such as its own remaining power status, the current running task load, and the pre-set charging strategy, and closely combine the received second gear information to carefully generate power request information containing appropriate power value, voltage value and current value information, and then accurately send it to the controller to provide key basis for subsequent charging decisions.

[0073] In particular, the S4 analyzes the power request information, selects the charging gear required by the charged main device, and returns the selected charging gear information to the charger. That is, after receiving the power request information sent by the charged main device, the controller uses the analysis algorithm to analyze the information. In view of the fact that the power request information may cover multiple roughly matching charging gears, when selecting the final charging gear, it is necessary to further refer to the adjusted threshold parameter for accurate judgment. When the voltage value of the selected charging gear is less than or equal to the adjusted threshold voltage, it indicates that the power demand of the charging gear is within the tolerable range of the charger. At this time, the charging gear information can be directly transmitted back to the charger to meet the charging demand of the charged main device. However, when the voltage value of the selected charging gear is greater than the adjusted threshold voltage, it means that the charging gear may require higher power support. At this time, the first gear information corresponding to the charging gear should be selected and fed back to the charger. Since the first gear information stores the maximum charging power gear that the charger can provide, it can ensure that the charged main device obtains sufficient power supply, thereby ensuring the smooth progress of the charging process and achieving efficient and stable charging goals.

[0074] In particular, in S5, the charger receives the charging gear information and starts charging the main device being charged. After the charger successfully receives the charging gear information returned by the controller, the charging protocol negotiation between the charger and the main device being charged is completed. At this time, the charger will accurately adjust its output voltage and current parameters according to the received charging gear information, and stably deliver power to the main device being charged, thereby achieving the expected goal of fast charging.

[0075] In summary, the fast charging control method according to the embodiment of the present application is explained, which first sets the initial threshold parameters according to the output power of the charger, and further introduces data processing technology based on deep learning to comprehensively analyze the battery operating temperature and user preference settings of the charging device, optimizes and adjusts the initial threshold parameters, and then based on the adjusted threshold parameters, adjusts the current parameters corresponding to the gears whose power consumption exceeds the threshold parameters in all the charging power gear information of the charger, and forms new gear information to provide to the charged main device for it to select a suitable charging gear for charging, so as to realize intelligent fast charging control. By dynamically adjusting the charging power gear by real-time monitoring of the battery temperature of the charging device and the user's usage habits, it can effectively avoid battery damage caused by overheating, while improving charging efficiency and user experience, and ensuring that the charging process is both efficient and safe.

[0076] Furthermore, a fast charging control system is also provided.

[0077] Figure 5 FIG. 1 is a block diagram of a fast charging control system according to an embodiment of the present application. Figure 5 As shown, according to the fast charging control system 300 of the embodiment of the present application, it includes: a threshold parameter adjustment module 310, which is used to set the initial threshold parameter according to the output power of the charger, and dynamically adjust the initial threshold parameter based on the working conditions and user preference settings to obtain the adjusted threshold parameter; a gear information modification module 320, which is used to obtain the gear information of the charging power of the charger, and record it as the first gear information, compare the first gear information with the adjusted threshold parameter in turn, and modify the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information; a power request information sending module 330, which is used for the charged main device to receive the second gear information, and send the power request information according to the received second gear information; a charging gear information return module 340, which is used to parse the power request information, select the charging gear required by the charged main device, and return the selected charging gear information to the charger; a charging module 350, which is used for the charger to receive the charging gear information and start charging the charged main device.

[0078] As described above, the fast charging control system 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a fast charging control algorithm. In one possible implementation, the fast charging control system 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the fast charging control system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the fast charging control system 300 can also be one of the many hardware modules of the wireless terminal.

[0079] Alternatively, in another example, the fast charging control system 300 and the wireless terminal may be separate devices, and the fast charging control system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0080] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A fast charging control method, characterized in that: include: Setting an initial threshold parameter according to the output power of the charger, and dynamically adjusting the initial threshold parameter based on working conditions and user preference settings to obtain an adjusted threshold parameter; The working condition is a time queue of the battery temperature of the charging device; Obtaining the gear information of the charging power of the charger and recording it as the first gear information, comparing the first gear information with the adjusted threshold parameter in sequence, and modifying the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information; The charged main device receives the second gear position information, and sends a power request message according to the received second gear position information; Parse the power request information, select the charging gear required by the main device being charged, and return the selected charging gear information to the charger; The charger receives the charging gear information and starts charging the main device being charged; The method of dynamically adjusting the initial threshold parameter based on the working conditions and the user preference setting to obtain the adjusted threshold parameter includes: Extracting time series features of the working conditions to obtain a time series implicit correlation feature vector of the battery temperature of the charging device; Performing semantic encoding on the user preference setting to obtain a user preference setting semantic encoding feature vector; A two-way attention interaction analysis is performed on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain a working condition-user preference semantic collaborative coding vector, which includes: Constructing a positive and negative bidirectional attention balance field between the charging device battery temperature time series implicit association feature vector and the user preference setting semantic encoding feature vector to obtain a working condition-user preference positive and negative bidirectional attention balance field; Based on the working condition-user preference positive and negative bidirectional attention balance field, a field mapping interactive analysis is performed on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain the working condition-user preference semantic collaborative coding vector; The initial threshold parameter is optimized and adjusted based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameter.

2. The fast charging control method according to claim 1, characterized in that: Extracting the time series features of the working conditions to obtain the time series implicit correlation feature vector of the battery temperature of the charging device includes: The working conditions are input into a sequence encoder based on an LSTM-RNN hybrid model to obtain an implicit correlation feature vector of the battery temperature time series of the charging device.

3. The fast charging control method according to claim 2, characterized in that: Semantically encoding the user preference setting to obtain a semantic encoding feature vector of the user preference setting includes: The user preference settings are semantically encoded using a language model based on a Transformer architecture to obtain a semantic encoding feature vector of the user preference settings.

4. The fast charging control method according to claim 3, characterized in that: Constructing a positive and negative bidirectional attention balance field between the charging device battery temperature time series implicit association feature vector and the user preference setting semantic encoding feature vector to obtain a working condition-user preference positive and negative bidirectional attention balance field, including: Performing homography projection transformation on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain a charging device battery temperature time series feature homography projection coding vector and a user preference setting semantic feature homography projection coding vector; Calculating the positive attention score field of the homography projection coding vector of the battery temperature time series feature of the charging device relative to the homography projection coding vector of the semantic feature of the user preference setting to obtain the working condition-user preference positive attention score field; Calculating the reverse attention score field of the homography projection coding vector of the user preference setting semantic feature relative to the homography projection coding vector of the charging device battery temperature time series feature to obtain the working condition-user preference reverse attention score field; Based on the working condition-user preference positive attention score field and the working condition-user preference reverse attention score field, the working condition-user preference positive and negative bidirectional attention balance field is constructed.

5. The fast charging control method according to claim 4, characterized in that: Based on the working condition-user preference positive and negative bidirectional attention balance field, the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector are subjected to field mapping interactive analysis to obtain the working condition-user preference semantic collaborative coding vector, including: Mapping the charging device battery temperature time series feature homography projection coding vector and the user preference setting semantic feature homography projection coding vector to the working condition-user preference positive and negative bidirectional attention balance field to obtain the charging device battery temperature homography projection attention modulation coding vector and the user preference setting semantic homography projection attention modulation coding vector; The working condition-user preference semantic collaborative coding vector is obtained by calculating the point-by-point division between the charging device battery temperature homography projection attention modulation coding vector and the user preference setting semantic homography projection attention modulation coding vector.

6. The fast charging control method according to claim 5, characterized in that: Optimizing and adjusting the initial threshold parameter based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameter includes: Inputting the working condition-user preference semantics co-encoding vector into a threshold dynamic optimizer based on a decoder to obtain an adjustment factor; The adjustment factor is multiplied by the initial threshold parameter to obtain the adjusted threshold parameter.

7. A fast charging control system, characterized in that: include: A threshold parameter adjustment module, configured to set an initial threshold parameter according to the output power of the charger, and dynamically adjust the initial threshold parameter based on working conditions and user preference settings to obtain an adjusted threshold parameter; the working condition is a time queue of a battery temperature of a charging device; a gear information modification module, used to obtain gear information of the charging power of the charger, and record it as the first gear information, compare the first gear information with the adjusted threshold parameter in sequence, and modify the corresponding current value in the first gear information whose power consumption value is greater than the adjusted threshold parameter to form the second gear information; A power request information sending module, configured to receive the second gear position information from the charged main device, and send power request information according to the received second gear position information; The charging gear information return module is used to parse the power request information, select the charging gear required by the charged main device, and return the selected charging gear information to the charger; The charging module is used for the charger to receive the charging gear information and start charging the main device to be charged; The threshold parameter adjustment module is used to: Extracting time series features of the working conditions to obtain a time series implicit correlation feature vector of the battery temperature of the charging device; Performing semantic encoding on the user preference setting to obtain a user preference setting semantic encoding feature vector; A two-way attention interaction analysis is performed on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain a working condition-user preference semantic collaborative coding vector, which includes: Constructing a positive and negative bidirectional attention balance field between the charging device battery temperature time series implicit association feature vector and the user preference setting semantic encoding feature vector to obtain a working condition-user preference positive and negative bidirectional attention balance field; Based on the working condition-user preference positive and negative bidirectional attention balance field, a field mapping interactive analysis is performed on the charging device battery temperature time series implicit association feature vector and the user preference setting semantic coding feature vector to obtain the working condition-user preference semantic collaborative coding vector; The initial threshold parameter is optimized and adjusted based on the working condition-user preference semantic collaborative coding vector to obtain the adjusted threshold parameter.

Images