Intelligent control system and method of DC liquid cooling charging based on deep learning

Abstract

The present invention discloses a DC liquid-cooled charging intelligent control system and method based on deep learning, which relates to the field of DC liquid-cooled charging technology. The control system includes a target monitoring period division module, a storage label establishment module, a liquid-cooled charging monitoring index analysis module, a charging safety monitoring module establishment module and a safety early warning evaluation module; the target monitoring period division module is used to extract the historical charging data records of the target charging device, and divide the historical charging data records into target monitoring periods; the storage label establishment module is used to store the charging events recorded by the charging device within the target monitoring period in the corresponding monitoring unit with each vehicle as the storage label; the liquid-cooled charging monitoring index analysis module is used to analyze the liquid-cooled charging monitoring index corresponding to each storage label; the charging safety monitoring module establishment module is used to construct a charging safety monitoring model corresponding to each storage label.

Description

Technical Field

[0001] The present invention relates to the field of direct current liquid-cooled charging technology, and in particular to a direct current liquid-cooled charging intelligent control system and method based on deep learning. Background Art

[0002] With the vigorous development of the electric vehicle industry, charging efficiency and infrastructure construction have become one of the key factors restricting the rapid growth of the industry. In this context, the 600KW DC group charging (liquid-cooled super-charging) charging pile came into being. Liquid-cooled super-charging technology, that is, liquid-cooled ultra-high power charging technology, is an efficient charging solution achieved through a liquid cooling mechanism. Its working principle is to effectively take away the heat generated during the charging process through a liquid circulation system, thereby achieving a fast and stable charging process; but at the same time, this efficient charging method also brings some problems, such as the liquid cooling method is still based on the principle of temperature difference heat exchange, and the heat transfer is a sensible heat method, so it is impossible to avoid the temperature difference problem inside the battery module. The temperature difference causes overcharging, over-discharging or insufficient charging and discharging of different single batteries in the module during the charging and discharging process. Overcharging and over-discharging of the battery will lead to battery safety problems and reduce battery life. Insufficient charging and discharging will reduce battery energy density and reduce battery life. Although there is an intelligent battery monitoring system in the prior art to achieve effective monitoring of local temperature based on sensors, it is only a simple adjustment of the coolant based on the temperature difference, and it cannot effectively perform battery safety assessment and early warning for a large number of vehicles based on liquid-cooled charging. Summary of the invention

[0003] The purpose of the present invention is to provide a DC liquid-cooled charging intelligent control system and method based on deep learning to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: a DC liquid-cooled charging intelligent control method based on deep learning, the method comprising the following steps:

[0005] Step S100: marking a charging device based on DC liquid cooling technology as a target charging device, extracting historical charging data records of the target charging device, and dividing the historical charging data records into target monitoring periods; the charging device is equipped with an intelligent battery management system, which monitors the temperature distribution of the battery pack in real time and adjusts the flow and distribution of the coolant as needed;

[0006] Step S200: storing the charging events recorded by the charging equipment within the target monitoring period in the corresponding monitoring unit with each vehicle as a storage tag; and analyzing the liquid cooling charging monitoring index Z corresponding to each storage tag; the charging event refers to the charging event in response to the intelligent battery management system implementing the coolant adjustment;

[0007] Step S300: constructing a charging safety monitoring model corresponding to each storage tag based on the liquid cooling charging monitoring index Z and in combination with historical charging data records;

[0008] Step S400: extract the charging safety monitoring model corresponding to all storage tags recorded by each charging device, generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

[0009] Furthermore, the division of the historical charging data records into target monitoring periods includes the following specific processes:

[0010] The period from the time of the first charging event recorded in the same charging device for each vehicle to the time when the charging efficiency Q is less than the charging efficiency threshold is extracted as the first monitoring period; and the charging event where the charging efficiency Q is less than the charging efficiency threshold is marked as the terminal charging event; the charging device recorded in the terminal charging event is the initial device;

[0011] The calculation process of the charging efficiency Q is as follows:

[0012] The historical charging data record includes the vehicle charging time T, the initial power W1 and the final power W2; Q = (W2-W1) / T;

[0013] Mark the charging event of the vehicle at different charging devices as a charging event to be analyzed, and extract the charging efficiency Q1 of the next charging event to be analyzed adjacent to the endpoint charging event; when Q1>Q, increase the monitoring time until the charging efficiency Q of the initial device is greater than or equal to the charging efficiency Q recorded by the adjacent different charging devices and the difference is less than the difference threshold, and the charging efficiency Q of the initial device is less than the charging efficiency threshold, then lock the charging event recorded by the corresponding initial device as the updated endpoint charging event;

[0014] and taking the updated endpoint charging event as the end time of the target monitoring cycle, taking the time of the first charging event of the first monitoring cycle as the initial time, and updating the first monitoring cycle to the target monitoring cycle;

[0015] When Q1≤Q, the first monitoring period is output as the target monitoring period.

[0016] Furthermore, the step S200 includes the following steps:

[0017] Step S210: a storage tag records charging events of the same vehicle in the same charging device in the target monitoring period, wherein the charging events record the temperature data of the battery pack and the coolant flow adjustment data; the temperature data of the battery pack is obtained by sensors distributed on the battery packs at different positions, and the coolant flow adjustment data includes an adjustment flow value and an adjustment path; the adjustment path is obtained by responding to the sensors on the battery pack;

[0018] Step S220: extract the battery pack temperature difference value E1, the adjusted path length L1, the adjusted flow value R1 and the temperature balance response time U1 recorded in each charging event; where E1=E max -E min , E max Indicates that the sensor obtains the maximum temperature of the battery pack in the battery pack during the charging event, E min Indicates that the sensor obtains the minimum temperature of the battery pack in the battery pack during the charging event, the adjustment path length L1 refers to the maximum distance of the coolant flowing through the battery pack, and the temperature balance response time U1 refers to the time from the moment when the intelligent battery management system responds to the coolant adjustment to the moment when the battery pack restores the temperature balance;

[0019] Step S230: Normalize E1, L1, R1 and U1 to obtain the corresponding normalized values ​​E0, L0, R0 and U0, using the formula:

[0020] Z=k1×(U0 / E0)+k2×(R0 / L0);

[0021] Calculate the liquid cooling charging monitoring index Z corresponding to each storage tag, where k1 and k2 represent corresponding reference coefficients, which are set by the system.

[0022] Furthermore, the step S300 includes the following specific processes:

[0023] Sort the liquid cooling charging monitoring index Z recorded for each vehicle within the target monitoring period according to the time sequence of the charging events, and extract the charging efficiency Q1 under the corresponding charging event for each liquid cooling charging monitoring index Z in the sequence; bind the charging efficiency Q1 with the liquid cooling charging monitoring index Z under the same charging event to form a data pair A, A=(Z,Q1), and obtain all data pairs within the target monitoring period;

[0024] A charging safety monitoring model q is established, q=a×z1+ε, where a represents the reference coefficient corresponding to the liquid-cooled charging monitoring index Z, ε represents the error term, and all data pairs A and Z within the target monitoring period are substituted as input items and Q1 as output items; a charging safety monitoring model containing specific error terms and reference coefficients is obtained.

[0025] Furthermore, the step S400 includes the following specific processes:

[0026] Obtain the reference coefficients in the charging safety monitoring model corresponding to all storage tags recorded by each charging device, calculate the coefficient average a0 and the error term average ε0, and generate the updated safety assessment model Y of the corresponding charging device, Y=a0×z1+ε0;

[0027] Select E0, L0, R0 and U0 recorded during real-time vehicle charging, solve to obtain the real-time liquid cooling charging monitoring index, and substitute it into the safety assessment model Y to calculate the output value;

[0028] The actual charging efficiency of the real-time vehicle is obtained. When the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; when the actual charging efficiency is less than the output value, the intelligent battery management system outputs an abnormal battery status signal to remind the charging vehicle.

[0029] A DC liquid-cooled charging intelligent control system based on deep learning, characterized in that: the control system includes a target monitoring period division module, a storage label establishment module, a liquid-cooled charging monitoring index analysis module, a charging safety monitoring module establishment module and a safety early warning evaluation module;

[0030] The target monitoring period division module is used to extract the historical charging data records of the target charging device and divide the historical charging data records into target monitoring periods;

[0031] The storage tag establishment module is used to store the charging events recorded by the charging equipment within the target monitoring period into the corresponding monitoring unit with each vehicle as a storage tag;

[0032] The liquid cooling charging monitoring index analysis module is used to analyze the liquid cooling charging monitoring index corresponding to each storage tag;

[0033] The charging safety monitoring module establishment module is used to construct a charging safety monitoring model corresponding to each storage tag;

[0034] The safety warning assessment module is used to generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

[0035] Further, the target monitoring period division module includes a charging efficiency calculation unit, a first monitoring period determination unit and a target monitoring period determination unit;

[0036] The charging efficiency calculation unit is used to calculate the charging efficiency based on the vehicle charging time, initial power and final power;

[0037] The first monitoring period determination unit is used to extract the period corresponding to the time when each vehicle records the first charging event in the same charging device to the time when the charging efficiency is less than the charging efficiency threshold as the first monitoring period;

[0038] The target monitoring period determination unit is used to analyze the charging efficiency difference and determine the target monitoring period.

[0039] Further, the liquid-cooled charging monitoring index analysis module includes a charging data extraction unit and a liquid-cooled charging monitoring index calculation unit;

[0040] The charging data extraction unit is used to extract the battery pack temperature difference value, adjustment path length, adjustment flow value and temperature balance response time recorded in each charging event;

[0041] The liquid-cooled charging monitoring index calculation unit is used to normalize the battery pack temperature difference value, the adjustment path length, the adjustment flow value and the temperature balance response time, and calculate the liquid-cooled charging monitoring index corresponding to each storage tag.

[0042] The safety early warning evaluation module includes a safety monitoring model updating unit and a real-time data input evaluation unit;

[0043] The safety monitoring model updating unit is used to update the reference coefficient and error term based on the safety monitoring model to build a safety assessment model;

[0044] The real-time data input evaluation unit is used to substitute into the safety evaluation model to calculate the output value; obtain the real-time actual charging efficiency of the vehicle, and when the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; when the actual charging efficiency is less than the output value, the intelligent battery management system outputs an abnormal battery status signal to alert the charging vehicle.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. The present invention firstly accurately extracts the target monitoring period to ensure that the error of the charging event variable in analyzing the target vehicle is reduced, and realizes the effective analysis of the data of the charging process of the target vehicle historical record;

[0047] 2. The present invention will use the efficiency data and characteristic data in the DC liquid-cooled charging process for combined analysis, build a safety monitoring model for the associated vehicle battery itself, and then correlate them based on the charging equipment to achieve effective model control of the charging equipment, so as to achieve real-time safety status assessment and analysis of different types of charging vehicles, effectively assess the impact of uniform temperature distribution on vehicle battery safety, and achieve timely warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1This is a structural schematic diagram of the DC liquid-cooled charging intelligent control system based on deep learning of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example: Figure 1 As shown, the present invention provides a DC liquid-cooled charging intelligent control system and method technical solution based on deep learning, and a DC liquid-cooled charging intelligent control method based on deep learning, the method comprising the following steps:

[0051] Step S100: marking a charging device based on DC liquid cooling technology as a target charging device, extracting historical charging data records of the target charging device, and dividing the historical charging data records into target monitoring periods; the charging device is equipped with an intelligent battery management system, which monitors the temperature distribution of the battery pack in real time and adjusts the flow and distribution of the coolant as needed;

[0052] Step S200: storing the charging events recorded by the charging equipment within the target monitoring period in the corresponding monitoring unit with each vehicle as a storage tag; and analyzing the liquid cooling charging monitoring index Z corresponding to each storage tag; the charging event refers to the charging event in response to the intelligent battery management system implementing the coolant adjustment;

[0053] Step S300: constructing a charging safety monitoring model corresponding to each storage tag based on the liquid cooling charging monitoring index Z and in combination with historical charging data records;

[0054] Step S400: extract the charging safety monitoring model corresponding to all storage tags recorded by each charging device, generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

[0055] The process of dividing the historical charging data records into target monitoring periods includes the following specific processes:

[0056] The period from the time of the first charging event recorded in the same charging device for each vehicle to the time when the charging efficiency Q is less than the charging efficiency threshold is extracted as the first monitoring period; and the charging event where the charging efficiency Q is less than the charging efficiency threshold is marked as the terminal charging event; the charging device recorded in the terminal charging event is the initial device;

[0057] The calculation process of the charging efficiency Q is as follows:

[0058] The historical charging data record includes the vehicle charging time T, the initial power W1 and the final power W2; Q = (W2-W1) / T;

[0059] Mark the charging event of the vehicle at different charging devices as a charging event to be analyzed, and extract the charging efficiency Q1 of the next charging event to be analyzed adjacent to the endpoint charging event; when Q1>Q, increase the monitoring time until the charging efficiency Q of the initial device is greater than or equal to the charging efficiency Q recorded by the adjacent different charging devices and the difference is less than the difference threshold, and the charging efficiency Q of the initial device is less than the charging efficiency threshold, then lock the charging event recorded by the corresponding initial device as the updated endpoint charging event;

[0060] and taking the updated endpoint charging event as the end time of the target monitoring cycle, taking the time of the first charging event of the first monitoring cycle as the initial time, and updating the first monitoring cycle to the target monitoring cycle;

[0061] When Q1≤Q, the first monitoring period is output as the target monitoring period.

[0062] When analyzing the target monitoring cycle, locking the same vehicle to the same charging device can effectively reduce the monitoring errors caused by objective factors. At the same time, after determining the first monitoring cycle, by looking for charging events of the same vehicle recorded on different charging devices, it is possible to more accurately determine whether the charging efficiency is reduced due to problems with the charging device itself, thereby reducing errors in analyzing battery pack safety monitoring judgments and achieving accurate analysis.

[0063] The step S200 includes the following steps:

[0064] Step S210: a storage tag records charging events of the same vehicle in the same charging device in the target monitoring period, wherein the charging events record the temperature data of the battery pack and the coolant flow adjustment data; the temperature data of the battery pack is obtained by sensors distributed on the battery packs at different positions, and the coolant flow adjustment data includes an adjustment flow value and an adjustment path; the adjustment path is obtained by responding to the sensors on the battery pack;

[0065] Step S220: extract the battery pack temperature difference value E1, the adjusted path length L1, the adjusted flow value R1 and the temperature balance response time U1 recorded in each charging event; where E1=E max -E min , E max Indicates that the sensor obtains the maximum temperature of the battery pack in the battery pack during the charging event, E minIndicates that the sensor obtains the minimum temperature of the battery pack in the battery pack during the charging event, the adjustment path length L1 refers to the maximum distance of the coolant flowing through the battery pack, and the temperature balance response time U1 refers to the time length from the moment when the intelligent battery management system responds to the coolant adjustment to the moment when the battery pack restores the temperature balance; restoring the temperature balance means that the temperature difference between the battery packs in the battery pack is less than the temperature difference threshold;

[0066] Step S230: Normalize E1, L1, R1 and U1 to obtain the corresponding normalized values ​​E0, L0, R0 and U0, using the formula:

[0067] Z=k1×(U0 / E0)+k2×(R0 / L0);

[0068] Calculate the liquid cooling charging monitoring index Z corresponding to each storage tag, where k1 and k2 represent corresponding reference coefficients, which are set by the system.

[0069] (U0 / E0) indicates the response time required for a unit temperature change. A larger value indicates that it is more difficult or consumes more energy for the intelligent battery management system to adjust the battery temperature of the vehicle in the charging event. Similarly, (R0 / L0) indicates the coolant flow required for a unit adjustment distance change. A larger value indicates that the degree of adjustment required is deeper and the system consumes more. The larger the liquid-cooled charging monitoring index Z, the greater the adjustment effort and the higher the difficulty of the vehicle in the corresponding charging event.

[0070] The step S300 includes the following specific processes:

[0071] Sort the liquid cooling charging monitoring index Z recorded for each vehicle within the target monitoring period according to the time sequence of the charging events, and extract the charging efficiency Q1 under the corresponding charging event for each liquid cooling charging monitoring index Z in the sequence; bind the charging efficiency Q1 with the liquid cooling charging monitoring index Z under the same charging event to form a data pair A, A=(Z,Q1), and obtain all data pairs within the target monitoring period;

[0072] A charging safety monitoring model q is established, q=a×z1+ε, where a represents the reference coefficient corresponding to the liquid-cooled charging monitoring index Z, ε represents the error term, and all data pairs A and Z within the target monitoring period are substituted as input items and Q1 as output items; a charging safety monitoring model containing specific error terms and reference coefficients is obtained.

[0073] The step S400 includes the following specific processes:

[0074] Obtain the reference coefficients in the charging safety monitoring model corresponding to all storage tags recorded by each charging device, calculate the coefficient average a0 and the error term average ε0, and generate the updated safety assessment model Y of the corresponding charging device, Y=a0×z1+ε0;

[0075] Select E0, L0, R0 and U0 recorded during real-time vehicle charging, solve to obtain the real-time liquid cooling charging monitoring index, and substitute it into the safety assessment model Y to calculate the output value;

[0076] The actual charging efficiency of the real-time vehicle is obtained. When the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; when the actual charging efficiency is less than the output value, the intelligent battery management system outputs an abnormal battery status signal to remind the charging vehicle.

[0077] A DC liquid-cooled charging intelligent control system based on deep learning, characterized in that: the control system includes a target monitoring period division module, a storage label establishment module, a liquid-cooled charging monitoring index analysis module, a charging safety monitoring module establishment module and a safety early warning evaluation module;

[0078] The target monitoring period division module is used to extract the historical charging data records of the target charging device and divide the historical charging data records into target monitoring periods;

[0079] The storage tag establishment module is used to store the charging events recorded by the charging equipment within the target monitoring period into the corresponding monitoring unit with each vehicle as a storage tag;

[0080] The liquid cooling charging monitoring index analysis module is used to analyze the liquid cooling charging monitoring index corresponding to each storage tag;

[0081] The charging safety monitoring module establishment module is used to construct a charging safety monitoring model corresponding to each storage tag;

[0082] The safety warning assessment module is used to generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

[0083] The target monitoring period division module includes a charging efficiency calculation unit, a first monitoring period determination unit and a target monitoring period determination unit;

[0084] The charging efficiency calculation unit is used to calculate the charging efficiency based on the vehicle charging time, initial power and final power;

[0085] The first monitoring period determination unit is used to extract the period corresponding to the time when each vehicle records the first charging event in the same charging device to the time when the charging efficiency is less than the charging efficiency threshold as the first monitoring period;

[0086] The target monitoring period determination unit is used to analyze the charging efficiency difference and determine the target monitoring period.

[0087] The liquid-cooled charging monitoring index analysis module includes a charging data extraction unit and a liquid-cooled charging monitoring index calculation unit;

[0088] The charging data extraction unit is used to extract the battery pack temperature difference value, adjustment path length, adjustment flow value and temperature balance response time recorded in each charging event;

[0089] The liquid-cooled charging monitoring index calculation unit is used to normalize the battery pack temperature difference value, the adjustment path length, the adjustment flow value and the temperature balance response time, and calculate the liquid-cooled charging monitoring index corresponding to each storage tag.

[0090] The safety early warning evaluation module includes a safety monitoring model updating unit and a real-time data input evaluation unit;

[0091] The safety monitoring model updating unit is used to update the reference coefficient and error term based on the safety monitoring model to build a safety assessment model;

[0092] The real-time data input evaluation unit is used to substitute into the safety evaluation model to calculate the output value; obtain the real-time actual charging efficiency of the vehicle, and when the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; when the actual charging efficiency is less than the output value, the intelligent battery management system outputs an abnormal battery status signal to alert the charging vehicle.

[0093] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A DC liquid-cooled charging intelligent control method based on deep learning, characterized in that: The method comprises the following steps: Step S100: marking a charging device based on DC liquid cooling technology as a target charging device, extracting historical charging data records of the target charging device, and dividing the historical charging data records into target monitoring periods; the charging device is equipped with an intelligent battery management system, which monitors the temperature distribution of the battery pack in real time and adjusts the flow and distribution of the coolant as needed; Step S200: storing the charging events recorded by the charging equipment within the target monitoring period in the corresponding monitoring unit with each vehicle as a storage tag; and analyzing the liquid cooling charging monitoring index Z corresponding to each storage tag; the charging event refers to the charging event in response to the intelligent battery management system implementing the coolant adjustment; Step S300: constructing a charging safety monitoring model corresponding to each storage tag based on the liquid cooling charging monitoring index Z and in combination with historical charging data records; Step S400: extract the charging safety monitoring model corresponding to all storage tags recorded by each charging device, generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

2. According to the deep learning-based DC liquid cooling charging intelligent control method of claim 1, it is characterized by: The process of dividing the historical charging data records into target monitoring periods includes the following specific processes: The period from the time of the first charging event recorded in the same charging device for each vehicle to the time when the charging efficiency Q is less than the charging efficiency threshold is extracted as the first monitoring period; and the charging event where the charging efficiency Q is less than the charging efficiency threshold is marked as the terminal charging event; the charging device recorded in the terminal charging event is the initial device; The calculation process of the charging efficiency Q is as follows: The historical charging data record includes the vehicle charging time T, the initial power W1 and the final power W2; Q = (W2-W1) / T; Mark the charging event of the vehicle at different charging devices as a charging event to be analyzed, and extract the charging efficiency Q1 of the next charging event to be analyzed adjacent to the endpoint charging event; when Q1>Q, increase the monitoring time until the charging efficiency Q of the initial device is greater than or equal to the charging efficiency Q recorded by the adjacent different charging devices and the difference is less than the difference threshold, and the charging efficiency Q of the initial device is less than the charging efficiency threshold, then lock the charging event recorded by the corresponding initial device as the updated endpoint charging event; and taking the updated endpoint charging event as the end time of the target monitoring cycle, taking the time of the first charging event of the first monitoring cycle as the initial time, and updating the first monitoring cycle to the target monitoring cycle; When Q1≤Q, the first monitoring period is output as the target monitoring period.

3. The DC liquid-cooled charging intelligent control method based on deep learning according to claim 2 is characterized in that: The step S200 includes the following steps: Step S210: a storage tag records charging events of the same vehicle in the same charging device in the target monitoring period, wherein the charging events record the temperature data of the battery pack and the coolant flow adjustment data; the temperature data of the battery pack is obtained by sensors distributed on the battery packs at different positions, and the coolant flow adjustment data includes an adjustment flow value and an adjustment path; the adjustment path is obtained by responding to the sensors on the battery pack; Step S220: extract the battery pack temperature difference value E1, the adjusted path length L1, the adjusted flow value R1 and the temperature balance response time U1 recorded in each charging event; where E1=E max -E min , E max Indicates that the sensor obtains the maximum temperature of the battery pack in the battery pack during the charging event, E min Indicates that the sensor obtains the minimum temperature of the battery pack in the battery pack during the charging event, the adjustment path length L1 refers to the maximum distance of the coolant flowing through the battery pack, and the temperature balance response time U1 refers to the time from the moment when the intelligent battery management system responds to the coolant adjustment to the moment when the battery pack restores the temperature balance; Step S230: Normalize E1, L1, R1 and U1 to obtain the corresponding normalized values ​​E0, L0, R0 and U0, using the formula: Z=k1×(U0 / E0)+k2×(R0 / L0); Calculate the liquid cooling charging monitoring index Z corresponding to each storage tag, where k1 and k2 represent corresponding reference coefficients, which are set by the system.

4. The DC liquid-cooled charging intelligent control method based on deep learning according to claim 3 is characterized in that: The step S300 includes the following specific processes: Sort the liquid cooling charging monitoring index Z recorded for each vehicle during the target monitoring period according to the time sequence of the charging events, and extract the charging efficiency Q1 under the corresponding charging event for each liquid cooling charging monitoring index Z in the sequence; Bind the charging efficiency Q1 with the liquid cooling charging monitoring index Z corresponding to the same charging event to form a data pair A, A=(Z,Q1), and obtain all data pairs within the target monitoring period; A charging safety monitoring model q is established, q=a×z1+ε, where a represents the reference coefficient corresponding to the liquid-cooled charging monitoring index Z, ε represents the error term, and all data pairs A and Z within the target monitoring period are substituted as input items and Q1 as output items; a charging safety monitoring model containing specific error terms and reference coefficients is obtained.

5. The DC liquid cooling charging intelligent control method based on deep learning according to claim 4 is characterized in that: The step S400 includes the following specific processes: Obtain the reference coefficients in the charging safety monitoring model corresponding to all storage tags recorded by each charging device, calculate the coefficient average a0 and the error term average ε0, and generate the updated safety assessment model Y of the corresponding charging device, Y=a0×z1+ε0; Select E0, L0, R0 and U0 recorded during real-time vehicle charging, solve to obtain the real-time liquid cooling charging monitoring index, and substitute it into the safety assessment model Y to calculate the output value; Obtain the actual charging efficiency of the real-time vehicle. When the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; When the actual charging efficiency is less than the output value, the intelligent battery management system outputs a battery status abnormality signal to alert the charging vehicle.

6. A DC liquid-cooled charging intelligent control system based on deep learning, such as using a DC liquid-cooled charging intelligent control method based on deep learning as claimed in any one of claims 1 to 5, characterized in that: The control system includes a target monitoring cycle division module, a storage tag establishment module, a liquid cooling charging monitoring index analysis module, a charging safety monitoring module establishment module and a safety early warning evaluation module; The target monitoring period division module is used to extract the historical charging data records of the target charging device and divide the historical charging data records into target monitoring periods; The storage tag establishment module is used to store the charging events recorded by the charging equipment within the target monitoring period into the corresponding monitoring unit with each vehicle as a storage tag; The liquid cooling charging monitoring index analysis module is used to analyze the liquid cooling charging monitoring index corresponding to each storage tag; The charging safety monitoring module establishment module is used to construct a charging safety monitoring model corresponding to each storage tag; The safety warning assessment module is used to generate a safety assessment model for each DC liquid-cooled charging device, and respond to the safety assessment model to issue a safety warning when the vehicle is charging.

7. The DC liquid-cooled charging intelligent control system based on deep learning according to claim 6 is characterized in that: The target monitoring period division module includes a charging efficiency calculation unit, a first monitoring period determination unit and a target monitoring period determination unit; The charging efficiency calculation unit is used to calculate the charging efficiency based on the vehicle charging time, initial power and final power; The first monitoring period determination unit is used to extract the period corresponding to the time when each vehicle records the first charging event in the same charging device to the time when the charging efficiency is less than the charging efficiency threshold as the first monitoring period; The target monitoring period determination unit is used to analyze the charging efficiency difference and determine the target monitoring period.

8. The DC liquid-cooled charging intelligent control system based on deep learning according to claim 6 is characterized in that: The liquid-cooled charging monitoring index analysis module includes a charging data extraction unit and a liquid-cooled charging monitoring index calculation unit; The charging data extraction unit is used to extract the battery pack temperature difference value, adjustment path length, adjustment flow value and temperature balance response time recorded in each charging event; The liquid-cooled charging monitoring index calculation unit is used to normalize the battery pack temperature difference value, the adjustment path length, the adjustment flow value and the temperature balance response time, and calculate the liquid-cooled charging monitoring index corresponding to each storage tag.

9. The DC liquid-cooled charging intelligent control system based on deep learning according to claim 6 is characterized in that: The safety early warning evaluation module includes a safety monitoring model updating unit and a real-time data input evaluation unit; The safety monitoring model updating unit is used to update the reference coefficient and error term based on the safety monitoring model to build a safety assessment model; The real-time data input evaluation unit is used to substitute the data into the safety evaluation model to calculate the output value; Obtain the actual charging efficiency of the real-time vehicle. When the actual charging efficiency is greater than or equal to the output value, the intelligent battery management system outputs a normal battery status signal; When the actual charging efficiency is less than the output value, the intelligent battery management system outputs a battery status abnormality signal to alert the charging vehicle.

Images