High-power direct current charger system capable of fast charging and fast charging method

Abstract

The application provides a high-power direct-current charger system and a fast charging method, and relates to the technical field of charging.The controller can count the N+1th variance of the N+1th residual power of the charging battery of the vehicle corresponding to the plurality of accessed high-power direct-current chargers; the difference between the Nth variance and the N+1th variance is determined as the Nth variance reduction rate; if the difference between the Nth variance reduction rate and the N-1th variance reduction rate is greater than the set rate threshold, the network parameters configured by the power distribution network model are updated by using the reinforcement learning model.The controller controls the power distribution unit to respectively distribute charging power to the plurality of accessed high-power direct-current chargers according to the power distribution strategy, so that the plurality of accessed high-power direct-current chargers respectively charge the charging batteries of the plurality of vehicles according to the distributed charging power, and thus the time for simultaneously charging the charging batteries of the plurality of vehicles to full capacity can be maximally shortened.

Description

Technical Field

[0001] This invention relates to the field of charging technology, and in particular to a high-power DC charger system and fast charging method capable of achieving fast charging. Background Technology

[0002] With societal progress and increased environmental awareness, electric vehicles (EVs) are gaining popularity due to their onboard power supply, which addresses the environmental pollution and high energy consumption issues associated with gasoline-powered vehicles. However, the charging issue for EVs is a major concern, as it directly impacts their widespread adoption and promotion. In recent years, high power, fast charging, and group charging capabilities have become fundamental requirements for high-power DC chargers. For industries like taxis and wedding car rentals that require the simultaneous use of multiple vehicles, high-power DC chargers are essential to charge the batteries of multiple vehicles at once, enabling simultaneous operation once fully charged.

[0003] Currently, the method for charging the batteries of multiple vehicles involves using multiple power harvesting modules to convert the DC power supplied by the power grid into the electrical energy required by the vehicles. The amount of electrical energy is directly proportional to the number of power harvesting modules. A power distribution unit then evenly distributes the electrical energy harvested by the multiple power harvesting modules to the various high-power DC chargers connected to the vehicles for charging. However, due to differences in remaining battery power, battery type, and temperature among the vehicles, the batteries of multiple vehicles cannot be fully charged simultaneously. This prolongs the time required to fully charge the batteries of multiple vehicles at the same time. Since multiple vehicles need to be fully charged simultaneously for operation, this affects the availability of multiple taxis, wedding cars, etc. for customers. Summary of the Invention

[0004] This invention provides a high-power DC charger system and fast charging method that enables fast charging, addressing the problem in the prior art where the batteries of multiple vehicles cannot be fully charged simultaneously, resulting in extended charging time for multiple vehicles and affecting customers' use of multiple taxis, wedding cars, etc.

[0005] In a first aspect, embodiments of the present invention provide a fast charging method applied to a high-power DC charger system capable of fast charging. The system includes a power distribution unit, multiple power acquisition modules, multiple high-power DC chargers, and a controller. The power distribution unit is electrically connected to the multiple power acquisition modules and the multiple high-power DC chargers, respectively. Each high-power DC charger includes a charging port. The controller is electrically connected to the power distribution unit and the multiple high-power DC chargers, respectively. The method provided by the present invention includes:

[0006] Step 1: At time N, the controller detects the type of the vehicle's battery connected to each high-power DC charger, the Nth temperature, and the Nth remaining charge through the charging port of each high-power DC charger, where N is a positive integer;

[0007] Step 2: The controller calculates the Nth variance of the remaining charge of the Nth battery in the vehicles corresponding to the multiple high-power DC chargers connected to the system.

[0008] Step 3: The controller inputs the type, temperature, and Nth variance of the remaining charge of the vehicles corresponding to each connected high-power DC charger into the pre-trained power allocation network model. This allows the power allocation network model to obtain the Nth power allocation strategy based on the pre-configured network parameters. The power allocation network model is trained by inputting the historical battery type, temperature at multiple historical moments, variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the corresponding power allocation strategies at multiple historical moments into the network to be trained. The power allocation strategy is used to indicate how to reduce the variance of the remaining charge of the batteries of multiple vehicles after power allocation.

[0009] Step 4: According to the Nth power allocation strategy, the controller controls the power allocation unit to allocate charging power to the multiple connected high-power DC chargers, so that the multiple connected high-power DC chargers can charge the charging batteries of multiple vehicles according to the allocated charging power.

[0010] Step 5: At time N+1, the controller detects the type of the vehicle's charging battery, the temperature at time N+1, and the remaining charge at time N+1 through the charging port of each connected high-power DC charger. The interval between time N+1 and time N is less than the set time.

[0011] Step 6: The controller calculates the variance of the (N+1)th remaining charge of the vehicle's battery corresponding to multiple connected high-power DC chargers.

[0012] Step 7: The difference between the Nth variance and the (N+1)th variance is determined as the (N+1)th variance reduction rate. If the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than the set rate threshold, the controller updates the network parameters configured in the power allocation network model using the reinforcement learning model based on the (N+1)th variance. The Nth variance reduction rate is determined based on the difference between the (N-1)th variance and the Nth variance.

[0013] Step 8: Increment N by 1, return to step 1, until the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is less than the set rate threshold.

[0014] Step 9: Repeat steps 1-4 until the remaining charge of each vehicle's battery is 100%.

[0015] In one possible implementation, before the controller detects the type, Nth temperature, and Nth remaining charge of the vehicle's charging battery corresponding to each high-power DC charger at time N via the charging port of each high-power DC charger, the method further includes:

[0016] The controller inputs the historical battery types of the vehicles corresponding to each connected high-power DC charger, the temperature at multiple historical moments, the variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategies corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain the power allocation network model.

[0017] In one possible implementation, before the controller detects the type, Nth temperature, and Nth remaining charge of the vehicle's charging battery corresponding to each high-power DC charger through the charging port of each high-power DC charger at time N, the method provided by the present invention further includes:

[0018] The controller inputs the historical battery types of the vehicles corresponding to each connected high-power DC charger, the temperature at multiple historical moments, the variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategies corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain the power allocation network model.

[0019] In one possible implementation, the power distribution unit includes multiple power regulation units connected in parallel. Each power regulation unit includes a number of switching modules connected in parallel with the power acquisition modules. Each switching module is electrically connected to a corresponding charging port and is connected in series with one of the power acquisition modules. The controller controls the power distribution unit to distribute charging power to the multiple high-power DC chargers according to the Nth power distribution strategy, including:

[0020] The controller determines the number of switching modules that need to be closed in each power regulation unit according to the Nth power allocation strategy;

[0021] The controller controls the closing of the switching modules of each power regulation unit according to the number of switching modules that need to be closed in each power regulation unit, so as to control the power distribution unit to distribute charging power to multiple high-power DC chargers respectively. The power acquisition modules connected to each closed switching module are different.

[0022] In one possible implementation, after the controller detects the type, Nth temperature, and Nth remaining charge of the vehicle's battery connected to each high-power DC charger via the charging port of each high-power DC charger at time N, the method further includes:

[0023] When the controller detects that the Nth battery charge of any connected high-power DC charger is 100%, it controls the high-power DC charger to disconnect from the corresponding vehicle.

[0024] In one possible implementation, the network to be trained is a convolutional neural network.

[0025] Secondly, the present invention provides a high-power DC charger system capable of fast charging. The system includes a power distribution unit, multiple power acquisition modules, multiple high-power DC chargers, and a controller. The power distribution unit is electrically connected to the multiple power acquisition modules and the multiple high-power DC chargers. Each high-power DC charger includes a charging port. The controller is electrically connected to the power distribution unit and the multiple high-power DC chargers.

[0026] The controller is used to detect the type of the charging battery of the vehicle connected to each high-power DC charger, the Nth temperature, and the Nth remaining charge at time N through the charging port of each high-power DC charger, where N is a positive integer and the initial value of N is 1;

[0027] The Nth variance of the remaining Nth charge of the vehicle's battery corresponding to multiple high-power DC chargers connected to the system is calculated.

[0028] The battery type, temperature, and Nth variance of the Nth remaining charge of each vehicle connected to the high-power DC charger are input into a pre-trained power allocation network model. This allows the power allocation network model to obtain the Nth power allocation strategy based on pre-configured network parameters. The power allocation network model is trained by inputting the historical battery type, temperature at multiple historical moments, variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the corresponding power allocation strategies at multiple historical moments into the network to be trained. The power allocation strategy is used to indicate how to reduce the variance of the remaining charge of the batteries of multiple vehicles after power allocation.

[0029] According to the Nth power allocation strategy, the control power allocation unit allocates charging power to multiple connected high-power DC chargers respectively, so that the multiple connected high-power DC chargers charge the charging batteries of multiple vehicles according to the allocated charging power.

[0030] At time N+1, the charging port of each connected high-power DC charger is used to detect the type of the vehicle's charging battery, the temperature at time N+1, and the remaining charge at time N+1. The interval between time N+1 and time N is less than the set time.

[0031] Calculate the variance of the (N+1)th remaining charge of the battery in the vehicle corresponding to multiple connected high-power DC chargers.

[0032] The difference between the Nth variance and the (N+1)th variance is determined as the (N+1)th variance reduction rate. If the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than a set rate threshold, then the network parameters configured in the power allocation network model are updated using a reinforcement learning model based on the (N+1)th variance. The Nth variance reduction rate is determined based on the difference between the (N-1)th variance and the Nth variance.

[0033] For N+1, return to the step of detecting the type of the vehicle's charging battery, the Nth temperature, and the Nth remaining charge at time N through the charging port of each connected high-power DC charger, until the difference between the variance reduction rate of N+1 and the variance reduction rate of N is less than the set rate threshold. Then, according to the configured power allocation strategy, control the power allocation unit to allocate charging power to the multiple connected high-power DC chargers, so that the multiple connected high-power DC chargers charge the charging batteries of multiple vehicles according to the allocated charging power, until the remaining charge of each vehicle's charging battery is 100%.

[0034] In one possible implementation, the controller is further configured to input the historical charging battery type of each vehicle corresponding to the historically connected high-power DC charger, the temperature at multiple historical moments, the variance of the remaining charge of the charging batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategy corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain a power allocation network model.

[0035] In one possible implementation, the power distribution unit includes multiple power regulation units connected in parallel. Each power regulation unit includes a number of switching modules connected in parallel with the power acquisition modules. Each switching module is electrically connected to a corresponding charging port, and each switching module is connected in series with one of the power acquisition modules.

[0036] The controller is also used to determine the number of switch modules that need to be closed in each power regulation unit according to the Nth power distribution strategy; and to control the closing of the switch modules of each power regulation unit according to the determined number of switch modules that need to be closed in each power regulation unit, so as to control the power distribution unit to distribute charging power to multiple high-power DC chargers respectively, wherein the power acquisition modules connected to each closed switch module are different.

[0037] In one possible implementation, the controller is further configured to disconnect the high-power DC charger from the corresponding vehicle when it is detected that the Nth remaining charge of the charging battery of any connected high-power DC charger is 100%.

[0038] In one possible implementation, the network to be trained is a convolutional neural network.

[0039] Compared to existing technologies, this invention offers the following advantages: The high-power DC charger system and fast-charging method provided by this invention utilizes a power allocation network model trained on historical data. This model is based on the types of batteries in vehicles connected to each high-power DC charger, temperatures at multiple historical moments, the variance of the remaining charge of the batteries after power allocation at multiple historical moments, and the corresponding power allocation strategies at multiple historical moments. By inputting these data as training samples into the network, the reliability of the output power allocation strategy is high. Consequently, after multiple high-power DC chargers charge the batteries of multiple vehicles according to their allocated power, the variance of the remaining charge of the batteries decreases, resulting in a more balanced distribution of remaining charge across the batteries. This allows for the simultaneous full charging of the batteries in multiple vehicles in the shortest possible time.

[0040] Additionally, the controller can calculate the (N+1)th variance of the remaining charge of the batteries of vehicles corresponding to multiple connected high-power DC chargers. The difference between the Nth variance and the (N+1)th variance is determined as the Nth variance reduction rate. Understandably, a larger variance reduction rate leads to a faster and more balanced distribution of remaining charge across the batteries of multiple vehicles. If the difference between the Nth variance reduction rate and the (N-1)th variance reduction rate exceeds a set rate threshold, it indicates that the variance reduction rate has room for improvement. Based on the Nth variance reduction rate, a reinforcement learning model is used to update the network parameters configured for the power allocation network model. Thus, the Nth variance reduction rate has been improved, and further improvement is limited. Therefore, updating the network parameters configured for the power allocation network model is stopped, i.e., the power allocation strategy is stopped. The controller continues to control the power distribution unit to allocate charging power to multiple connected high-power DC chargers according to the configured power distribution strategy. This allows the multiple connected high-power DC chargers to charge the batteries of multiple vehicles according to the allocated charging power until the remaining charge of each vehicle's battery is 100%. In this way, the time to charge the batteries of multiple vehicles at the same time can be minimized, improving the efficiency of charging the batteries of multiple vehicles at the same time, without affecting the customer's use of multiple taxis, wedding cars, etc. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 One of the circuit connection block diagrams of a high-power DC charger system capable of fast charging provided in an embodiment of this application;

[0043] Figure 2 A flowchart of a fast charging method provided in an embodiment of this application;

[0044] Figure 3 The second circuit connection block diagram of a high-power DC charger system capable of fast charging provided in the embodiments of this application.

[0045] The correspondence between the reference numerals and component names in the attached drawings is as follows: power acquisition module 101, power grid 102, power distribution unit 103, power adjustment unit 104, switch module 105, high-power DC charger 106, controller 107. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments made by those skilled in the art under the guidance of these embodiments are within the scope of protection of the present invention.

[0047] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0048] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0049] This invention provides a fast charging method applicable to a high-power DC charger system capable of fast charging. For example... Figure 1 As shown, the system includes a power distribution unit 103, multiple power acquisition modules 101, multiple high-power DC chargers 106, and a controller 107. The power distribution unit 103 is electrically connected to the multiple power acquisition modules 101 and the multiple high-power DC chargers 106, respectively. Each high-power DC charger 106 includes a charging port. The controller 107 is electrically connected to the power distribution unit 103 and the multiple high-power DC chargers 106, respectively.

[0050] In this embodiment, the power acquisition module 101 includes, but is not limited to, 20 modules. The power acquisition module 101 is used to convert the DC power supplied by the power grid 102 into electrical energy required for the vehicle's rechargeable battery. In this embodiment, the rated output power of a single power acquisition module 101 is set to 20 kW, so the total output power of the 20 power acquisition modules 101 is 400 kW.

[0051] like Figure 2As shown, the method provided in this embodiment of the invention includes:

[0052] S201: At time N, the controller 107 detects the type of the vehicle's charging battery connected to each high-power DC charger 106, the Nth temperature, and the Nth remaining charge through the charging port of each high-power DC charger 106, where N is a positive integer.

[0053] The types of rechargeable batteries used in each vehicle are not limited to lead-acid batteries, sodium-sulfur batteries, and secondary lithium batteries, and are not specified here. Furthermore, the temperature and remaining charge of the rechargeable batteries in each vehicle will vary at different times.

[0054] S202: The controller 107 calculates the Nth variance of the Nth remaining charge of the vehicle's charging battery corresponding to the multiple high-power DC chargers 106 connected to the system.

[0055] Among them, variance can express the degree of consistency (i.e., dispersion) among the Nth remaining charge of the charging battery of the vehicle corresponding to multiple high-power DC chargers 106.

[0056] S203: The controller 107 inputs the type, temperature, and Nth variance of the charging battery of each connected high-power DC charger 106 into the pre-trained power allocation network model, so that the power allocation network model can obtain the Nth power allocation strategy according to the pre-configured network parameters.

[0057] The power allocation network model is trained by inputting the historical charging battery type of each vehicle corresponding to the high-power DC charger 106 that has been connected in history, the temperature at multiple historical moments, the variance of the remaining charge of the charging batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategy corresponding to multiple historical moments as training samples into the network to be trained. The power allocation strategy is used to indicate how to reduce the variance of the remaining charge of the charging batteries of multiple vehicles after power allocation.

[0058] The network to be trained can be, but is not limited to, a convolutional neural network.

[0059] That is to say, before S201 above, the method provided by the embodiments of the present invention further includes: the controller 107 inputs the historical charging battery type of each vehicle corresponding to the historically connected high-power DC charger 106, the temperature at multiple historical times, the variance of the remaining charge of the charging batteries of multiple vehicles at multiple historical times after power allocation, and the power allocation strategy corresponding to multiple historical times as training samples into the network to be trained, so as to obtain a power allocation network model.

[0060] S204: According to the Nth power allocation strategy, the controller 107 controls the power allocation unit 103 to allocate charging power to the multiple connected high-power DC chargers 106 respectively, so that the multiple connected high-power DC chargers 106 can charge the charging batteries of multiple vehicles according to the allocated charging power.

[0061] like Figure 3 As shown, the power distribution unit 103 includes multiple power adjustment units 104 connected in parallel. Each power adjustment unit 104 includes the same number of switching modules as the power acquisition modules 101, connected in parallel with each other. Each switching module is electrically connected to a corresponding charging port, and each switching module is connected in series with one of the power acquisition modules 101. Thus, S204 can be specifically implemented as follows:

[0062] According to the Nth power allocation strategy, the controller 107 determines the number of switch modules that need to be closed in each power adjustment unit 104. Based on the determined number of switch modules that need to be closed in each power adjustment unit 104, the controller 107 controls the switch modules of each power adjustment unit 104 to close, so as to control the power allocation unit 103 to allocate charging power to multiple high-power DC chargers 106 respectively, wherein the power acquisition modules 101 connected to each closed switch module are different.

[0063] For example, when the rated output power of a single power acquisition module 101 is 20KW and a single power regulation unit 104 includes 20 switching modules, for the power distribution unit 103, the power output from the charging port of each high-power DC charger 106 comes from one or more power acquisition modules 101. When one of the switching modules 105 of the power regulation unit 104 is closed, the maximum output power from the charging port of the high-power DC charger 106 is 20KW. When two of the switching modules 105 of the same power regulation unit 104 are closed, the maximum output power from the charging port of the high-power DC charger 106 is 40KW. When all the switch modules 105 of the same power regulation unit 104 are closed, the maximum power output of the charging port of the high-power DC charger 106 is 400KW. When one switch module 105 of each of the two power regulation units 104 is closed, the maximum power output of the charging port of the high-power DC charger 106 connected to each power regulation unit 104 is 20KW. When one switch module 105 of each of the three power regulation units 104 is closed, the maximum power output of the charging port of the high-power DC charger 106 connected to each power regulation unit 104 is 20KW, but the total output power does not exceed 400KW.

[0064] S205: At time N+1, the controller 107 detects the type of the vehicle's charging battery, the N+1th temperature, and the N+1th remaining charge through the charging port of each connected high-power DC charger 106.

[0065] The interval between the (N+1)th time and the Nth time is less than the set duration (e.g., 1ms, 2ms, etc.).

[0066] S206: Controller 107 calculates the N+1 variance of the remaining charge of the charging battery of the vehicle corresponding to multiple connected high-power DC chargers 106.

[0067] S207: The difference between the Nth variance and the (N+1)th variance is determined as the (N+1)th variance reduction rate. If the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than the set rate threshold, the controller 107 updates the network parameters configured in the power allocation network model using a reinforcement learning model based on the (N+1)th variance.

[0068] The rate of decrease of the Nth variance is determined based on the difference between the (N-1)th variance and the Nth variance.

[0069] S208: Increment N by 1, then return to execute S201 until the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is less than the set rate threshold.

[0070] S209: Repeat S201-S204 until the remaining charge of each vehicle's battery is 100%.

[0071] In summary, the high-power DC charger system and fast charging method provided by this invention enable rapid charging. The power allocation network model is trained by inputting the historical battery types, temperatures at multiple historical moments, variances of the remaining charge levels of the batteries at multiple historical moments after power allocation, and corresponding power allocation strategies into the network. This ensures high reliability of the output Nth power allocation strategy by inputting the battery types, temperatures, and Nth variances of the remaining charge levels of the vehicles corresponding to each high-power DC charger 106 into the power allocation network model. Consequently, after multiple high-power DC chargers 106 charge the batteries of multiple vehicles according to their allocated charging power, the variances of the remaining charge levels of the batteries decrease, resulting in a more balanced distribution of remaining charge levels and enabling simultaneous full charging of multiple vehicles' batteries in the shortest possible time.

[0072] Furthermore, the controller 107 can calculate the (N+1)th variance of the remaining charge of the charging batteries of the vehicles corresponding to the multiple connected high-power DC chargers 106. The difference between the Nth variance and the (N+1)th variance is determined as the Nth variance reduction rate. Understandably, a larger variance reduction rate leads to a faster and more balanced distribution of the remaining charge of the charging batteries across multiple vehicles. If the difference between the Nth variance reduction rate and the (N-1)th variance reduction rate is greater than a set rate threshold, it indicates that the variance reduction rate has room for improvement. Therefore, based on the Nth variance reduction rate, a reinforcement learning model is used to update the network parameters configured for the power allocation network model. Thus, the Nth variance reduction rate has been improved, and there is little room for further improvement. Consequently, updating the network parameters configured for the power allocation network model is stopped, i.e., the power allocation strategy is stopped. Furthermore, the controller 107 continues to control the power distribution unit 103 to allocate charging power to the multiple connected high-power DC chargers 106 according to the configured power distribution strategy, so that the multiple connected high-power DC chargers 106 charge the batteries of multiple vehicles according to the allocated charging power until the remaining charge of each vehicle's battery is 100%. In this way, the time to charge the batteries of multiple vehicles at the same time can be minimized, the efficiency of charging the batteries of multiple vehicles at the same time can be improved, and it will not affect the customer's use of multiple taxis, wedding cars, etc.

[0073] Furthermore, when the controller 107 detects that the Nth remaining charge of the vehicle's rechargeable battery corresponding to any of the connected high-power DC chargers 106 is 100%, it controls the high-power DC charger 106 to disconnect from the corresponding vehicle. This avoids charging the vehicle's rechargeable battery even when the Nth remaining charge is 100%, thus extending the battery's lifespan and saving energy.

[0074] like Figure 2 As shown, this invention provides a high-power DC charger system capable of fast charging. The system includes a power distribution unit 103, multiple power acquisition modules 101, multiple high-power DC chargers 106, and a controller 107. The power distribution unit 103 is electrically connected to the multiple power acquisition modules 101 and the multiple high-power DC chargers 106. Each high-power DC charger 106 includes a charging port. The controller 107 is electrically connected to the power distribution unit 103 and the multiple high-power DC chargers 106. It should be noted that the high-power DC charger system capable of fast charging provided in this embodiment has the same basic principle and technical effects as the above embodiments. For the sake of brevity, any parts not mentioned in the embodiments of this application can be referred to the corresponding content in the above embodiments.

[0075] The controller 107 is used to detect the type of the charging battery of the vehicle connected to each high-power DC charger 106, the Nth temperature, and the Nth remaining charge through the charging port of each high-power DC charger 106 at time N, where N is a positive integer and the initial value of N is 1.The Nth variance of the remaining charge capacity of the vehicles corresponding to the multiple high-power DC chargers 106 connected to the network is statistically analyzed. The type, temperature, and Nth variance of the remaining charge capacity of the vehicles corresponding to each high-power DC charger 106 are then input into a pre-trained power allocation network model. This allows the power allocation network model to obtain the Nth power allocation strategy based on pre-configured network parameters. The power allocation network model is based on the historical battery type, temperature at multiple historical moments, and the variance of the remaining charge capacity of the vehicles' batteries at multiple historical moments after power allocation. The network is trained by inputting power allocation strategies corresponding to multiple historical moments as training samples; wherein, the power allocation strategy is used to indicate the variance of the remaining charge of the charging batteries of multiple vehicles after reducing power allocation; according to the Nth power allocation strategy, the power allocation unit 103 allocates charging power to multiple connected high-power DC chargers 106 respectively, so that the multiple connected high-power DC chargers 106 charge the charging batteries of multiple vehicles according to the allocated charging power; at the N+1th moment, the type of charging battery of the vehicle corresponding to each connected high-power DC charger 106 is detected through the charging port of each connected high-power DC charger 106. The system calculates the (N+1)th temperature and the (N+1)th remaining battery power, where the interval between the (N+1)th time and the Nth time is less than a set duration; it calculates the (N+1)th variance of the (N+1)th remaining battery power of the vehicles corresponding to multiple accessed high-power DC chargers 106; it determines the (N+1)th variance reduction rate by the difference between the Nth variance and the (N+1)th variance reduction rate; if the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than a set rate threshold, it updates the network parameters of the power allocation network model configuration using a reinforcement learning model based on the (N+1)th variance, where the Nth variance reduction rate is determined based on the difference between the (N-1)th variance and the Nth variance; it increments N by 1 and returns the result. The process involves detecting the type, Nth temperature, and Nth remaining charge of the vehicle's charging battery at time N by passing through the charging port of each connected high-power DC charger 106. This process continues until the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is less than a set rate threshold. Based on the configured power allocation strategy, the power allocation unit 103 allocates charging power to the multiple connected high-power DC chargers 106, so that the multiple connected high-power DC chargers 106 charge the charging batteries of the multiple vehicles according to the allocated charging power, until the remaining charge of each vehicle's charging battery is 100%.

[0076] In one possible implementation, the controller 107 is further configured to input the historical charging battery type of each vehicle corresponding to the historically connected high-power DC charger 106, the temperature at multiple historical moments, the variance of the remaining charge of the charging batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategy corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain a power allocation network model.

[0077] In one possible implementation, such as Figure 3 As shown, the power distribution unit 103 includes multiple power adjustment units 104 connected in parallel. Each power adjustment unit 104 includes the same number of switching modules as the power acquisition modules 101, which are connected in parallel with each other. Each switching module is electrically connected to a corresponding charging port, and each switching module is connected in series with one of the power acquisition modules 101.

[0078] The controller 107 is also configured to determine the number of switch modules that need to be closed in each power adjustment unit 104 according to the Nth power allocation strategy; and control the switch modules of each power adjustment unit 104 to close according to the determined number of switch modules that need to be closed in each power adjustment unit 104, so as to control the power allocation unit 103 to allocate charging power to multiple high-power DC chargers 106 respectively, wherein the power acquisition modules 101 connected to each closed switch module are different.

[0079] In one possible implementation, the controller 107 is further configured to disconnect the high-power DC charger 106 from the corresponding vehicle when it detects that the Nth remaining charge of the charging battery of any connected high-power DC charger 106 is 100%.

[0080] In one possible implementation, the network to be trained is a convolutional neural network.

[0081] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fast charging method, characterized in that, A high-power DC charger system capable of fast charging is provided. The system includes a power distribution unit, multiple power acquisition modules, multiple high-power DC chargers, and a controller. The power distribution unit is electrically connected to the multiple power acquisition modules and the multiple high-power DC chargers. Each high-power DC charger includes a charging port. The controller is electrically connected to the power distribution unit and the multiple high-power DC chargers. The method includes: Step 1: At time N, the controller detects the type of the charging battery of the vehicle connected to each high-power DC charger, the Nth temperature, and the Nth remaining charge through the charging port of each high-power DC charger, where N is a positive integer; Step 2: The controller calculates the Nth variance of the remaining Nth charge of the vehicle's charging battery corresponding to the multiple high-power DC chargers connected to the system. Step 3: The controller inputs the type, temperature, and Nth variance of the remaining charge of the vehicles corresponding to each connected high-power DC charger into the pre-trained power allocation network model, so that the power allocation network model can obtain the Nth power allocation strategy according to the pre-configured network parameters. The power allocation network model is trained by inputting the historical battery type, temperature at multiple historical times, variance of the remaining charge of the batteries of multiple vehicles at multiple historical times after power allocation, and the power allocation strategy corresponding to multiple historical times as training samples into the network to be trained. The power allocation strategy is used to indicate the reduction of the variance of the remaining charge of the batteries of multiple vehicles after power allocation. Step 4: The controller controls the power allocation unit to allocate charging power to the multiple connected high-power DC chargers according to the Nth power allocation strategy, so that the multiple connected high-power DC chargers can charge the charging batteries of multiple vehicles according to the allocated charging power. Step 5: At time N+1, the controller detects the type of the vehicle's charging battery, the temperature at time N+1, and the remaining charge at time N+1 through the charging port of each connected high-power DC charger. The interval between time N+1 and time N is less than the set time. Step 6: The controller calculates the (N+1)th variance of the remaining charge of the charging battery of the vehicle corresponding to the multiple connected high-power DC chargers; Step 7: The difference between the Nth variance and the (N+1)th variance is determined as the (N+1)th variance reduction rate. If the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than a set rate threshold, the controller updates the network parameters configured in the power allocation network model using a reinforcement learning model based on the (N+1)th variance. The Nth variance reduction rate is determined based on the difference between the (N-1)th variance and the Nth variance. Step 8: Increment N by 1, return to step 1, until the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is less than the set rate threshold. Step 9: Repeat steps 1-4 until the remaining charge of each vehicle's battery is 100%.

2. The method according to claim 1, characterized in that, Before the controller detects the type, Nth temperature, and Nth remaining charge of the vehicle's charging battery corresponding to each high-power DC charger at time N via the charging port of each high-power DC charger, the method further includes: The controller inputs the historical battery types of the vehicles corresponding to each connected high-power DC charger, the temperatures at multiple historical moments, the variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategies corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain the power allocation network model.

3. The method according to claim 1, characterized in that, The power distribution unit includes multiple power regulation units connected in parallel. Each power regulation unit includes a number of switching modules connected in parallel with the power acquisition modules. Each switching module is electrically connected to a corresponding charging port and is connected in series with one of the power acquisition modules. The controller controls the power distribution unit to distribute charging power to the multiple high-power DC chargers according to the Nth power distribution strategy, including: The controller determines the number of switch modules that need to be closed in each power regulation unit according to the Nth power allocation strategy. The controller controls the closing of the switch modules of each power regulation unit according to the determined number of switch modules that need to be closed in each power regulation unit, so as to control the power distribution unit to distribute charging power to the multiple high-power DC chargers respectively, wherein the power acquisition modules connected to each closed switch module are different.

4. The method according to claim 1, characterized in that, After the controller detects the type, Nth temperature, and Nth remaining charge of the vehicle's battery connected to each high-power DC charger at time N via the charging port of each high-power DC charger, the method further includes: When the controller detects that the Nth battery charge of any connected high-power DC charger is 100%, it controls the high-power DC charger to disconnect from the corresponding vehicle.

5. The method according to claim 1, characterized in that, The network to be trained is a convolutional neural network.

6. A high-power DC charger system capable of fast charging, characterized in that, The system includes a power distribution unit, multiple power acquisition modules, multiple high-power DC chargers, and a controller. The power distribution unit is electrically connected to the multiple power acquisition modules and the multiple high-power DC chargers. Each high-power DC charger includes a charging port. The controller is electrically connected to the power distribution unit and the multiple high-power DC chargers. The controller is used to detect the type of the charging battery of the vehicle connected to each high-power DC charger, the Nth temperature, and the Nth remaining charge at time N through the charging port of each high-power DC charger, where N is a positive integer and the initial value of N is 1. The Nth variance of the remaining Nth charge of the vehicle's battery corresponding to multiple high-power DC chargers connected to the system is calculated. The battery type, temperature, and Nth variance of the Nth remaining charge of the vehicles corresponding to each connected high-power DC charger are input into a pre-trained power allocation network model. This allows the power allocation network model to obtain the Nth power allocation strategy based on pre-configured network parameters. The power allocation network model is trained by inputting the historical battery type, temperature at multiple historical moments, variance of the remaining charge of the batteries of multiple vehicles at multiple historical moments after power allocation, and the corresponding power allocation strategies at multiple historical moments into the network to be trained as training samples. The power allocation strategy is used to indicate how to reduce the variance of the remaining charge of the batteries of multiple vehicles after power allocation. According to the Nth power allocation strategy, the power allocation unit is controlled to allocate charging power to multiple connected high-power DC chargers, so that the multiple connected high-power DC chargers can charge the charging batteries of multiple vehicles according to the allocated charging power. At time N+1, the charging port of each connected high-power DC charger is used to detect the type of the vehicle's charging battery, the temperature at time N+1, and the remaining charge at time N+1. The interval between time N+1 and time N is less than the set time. Calculate the variance of the (N+1)th remaining charge of the vehicle's charging battery corresponding to the multiple accessed high-power DC chargers. The difference between the Nth variance and the (N+1)th variance is determined as the (N+1)th variance reduction rate. If the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is greater than a set rate threshold, then the network parameters configured in the power allocation network model are updated using a reinforcement learning model based on the (N+1)th variance. The Nth variance reduction rate is determined based on the difference between the (N-1)th variance and the Nth variance. For N+1, return to the step of detecting the type, Nth temperature, and Nth remaining charge of the vehicle's charging battery corresponding to each connected high-power DC charger at time N through the charging port of each connected high-power DC charger, until the difference between the (N+1)th variance reduction rate and the Nth variance reduction rate is less than the set rate threshold, and according to the configured power allocation strategy, control the power allocation unit to allocate charging power to the multiple connected high-power DC chargers respectively, so that the multiple connected high-power DC chargers charge the charging batteries of multiple vehicles according to the allocated charging power, until the remaining charge of each vehicle's charging battery is 100%.

7. The system according to claim 6, characterized in that, The controller is further configured to input the historical charging battery type of each vehicle corresponding to the high-power DC charger accessed in history, the temperature at multiple historical moments, the variance of the remaining charge of the charging batteries of multiple vehicles at multiple historical moments after power allocation, and the power allocation strategy corresponding to multiple historical moments as training samples into the network to be trained, so as to obtain the power allocation network model.

8. The system according to claim 6, characterized in that, The power distribution unit includes multiple power regulation units connected in parallel. Each power regulation unit includes the same number of switching modules as the power acquisition modules, all connected in parallel. Each switching module is electrically connected to a corresponding charging port, and each switching module is connected in series with one of the power acquisition modules. The controller is further configured to determine the number of switch modules that need to be closed in each power adjustment unit according to the Nth power allocation strategy; and control the switch modules of each power adjustment unit to close according to the determined number of switch modules that need to be closed in each power adjustment unit, so as to control the power allocation unit to allocate charging power to the plurality of high-power DC chargers respectively, wherein the power acquisition modules connected to each closed switch module are different.

9. The system according to claim 6, characterized in that, The controller is also configured to disconnect the high-power DC charger from the corresponding vehicle when it is detected that the Nth remaining charge of the charging battery of any connected high-power DC charger is 100%.

10. The system according to claim 6, characterized in that, The network to be trained is a convolutional neural network.

Images