Charging optimization method for starting power supply system

Abstract

The invention relates to a charging optimization method for a starting power supply system, and belongs to the field of power transmission, conversion, distribution and control equipment, and the method comprises the steps: operating an AI prediction model to obtain a success and failure identifier of a next starting vehicle battery of the starting power supply system; and when the successful failure identifier marks that the starting of the power supply system fails, forbidding the starting of the power supply system and executing an early warning operation corresponding to the next starting failure, otherwise, continuing to enable the starting of the power supply system. According to the invention, for the technical problem that the charging optimization space is limited because the future starting success or failure of the starting power supply system is difficult to predict, the success and failure of the next starting of different starting power supply systems are predicted by adopting the basic data screened in a targeted manner and utilizing the AI prediction model designed by a customized structure; and on the basis of prediction, the charging mode for starting the power supply system is optimized, so that the technical problem is solved.

Description

Technical Field

[0001] The invention relates to the field of power transmission, conversion, distribution and control equipment, and in particular to a charging optimization method for a starting power supply system. Background Art

[0002] Generally, no matter gasoline vehicles, diesel vehicles or gas vehicles, they all need to use a starting power system to start the vehicle battery into working state, thereby driving the normal operation of the entire vehicle. The starting power system is powerful and relatively safe. It can effectively start the car battery. At the same time, it has a forced start function, that is, when the car user cannot start the car battery, the car battery can be started by pressing the forced start button for about 3 seconds. The reliable starting power system can start the car battery while reducing the temperature of the car to ensure the safe operation of the car battery.

[0003] For example, the invention patent with publication number CN106374604A proposes a starting method for an intelligent starting power supply system, and its application objects include diesel (gasoline) engines / generator sets for vehicles, ships, and buildings, which can replace the existing engine / generator starting batteries. The intelligent starting power supply system device includes a supercapacitor module, a supercapacitor balancing circuit, a battery, a DC bidirectional converter, an intelligent control module, and a switch. A DC bidirectional converter is connected between the supercapacitor module and the battery. The DC bidirectional converter realizes bidirectional energy transfer between the supercapacitor module and the battery. The intelligent control module detects the voltage of the supercapacitor module and the battery, determines the starting state of the engine, and sends instructions to the DC bidirectional converter to control the flow of electrical energy between the supercapacitor module and the battery. The starting power supply system device can realize parallel state starting between the supercapacitor module and the battery through a parallel switch through intelligent conversion.

[0004] For example, the invention patent with publication number CN116707105A proposes an intelligent starting power supply system for fuel vehicles, including a battery pack, a supercapacitor module, a DC-DC module, a BMS power management module, a main control unit, a T-BOX and a remote monitoring module; the two ends of the supercapacitor module are respectively connected to the positive and negative poles of the power interface; the battery pack and the DC-DC module are connected in series and in parallel with the supercapacitor module; the BMS power management module is respectively connected to the battery pack, the main control unit and the T-BOX; the main control unit is respectively connected to the supercapacitor module and the DC-DC module and the T-BOX, and controls the DC-DC module according to the received charge and discharge instructions, so that when discharging, the battery pack is controlled to charge the supercapacitor module, so that when charging, the battery pack is charged by taking power from the voltage at the end of the supercapacitor module; the T-BOX receives the battery pack information, the data of the supercapacitor module and the DC-DC module, and packages and uploads it to the remote monitoring module. The intelligent starting power supply system has a long service life and can start the vehicle normally at low temperatures to avoid pollution to the environment.

[0005] It can be seen that the above-mentioned technical solutions are only limited to the specific design of the internal structure and starting mode of the starting power system, and cannot provide more prediction data, resulting in a lack of basic information for subsequent charging optimization. In fact, different starting power systems have different parameters such as maximum battery capacity, starting voltage, peak current and system weight. At the same time, the types of vehicles being started are different, and the various data of the vehicle batteries used are different, resulting in difficulty in predicting whether the start will be successful under different battery capacities. If the car user cannot obtain the prediction information of whether the start will be successful, it is easy to cause the start to fail but still insist on starting, or cause the start to succeed but the car user has no confidence in executing the start. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a charging optimization method for a starting power system. By adopting targeted screened basic data and utilizing an AI prediction model with a customized structure, the success and failure of the next startup of different starting power systems are predicted. Based on the prediction, the charging mode of the starting power system is optimized. For example, when it is predicted that the next startup will be successful, the starting power system continues to be enabled for startup. When it is predicted that the next startup will fail, the starting power system is disabled and the next startup is not performed, thereby avoiding meaningless startup in the event of a next startup failure and reducing damage to the vehicle charging components.

[0007] According to the present invention, a charging optimization method for a starting power supply system is provided, the method comprising:

[0008] Capturing various configuration data of a starting power system for starting a vehicle battery, wherein the various configuration data of the starting power system are a maximum battery capacity, a starting voltage, a peak current, and a system weight of the starting power system;

[0009] Collecting the vehicle type identification, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located of the target vehicle where the vehicle battery is located as a plurality of on-site information outputs of the target vehicle;

[0010] Performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process is completed, and outputting the deep neural network after each learning process as an AI prediction model;

[0011] The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model and obtain the success or failure indicator of the next vehicle battery starting of the starting power system output by the AI ​​prediction model;

[0012] When the received success or failure identification mark of the starting power system for the next time starting the vehicle battery indicates that the starting power system fails to start the vehicle battery next time, disabling the starting power system and performing a warning operation corresponding to the next start failure to suspend the next start; when the received success or failure identification mark of the starting power system for the next time starting the vehicle battery indicates that the starting power system succeeds in starting the vehicle battery next time, continuing to enable the starting power system;

[0013] Among them, the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located are collected as multiple on-site information outputs of the target vehicle, including: when the vehicle type identifications of the target vehicles where the vehicle battery is located are 0B01, 0B10 and 0B11 respectively, the vehicle types of the target vehicles where the vehicle battery is located are gasoline vehicles, diesel vehicles and gas vehicles respectively.

[0014] Compared with the prior art, the present invention has at least the following four key inventive features:

[0015] Invention point 1: An artificial intelligence prediction mechanism is used to accurately predict the success or failure of the next start-up of the vehicle battery by the starting power system, and when it is predicted that the next start-up of the vehicle battery by the starting power system fails, the starting power system is disabled and a warning operation corresponding to the next start failure is performed to suspend the next start-up, and when it is predicted that the next start-up of the vehicle battery by the starting power system is successful, the starting power system continues to be enabled to prepare for the next start-up, thereby completing the targeted optimization processing of the charging process of the vehicle battery by the starting power system, avoiding entering an invalid charging process of a start failure, and reducing damage to the vehicle battery and associated vehicle components;

[0016] Invention point 2: An AI prediction model with a customized structure is used to implement an artificial intelligence prediction mechanism. The structural customization of the AI ​​prediction model is reflected in the following aspects: the AI ​​prediction model is a deep neural network after completing each learning process, the number of learning times of the deep neural network is monotonically positively correlated with the maximum battery capacity of the starting power system, and the deep neural network includes multiple hidden layers, a single input layer and a single output layer, and the number of the multiple hidden layers is proportional to the peak current of the starting power system, thereby customizing AI prediction models with different structures for different starting power systems, ensuring the reliability and stability of the prediction results;

[0017] Invention point three: multiple basic data are targeted for the implementation of the artificial intelligence prediction mechanism, and the multiple basic data include the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle, wherein the various configuration data of the starting power system are the maximum battery capacity, starting voltage, peak current and system weight of the starting power system, and the multiple on-site information of the target vehicle are the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinder, the current vehicle body temperature and the humidity of the environment where the vehicle body is located. The full and comprehensive screening of the above multiple basic data further ensures the reliability and stability of the prediction results;

[0018] Invention point 4: In each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network to complete this learning process, thereby ensuring the learning effect of the deep neural network every time. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0020] Figure 1 The present invention is a technical flow chart of a charging optimization method for a starting power system according to the present invention.

[0021] Figure 2 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 1 of the present invention.

[0022] Figure 3 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 2 of the present invention.

[0023] Figure 4 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 3 of the present invention.

[0024] Figure 5 The present invention is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 4 of the present invention.

[0025] Figure 6 A flowchart of the steps of the charging optimization method of the starting power system according to Embodiment 5 of the present invention is shown. DETAILED DESCRIPTION

[0026] like Figure 1 As shown, a technical flow chart of a charging optimization method for a starting power system according to the present invention is given.

[0027] Figure 1 A force button is provided on the front panel of the starting power supply system in the invention, and is used to start the car battery by pressing the force button for about 3 seconds when the car user cannot start the car battery.

[0028] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0029] Technical process 1: Designing a customized AI prediction model for predicting the success or failure of the next vehicle battery start-up by the starting power system. The structure customization of the AI ​​prediction model is reflected in the following aspects:

[0030] First aspect: the AI ​​prediction model is a deep neural network after completing each learning process;

[0031] Second aspect: the number of times the deep neural network is learned is monotonically positively correlated with the maximum battery capacity of the starting power system;

[0032] A third aspect: the deep neural network includes multiple hidden layers, a single input layer and a single output layer, and the number of the multiple hidden layers is proportional to the peak current of the starting power supply system;

[0033] Fourth aspect: In each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network to complete this learning process, thereby ensuring the learning effect of the deep neural network every time;

[0034] In this way, through the customized design of the above structure, AI prediction models with different structures are customized for different starting power supply systems, thereby ensuring the reliability and stability of the prediction results;

[0035] Technical process 2: Targeted selection of multiple basic data for predicting the success or failure of the next vehicle battery start by the starting power system;

[0036] Specifically, the multiple basic data include the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle;

[0037] More specifically, the configuration data of the starting power system are the maximum battery capacity, starting voltage, peak current and system weight of the starting power system;

[0038] And further specifically, the multiple pieces of on-site information of the target vehicle are the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinders, the current vehicle body temperature, and the humidity of the environment where the vehicle body is located;

[0039] In this way, through the full and comprehensive selection of the above-mentioned multiple basic data, the reliability and stability of the prediction results are further guaranteed;

[0040] Technical process three: The AI ​​prediction model designed in technical process one is used to intelligently predict whether the starting power system will fail to start the vehicle battery next time based on multiple basic data selected in technical process two;

[0041] Technical process 4: when it is predicted that the starting power system fails to start the vehicle battery next time, disabling the starting power system and performing a warning operation corresponding to the next start failure to suspend the next start; and when it is predicted that the starting power system succeeds in starting the vehicle battery next time, continuing to enable the starting power system to prepare for the next start;

[0042] In this way, based on the intelligent prediction results of whether the starting power system will fail to start the vehicle battery next time, the starting power system completes the targeted optimization processing of the charging process of the vehicle battery, avoiding entering the invalid charging process of starting failure and reducing damage to the vehicle battery and related vehicle components.

[0043] The key points of the present invention are: based on the intelligent prediction result of whether the vehicle battery will fail to start next time, targeted optimization of the charging mode of the starting power system, specific structural customization of the AI ​​prediction model, and sufficient and comprehensive targeted selection of multiple basic data of the AI ​​prediction model are completed.

[0044] The charging optimization method of the starting power system of the present invention will be specifically described below by way of an embodiment.

[0045] Example 1

[0046] Figure 2 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 1 of the present invention.

[0047] like Figure 2 As shown, the charging optimization method of the starting power system includes the following specific steps:

[0048] Step S11: capturing various configuration data of a starting power system for starting a vehicle battery, wherein the various configuration data of the starting power system are a maximum battery capacity, a starting voltage, a peak current, and a system weight of the starting power system;

[0049] For example, capturing various configuration data of a starting power system for starting a vehicle battery, wherein the various configuration data of the starting power system are a maximum battery capacity, a starting voltage, a peak current, and a system weight of the starting power system, including: the maximum battery capacity of the starting power system can be selected as 23800 mAh, the starting voltage of the starting power system can be selected as 12V, the peak current of the starting power system can be selected as 3000A, and the system weight of the starting power system can be selected as 3.3 pounds;

[0050] Step S12: collecting the vehicle type identification, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located of the target vehicle where the vehicle battery is located as a plurality of on-site information outputs of the target vehicle;

[0051] For example, collecting the vehicle type identification, the total volume of the vehicle engine cylinders, the current vehicle body temperature, and the humidity of the environment where the vehicle body is located of the target vehicle as multiple on-site information outputs of the target vehicle includes: the total volume of the vehicle engine cylinders of the target vehicle where the vehicle battery is located is 8 liters;

[0052] Specifically, the total volume of a vehicle engine cylinder is usually expressed in liters (L), that is, the displacement of the engine, which refers to the volume of space that the piston passes through when it moves from the top dead center to the bottom dead center of the cylinder. Generally, the engine contains multiple cylinders, and the sum of the working volumes of all cylinders constitutes the displacement of the engine;

[0053] Step S13: performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process is completed, and outputting the deep neural network after each learning process as an AI prediction model;

[0054] For example, performing a set number of learning processes on the deep neural network to obtain the deep neural network after each learning process, and outputting the deep neural network after each learning process as the AI ​​prediction model includes: using a numerical simulation mode to complete the simulation of a data processing process of performing a set number of learning processes on the deep neural network to obtain the deep neural network after each learning process, and outputting the deep neural network after each learning process as the AI ​​prediction model;

[0055] Step S14: synchronously inputting the current battery capacity of the starting power system, various configuration data of the starting power system and multiple on-site information of the target vehicle into the AI ​​prediction model to run the AI ​​prediction model and obtain the success or failure mark of the next vehicle battery starting of the starting power system output by the AI ​​prediction model;

[0056] Specifically, the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and the success or failure mark of the next vehicle battery start-up by the starting power system output by the AI ​​prediction model includes: the success or failure mark of the next vehicle battery start-up by the starting power system, the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are all in the form of numerical representation after numerical normalization;

[0057] Step S15: when the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system fails to start the vehicle battery next time, disabling the starting power system and performing a warning operation corresponding to the next start failure to suspend the next start; when the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system succeeds in starting the vehicle battery next time, continuing to enable the starting power system;

[0058] In this way, the charging optimization processing of the starting power system based on the intelligent prediction result of the success or failure mark of the next starting of the vehicle battery is completed;

[0059] Wherein, collecting the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinder, the current vehicle body temperature and the humidity of the environment where the vehicle body is located as multiple on-site information outputs of the target vehicle includes: when the vehicle type identifications of the target vehicle where the vehicle battery is located are 0B01, 0B10 and 0B11 respectively, the vehicle types of the target vehicle where the vehicle battery is located are gasoline vehicles, diesel vehicles and gas vehicles respectively;

[0060] Among them, performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model includes: in each learning process performed on the deep neural network, taking the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle as the input content of the deep neural network item by item, taking the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery as the single output content of the deep neural network, to complete this learning process;

[0061] The method further comprises: performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model; the value of the set number is monotonically positively correlated with the maximum battery capacity of the starting power system;

[0062] For example, the value of the set number is monotonically positively correlated with the maximum battery capacity of the starting power system, including: when the maximum battery capacity of the starting power system is 23,800 mAh, the value of the set number is 150; when the maximum battery capacity of the starting power system is 27,800 mAh, the value of the set number is 300, and so on;

[0063] The method further comprises: performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model. The deep neural network comprises a plurality of hidden layers, a single input layer, and a single output layer, and the number of the plurality of hidden layers is proportional to the peak current of the starting power supply system.

[0064] Among them, performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model also includes: using a numerical conversion function to represent a numerical conversion relationship of a monotonically positive correlation between the value of the set number and the maximum battery capacity of the starting power system;

[0065] And wherein, performing a set number of learning processes on the deep neural network to obtain the deep neural network after completing each learning process, and outputting the deep neural network after completing each learning process as an AI prediction model also includes: in the numerical conversion function, the maximum battery capacity of the starting power system is the input data of the numerical conversion function, and the set number of values ​​is the output data of the numerical conversion function.

[0066] Example 2

[0067] Figure 3 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 2 of the present invention.

[0068] like Figure 3 As shown, Figure 2 Different from the embodiment in, before collecting the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located as the output of multiple on-site information of the target vehicle, that is, before step S12, the method further includes:

[0069] Step S16: a plurality of temperature sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining temperature data obtained after removing the maximum and minimum values ​​of the plurality of temperature data corresponding to the plurality of temperature sensing mechanisms at the current moment is used as the current body temperature of the target vehicle where the vehicle battery is located;

[0070] Wherein, a plurality of temperature sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining temperature data obtained after removing the maximum value and the minimum value of the plurality of temperature data respectively corresponding to the plurality of temperature sensing mechanisms at the current moment is used as the current body temperature of the target vehicle where the vehicle battery is located comprises: the plurality of temperature sensing mechanisms have the same structure;

[0071] Specifically, a plurality of temperature sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining temperature data obtained after removing the maximum value and the minimum value of the multiple temperature data corresponding to the multiple temperature sensing mechanisms at the current moment is used as the current body temperature of the target vehicle where the vehicle battery is located, including: the multiple temperature sensing mechanisms are all non-contact temperature sensing mechanisms.

[0072] Example 3

[0073] Figure 4 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 3 of the present invention.

[0074] like Figure 4 As shown, Figure 2 Different from the embodiment in, before collecting the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located as the output of multiple on-site information of the target vehicle, that is, before step S12, the method further includes:

[0075] Step S17: a plurality of humidity sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining humidity data obtained after removing the maximum and minimum values ​​of the plurality of humidity data corresponding to the plurality of humidity sensing mechanisms at the current moment is used as the environmental humidity of the body of the target vehicle where the vehicle battery is located;

[0076] Wherein, a plurality of humidity sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining humidity data obtained after removing the maximum and minimum values ​​of the plurality of humidity data respectively corresponding to the plurality of humidity sensing mechanisms at the current moment is used as the environmental humidity of the body of the target vehicle where the vehicle battery is located, including: the plurality of humidity sensing mechanisms have the same structure;

[0077] Specifically, a plurality of humidity sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining humidity data obtained after removing the maximum and minimum values ​​of the multiple humidity data corresponding to the multiple humidity sensing mechanisms at the current moment is used as the ambient humidity of the body of the target vehicle where the vehicle battery is located, including: the multiple humidity sensing mechanisms are evenly spaced.

[0078] Example 4

[0079] Figure 5 The present invention is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 4 of the present invention.

[0080] like Figure 5 As shown, Figure 2 Different from the embodiment in, after performing a set number of learning processes on the deep neural network to obtain the deep neural network after each learning process, and outputting the deep neural network after each learning process as the AI ​​prediction model, that is, after step S13, the method further includes:

[0081] Step S18: using a data storage mechanism to store various model parameters of the AI ​​prediction model to implement model storage of the AI ​​prediction model;

[0082] Specifically, the data storage mechanism is used to store various model parameters of the AI ​​prediction model to realize model storage of the AI ​​prediction model, including: the data storage mechanism is a TF storage chip or a FLASH storage chip.

[0083] Example 5

[0084] Figure 6 The figure is a flowchart of the steps of a charging optimization method for a starting power supply system according to Embodiment 5 of the present invention.

[0085] like Figure 6 As shown, Figure 2 Different from the embodiment in, when the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system fails to start the vehicle battery next time, the starting power system is disabled and a warning operation corresponding to the next start failure is performed to suspend the next start, and when the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system successfully starts the vehicle battery next time, after continuing to enable the starting power system, that is, after step S15, the method further includes:

[0086] Step S19: using a field display mechanism to display advance notification information of success or failure of the starting power system to start the vehicle battery next time corresponding to the received success or failure identification mark of the starting power system to start the vehicle battery next time;

[0087] Specifically, the on-site display mechanism is used to display the advance notification information of the success or failure of the next vehicle battery starting by the starting power system corresponding to the received success or failure identification mark of the next vehicle battery starting by the starting power system, including: the on-site display mechanism is an LCD display array or an LED display array.

[0088] Next, various method embodiments of the present invention are described in detail.

[0089] In the charging optimization method of the starting power system according to various method embodiments of the present invention:

[0090] The current battery capacity of the starting power system, various configuration data of the starting power system and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and obtaining the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model also includes: when the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model is OBFF, it indicates that the starting power system fails to start the vehicle battery next time, and when the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model is OB00, it indicates that the starting power system successfully starts the vehicle battery next time;

[0091] Wherein, in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network, and completing this learning process includes: in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark with a value of OB00 or OBFF is used as the single output content of the deep neural network to complete this learning process;

[0092] It can be seen that in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark with a value of OB00 or OBFF is used as the single output content of the deep neural network. Completing this learning process includes: a successful start in history corresponds to a positive training process of the deep neural network, and a failed start in history corresponds to a negative training process of the deep neural network. The number of positive training processes performed on the deep neural network is equal to the number of negative training processes performed on the deep neural network.

[0093] And wherein, in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network. Completing this learning process also includes: using the MATLAB toolbox to complete the testing and simulation of each learning process performed on the deep neural network.

[0094] And in the charging optimization method of the starting power system according to each method embodiment of the present invention:

[0095] The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and the success or failure indicator of the next vehicle battery start-up by the starting power system output by the AI ​​prediction model is obtained, which includes: performing binary value conversion on the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle, respectively, and then synchronously inputting them into the AI ​​prediction model to run the AI ​​prediction model;

[0096] Wherein, the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and obtaining the success or failure mark of the next vehicle battery starting of the starting power system output by the AI ​​prediction model also includes: using a synchronous drive mechanism to drive the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle to be synchronously input into the AI ​​prediction model;

[0097] And wherein, a synchronous driving mechanism is used to drive the current battery capacity of the starting power system, various configuration data of the starting power system and multiple on-site information of the target vehicle to the synchronous input of the AI ​​prediction model, including: the synchronous driving mechanism is a CPLD chip designed using VHDL language.

[0098] And in the charging optimization method of the starting power system according to each method embodiment of the present invention:

[0099] The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are converted into binary numbers respectively and then synchronously input into the AI ​​prediction model to run the AI ​​prediction model, including: using a numerical conversion interface to perform binary number conversion of the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle.

[0100] In addition, the present invention may also cite the following technical contents to highlight the significant technical progress of the present invention:

[0101] The deep neural network includes a plurality of hidden layers, a single input layer and a single output layer, and the number of the plurality of hidden layers is proportional to the peak current of the starting power system, including: using a content mapping formula to express a content mapping relationship in which the number of the plurality of hidden layers is proportional to the peak current of the starting power system;

[0102] For example, when a content mapping formula is used to represent a content mapping relationship in which the number of the plurality of hidden layers is proportional to the peak current of the starting power system, the peak current of the starting power system is an input parameter of the content mapping formula, and the number of the plurality of hidden layers corresponding to the peak current of the starting power system is an output parameter of the content mapping formula;

[0103] Among them, performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model also includes: selecting a programmable logic device to implement a data processing process of performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model;

[0104] For example, a programmable logic device is selected to implement a set number of learning processes on a deep neural network to obtain a deep neural network after each learning process, and the deep neural network after each learning process is used as a data processing process output by an AI prediction model, including: the programmable logic device is an FPGA device.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A charging optimization method for a starting power supply system, characterized in that: The method comprises: Capturing various configuration data of a starting power system for starting a vehicle battery, wherein the various configuration data of the starting power system are a maximum battery capacity, a starting voltage, a peak current, and a system weight of the starting power system; Collecting the vehicle type identification, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located of the target vehicle where the vehicle battery is located as a plurality of on-site information outputs of the target vehicle; Performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process is completed, and outputting the deep neural network after each learning process as an AI prediction model; The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model and obtain the success or failure indicator of the next vehicle battery starting of the starting power system output by the AI ​​prediction model; When the received success or failure identification mark of the starting power system for the next time starting the vehicle battery indicates that the starting power system fails to start the vehicle battery next time, disabling the starting power system and performing a warning operation corresponding to the next start failure to suspend the next start; when the received success or failure identification mark of the starting power system for the next time starting the vehicle battery indicates that the starting power system succeeds in starting the vehicle battery next time, continuing to enable the starting power system; Among them, the vehicle type identification of the target vehicle where the vehicle battery is located, the total volume of the vehicle engine cylinders, the current vehicle body temperature and the humidity of the environment where the vehicle body is located are collected as multiple on-site information outputs of the target vehicle, including: when the vehicle type identifications of the target vehicles where the vehicle battery is located are 0B01, 0B10 and 0B11 respectively, the vehicle types of the target vehicles where the vehicle battery is located are gasoline vehicles, diesel vehicles and gas vehicles respectively.

2. The charging optimization method of the starting power system according to claim 1, characterized in that: Performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model includes: in each learning process performed on the deep neural network, taking the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle as the input content of the deep neural network item by item, taking the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery as the single output content of the deep neural network, to complete this learning process; Among them, performing a set number of learning processes on the deep neural network to obtain the deep neural network after completing each learning process, and outputting the deep neural network after completing each learning process as an AI prediction model also includes: the value of the set number is monotonically positively correlated with the maximum battery capacity of the starting power system.

3. The charging optimization method of the starting power system according to claim 2, characterized in that: Performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model also includes: the deep neural network includes multiple hidden layers, a single input layer and a single output layer, and the number of the multiple hidden layers is proportional to the peak current of the starting power supply system; Among them, performing a set number of learning processes on the deep neural network to obtain a deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model also includes: using a numerical conversion function to represent a numerical conversion relationship of a monotonically positive correlation between the value of the set number and the maximum battery capacity of the starting power system; Among them, performing a set number of learning processes on the deep neural network to obtain the deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model also includes: in the numerical conversion function, the maximum battery capacity of the starting power system is the input data of the numerical conversion function, and the set number of values ​​is the output data of the numerical conversion function.

4. The charging optimization method of the starting power system according to claim 3, characterized in that: Before collecting the vehicle type identification, the total cylinder volume of the vehicle engine, the current vehicle body temperature, and the humidity of the environment where the vehicle body is located of the target vehicle where the vehicle battery is located as the output of multiple on-site information of the target vehicle, the method further includes: A plurality of temperature sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining temperature data obtained after removing the maximum and minimum values ​​of the plurality of temperature data corresponding to the plurality of temperature sensing mechanisms at the current moment is used as the current body temperature of the target vehicle where the vehicle battery is located; Among them, multiple temperature sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining temperature data obtained after removing the maximum value and the minimum value of multiple temperature data corresponding to the multiple temperature sensing mechanisms at the current moment is used as the current body temperature of the target vehicle where the vehicle battery is located, including: the multiple temperature sensing mechanisms have the same structure.

5. The charging optimization method of the starting power system according to claim 3, characterized in that: Before collecting the vehicle type identification, the total cylinder volume of the vehicle engine, the current vehicle body temperature, and the humidity of the environment where the vehicle body is located of the target vehicle where the vehicle battery is located as the output of multiple on-site information of the target vehicle, the method further includes: A plurality of humidity sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining humidity data obtained after removing the maximum and minimum values ​​of the plurality of humidity data corresponding to the plurality of humidity sensing mechanisms at the current moment is used as the environmental humidity of the body of the target vehicle where the vehicle battery is located; Among them, multiple humidity sensing mechanisms are respectively arranged at different positions of the body of the target vehicle, and the average of the remaining humidity data obtained after removing the maximum and minimum values ​​of multiple humidity data corresponding to the multiple humidity sensing mechanisms at the current moment is used as the ambient humidity of the body of the target vehicle where the vehicle battery is located, including: the multiple humidity sensing mechanisms have the same structure.

6. The charging optimization method of the starting power system according to claim 3, characterized in that: After performing a set number of learning processes on the deep neural network to obtain the deep neural network after each learning process, and outputting the deep neural network after each learning process as an AI prediction model, the method further includes: A data storage mechanism is used to store various model parameters of the AI ​​prediction model to achieve model storage of the AI ​​prediction model.

7. The charging optimization method of the starting power system according to claim 3, characterized in that: When the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system fails to start the vehicle battery next time, disabling the starting power system and performing a warning operation corresponding to the next start failure to suspend the next start; when the received success or failure identification mark of the starting power system for starting the vehicle battery next time indicates that the starting power system successfully starts the vehicle battery next time, continuing to enable the starting power system, the method further includes: An on-site display mechanism is used to display advance notification information of success or failure of the starting power system to start the vehicle battery next time, corresponding to the received success or failure identification mark of the starting power system to start the vehicle battery next time.

8. The charging optimization method of the starting power system according to claim 7, characterized in that: The current battery capacity of the starting power system, various configuration data of the starting power system and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and obtaining the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model also includes: when the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model is OBFF, it indicates that the starting power system fails to start the vehicle battery next time, and when the success or failure mark of the starting power system for starting the vehicle battery next time output by the AI ​​prediction model is OB00, it indicates that the starting power system successfully starts the vehicle battery next time; Wherein, in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network, and completing this learning process includes: in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system, and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark with a value of OB00 or OBFF is used as the single output content of the deep neural network to complete this learning process; Among them, in each learning process performed on the deep neural network, the current battery capacity of the starting power system before a successful or failed start of the vehicle battery in history, various configuration data of the starting power system and multiple on-site information of the target vehicle are used as the input content of the deep neural network item by item, and the success or failure mark indicating the success or failure of the starting power system to start the vehicle battery is used as the single output content of the deep neural network. Completing this learning process also includes: using the MATLAB toolbox to complete the testing and simulation of each learning process performed by the deep neural network.

9. The charging optimization method of the starting power system according to claim 7, characterized in that: The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and the success or failure indicator of the next vehicle battery start-up by the starting power system output by the AI ​​prediction model is obtained, which includes: performing binary value conversion on the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle, respectively, and then synchronously inputting them into the AI ​​prediction model to run the AI ​​prediction model; Wherein, the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are synchronously input into the AI ​​prediction model to run the AI ​​prediction model, and obtaining the success or failure mark of the next vehicle battery starting of the starting power system output by the AI ​​prediction model also includes: using a synchronous drive mechanism to drive the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle to be synchronously input into the AI ​​prediction model; Among them, the synchronous driving mechanism is used to drive the current battery capacity of the starting power system, various configuration data of the starting power system and multiple on-site information of the target vehicle to the synchronous input of the AI ​​prediction model, including: the synchronous driving mechanism is a CPLD chip designed with VHDL language.

10. The charging optimization method of the starting power system according to claim 9, characterized in that: The current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle are converted into binary numbers respectively and then synchronously input into the AI ​​prediction model to run the AI ​​prediction model, including: using a numerical conversion interface to perform binary number conversion of the current battery capacity of the starting power system, various configuration data of the starting power system, and multiple on-site information of the target vehicle.

Images