Range extender control method, device, controller, storage medium and program product

Abstract

The application relates to a range extender control method and device, a controller, a storage medium and a program product. The method comprises the following steps: acquiring real-time driving behavior state data of a target vehicle, acquiring real-time attention degree data of a driver of the target vehicle to different vehicle performance dimensions according to the real-time driving behavior state data; acquiring real-time attention degree weights of the driver to the vehicle performance dimensions according to the real-time attention degree data, and obtaining a real-time attention degree weight combination of the driver; starting a range extender of the target vehicle in the case that the target vehicle meets a preset range extender starting condition, and acquiring target power generation power and target engine rotating speed corresponding to the real-time attention degree weight combination according to a first mapping relationship constructed in advance; and controlling the range extender by using the target power generation power and the target engine rotating speed. The range extender control method can cope with instantaneous attention changes and meet real-time requirements of users.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a range extender control method, device, controller, storage medium, and program product. Background Technology

[0002] With the development of vehicle control technology, a technology for controlling electric vehicle range extenders has emerged. Range extenders typically directly determine a vehicle's performance in three core indicators: NVH performance, fuel economy, and battery retention.

[0003] In traditional technologies, the control of vehicle range extenders is usually based on fixed rules or driving style recognition technology based on machine learning. For example, power matching can be performed based on the SOC (state of charge) threshold and vehicle speed, or the user's driving habits can be learned through historical data to adjust the control parameters according to the driving habits.

[0004] However, the aforementioned range extender control methods, which rely on fixed rules, are unable to adapt to the personalized needs of different drivers. On the other hand, the method of learning user driving habits from historical data can only reflect long-term statistical preferences and cannot cope with instantaneous changes in attention. This results in a significant disconnect from the user's real-time subjective perception. Therefore, the existing range extender control methods cannot meet the user's real-time needs. Summary of the Invention

[0005] Based on this, this application addresses the aforementioned technical problems by providing a range extender control method, apparatus, controller, computer-readable storage medium, and computer program product that can meet the real-time needs of users.

[0006] Firstly, this application provides a range extender control method, including:

[0007] Acquire real-time driving behavior status data for the target vehicle, and obtain real-time attention data of the driver of the target vehicle to different vehicle performance dimensions based on the real-time driving behavior status data.

[0008] Based on the real-time attention data, the real-time attention weights of the driver for each of the vehicle performance dimensions are obtained, and the real-time attention weight combination of the driver is obtained.

[0009] When the target vehicle meets the preset range extender start-up conditions, the range extender of the target vehicle is started, and the target power generation and target engine speed corresponding to the real-time attention weight combination are obtained according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations.

[0010] The range extender is controlled using the target power generation and the target engine speed.

[0011] In one embodiment, obtaining the driver's real-time attention weights for each of the vehicle performance dimensions based on the real-time attention data includes: obtaining the driver's first attention weights for each of the vehicle performance dimensions based on the real-time attention data when any of the real-time attention data is greater than or equal to a first attention threshold and any of the real-time attention data is less than a second attention threshold; obtaining a pre-set uniform attention weight for each of the vehicle performance dimensions; weighting each of the first attention weights and each of the uniform attention weights using a pre-set residual mixing coefficient to obtain the driver's second attention weights for each of the vehicle performance dimensions; and obtaining each of the real-time attention weights using the second attention weights.

[0012] In one embodiment, obtaining the driver's first attention weight for each of the vehicle performance dimensions based on the real-time attention data includes: mapping each of the real-time attention data using a preset temperature coefficient to obtain real-time attention mapping data corresponding to each of the vehicle performance dimensions; and normalizing each of the real-time attention mapping data to obtain each of the first attention weights.

[0013] In one embodiment, the real-time attention weight is the driver's attention weight for different vehicle performance dimensions in the current sampling period; obtaining each real-time attention weight using each of the second attention weights includes: acquiring a preset number of historical attention weights; wherein each of the historical attention weights is the driver's attention weight for different vehicle performance dimensions in each historical sampling period, and each of the historical sampling periods is a preset number of sampling periods prior to the current sampling period; and smoothing each of the second attention weights using each of the historical attention weights to obtain the real-time attention weight.

[0014] In one embodiment, obtaining the driver's real-time attention weight corresponding to each of the vehicle performance dimensions based on the real-time attention data further includes: when all the real-time attention data are less than the first attention threshold, obtaining a first default attention weight pre-set for each of the vehicle performance dimensions, and using each of the first default attention weights as the real-time attention weights; when all the real-time attention data are greater than the second attention threshold, obtaining a second default attention weight pre-set for each of the vehicle performance dimensions, and using each of the second default attention weights as the real-time attention weights; wherein the second default attention weights pre-set for each of the vehicle performance dimensions are evenly distributed.

[0015] In one embodiment, obtaining real-time attention data of the driver of the target vehicle for different vehicle performance dimensions based on the real-time driving behavior state data includes: obtaining real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions based on the real-time driving behavior state data; fusing the real-time environmental impact attention data and real-time event impact attention data of each vehicle performance dimension using a pre-set fusion coefficient to obtain initial real-time attention data of the driver for different vehicle performance dimensions; obtaining a pre-set attention data threshold for each vehicle performance dimension, and using the smaller value between each attention data threshold and each initial real-time attention data as the real-time attention data of the driver for each vehicle performance dimension.

[0016] In one embodiment, the vehicle performance dimension includes an NVH performance dimension. The step of obtaining real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions based on the real-time driving behavior state data includes: obtaining first environmental impact data associated with the NVH performance dimension based on the real-time driving behavior state data; the first environmental impact data includes at least one of the following: vehicle speed, window opening, external microphone decibel value, and music volume; using the first environmental impact data, obtaining real-time environmental impact attention data corresponding to the NVH performance dimension; obtaining each first short-term behavioral event associated with the NVH performance dimension triggered within a preset time window based on the real-time driving behavior state data; the first short-term behavioral events include at least one of the following: sudden increase in music volume, sudden change in accelerator pedal position, driver looking at the central control screen, air conditioning adjustment and phone call connection, and changes in facial expression and grip strength; using the trigger time difference and event type weight associated with each of the first short-term behavioral events, obtaining real-time event impact attention data corresponding to the NVH performance dimension.

[0017] In one embodiment, the vehicle performance dimension includes a fuel consumption performance dimension. The step of obtaining real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions based on the real-time driving behavior state data includes: obtaining second environmental impact data associated with the fuel consumption performance dimension based on the real-time driving behavior state data; the second environmental impact data includes: the number of times the driver views the fuel consumption details page of the target vehicle within a preset historical time period, the actual fuel consumption data of the target vehicle, and the cumulative mileage data driven by fuel; using the second environmental impact data, obtaining real-time environmental impact attention data corresponding to the fuel consumption performance dimension; obtaining each second short-term behavioral event associated with the fuel consumption performance dimension triggered within a preset time window based on the real-time driving behavior state data; the second short-term behavioral events include: driver viewing fuel consumption events, rapid acceleration events, and emergency braking events; using the trigger time difference and event type weight associated with each second short-term behavioral event, obtaining real-time event impact attention data corresponding to the fuel consumption performance dimension.

[0018] In one embodiment, the vehicle performance dimension includes a battery maintenance performance dimension. The step of obtaining real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions based on the real-time driving behavior state data includes: obtaining third environmental impact data associated with the battery maintenance performance dimension based on the real-time driving behavior state data; the third environmental impact data includes: the number of times the driver views the target vehicle's state of charge within a preset historical time period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power; using the third environmental impact data, obtaining real-time environmental impact attention data corresponding to the battery maintenance performance dimension; obtaining each third short-term behavioral event associated with the battery maintenance performance dimension triggered within a preset time window based on the real-time driving behavior state data; the third short-term behavioral events include: driver viewing state of charge events and continuous high-speed driving events; using the trigger time difference and event type weight associated with each of the third short-term behavioral events, obtaining real-time event impact attention data corresponding to the battery maintenance performance dimension.

[0019] In one embodiment, activating the range extender of the target vehicle when the target vehicle meets preset range extender activation conditions includes: acquiring NVH performance attention data of the driver during the range extender activation process based on the real-time driving behavior state data; activating the range extender when the real-time state of charge of the target vehicle is less than a first target state of charge and the NVH performance attention data is less than a preset activation attention threshold; the method further includes: deactivating the range extender when the real-time state of charge of the target vehicle is greater than or equal to a second target state of charge, or when the NVH performance attention data is greater than or equal to a shutdown attention threshold; wherein the shutdown attention threshold is greater than the activation attention threshold, and the second target state of charge is greater than the first target state of charge.

[0020] In one embodiment, the first target state of charge is obtained through the following steps: obtaining the starting state of charge corresponding to the real-time driving mode of the target vehicle; setting the starting state of charge as the first target state of charge when the driver has not set a destination for the target vehicle; obtaining the remaining distance between the target vehicle and the destination when the driver has set a destination for the target vehicle; and obtaining the first target state of charge based on the distance relationship between the starting state of charge, a pre-set maximum state of charge, the remaining distance, and the nominal pure electric driving distance of the target vehicle.

[0021] In one embodiment, obtaining the driver's NVH performance attention data during the range extender startup process based on the real-time driving behavior state data includes: obtaining the target vehicle's speed, window opening, longitudinal acceleration, and driving time based on the real-time driving behavior state data; and using the vehicle speed, window opening, longitudinal acceleration, and driving time to obtain the NVH performance attention data.

[0022] In one embodiment, the vehicle performance dimensions include NVH performance, fuel consumption performance, and battery life performance. The first mapping relationship is constructed through the following steps: obtaining a first cost value for each power-speed combination corresponding to the NVH performance dimension, a second cost value corresponding to the fuel consumption performance dimension, and a third cost value corresponding to the battery life performance dimension; each power-speed combination includes generator power and engine speed; obtaining a current attention weight combination, and using the current attention weight combination to weight the first cost value, second cost value, and third cost value corresponding to each power-speed combination to obtain the cost value of the current attention weight combination corresponding to each power-speed combination; the current attention weight combination is any one of the different attention weight combinations; taking the minimum cost value among the cost values ​​as the target cost value corresponding to the current attention weight combination, and taking the power-speed combination corresponding to the target cost value as the target power-speed combination corresponding to the current attention weight combination; constructing the first mapping relationship using the correspondence between each current attention weight combination and the target power-speed combination.

[0023] In one embodiment, obtaining the first cost value corresponding to the NVH performance dimension, the second cost value corresponding to the fuel consumption performance dimension, and the third cost value corresponding to the battery life performance dimension for each power speed combination includes: for any power speed combination, obtaining the first cost value corresponding to the NVH performance dimension from a pre-built second mapping relationship; the second mapping relationship pre-stores the correspondence between different power speed combinations and the first cost value; obtaining the second cost value corresponding to the fuel consumption performance dimension from a pre-built third mapping relationship; the third mapping relationship pre-stores the correspondence between different power speed combinations and the second cost value; obtaining the third cost value corresponding to the battery life performance dimension from a pre-built fourth mapping relationship; the fourth mapping relationship pre-stores the correspondence between different power speed combinations and the third cost value.

[0024] In one embodiment, controlling the range extender using the target power generation and the target engine speed includes: controlling the range extender using the maximum power generation and the target engine speed when the target power generation exceeds the maximum power generation matching the vehicle speed of the target vehicle; and controlling the range extender using the target power generation and the target engine speed when the target power generation does not exceed the maximum power generation.

[0025] In one embodiment, the target power generation and target engine speed are the target power generation value and target engine speed value corresponding to the current sampling period; controlling the range extender using the target power generation and target engine speed includes: if the target power generation is greater than the historical power generation, controlling the range extender using the target power generation and target engine speed after a preset time delay; the historical power generation is the target power generation value corresponding to the previous sampling period; if the target power generation is not greater than the historical power generation, immediately controlling the range extender using the target power generation and target engine speed.

[0026] Secondly, this application also provides a range extender control device, comprising:

[0027] The attention data acquisition module is used to acquire real-time driving behavior status data for the target vehicle, and to acquire real-time attention data of the driver of the target vehicle on different vehicle performance dimensions based on the real-time driving behavior status data.

[0028] The attention weight acquisition module is used to obtain the driver's real-time attention weight for each of the vehicle performance dimensions based on the real-time attention data, and to obtain the driver's real-time attention weight combination.

[0029] The target acquisition module is used to start the range extender of the target vehicle when the target vehicle meets the preset range extender start conditions, and to obtain the target power generation and target engine speed corresponding to the real-time attention weight combination according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations.

[0030] The range extender control module is used to control the range extender using the target power generation and the target engine speed.

[0031] Thirdly, this application also provides a controller, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the methods described above.

[0032] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0033] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the above aspects.

[0034] The aforementioned range extender control method, device, controller, storage medium, and program product acquire real-time driving behavior state data for a target vehicle, and obtain real-time attention data of the driver for different vehicle performance dimensions based on the real-time driving behavior state data; based on each real-time attention data, obtain the driver's real-time attention weight for each vehicle performance dimension, and obtain a combination of driver's real-time attention weights; when the target vehicle meets the preset range extender start-up conditions, start the target vehicle's range extender, and obtain the target power generation and target engine speed corresponding to the real-time attention weight combination according to a pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations; and uses the target power generation and target engine speed to control the range extender. This application can determine the driver's real-time attention to different vehicle performance dimensions and the real-time attention weight of each vehicle performance dimension by collecting the real-time driving behavior status of the target vehicle. Then, when the range extender start-up conditions are met, the target power generation and target engine speed can be obtained according to the real-time attention weight and the first mapping relationship. The target power generation and target engine speed can then be used to control the range extender. This method can determine the importance of the driver's attention to different vehicle performance dimensions by combining the real-time driving behavior status, and thus complete the range extender control. Therefore, it can cope with instantaneous changes in attention, and the range extender control method can meet the user's real-time needs. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic diagram of an optional application environment for the range extender control method in one embodiment;

[0037] Figure 2 This is a schematic diagram of an optional flow chart of a range extender control method in one embodiment;

[0038] Figure 3 This is a schematic diagram of an optional process for obtaining real-time attention weights in one embodiment;

[0039] Figure 4 This is a schematic diagram of an optional process for obtaining real-time attention data in one embodiment;

[0040] Figure 5This is a schematic diagram of an optional process for the start-up and shutdown control of the range extender in one embodiment;

[0041] Figure 6 This is a schematic diagram of an optional process for constructing a first mapping relationship in one embodiment;

[0042] Figure 7 This is a schematic diagram of an optional architecture for a range-extending control system in one embodiment;

[0043] Figure 8 This is a schematic diagram of an optional structure of the range extender control device in one embodiment;

[0044] Figure 9 This is a schematic diagram of an optional internal structure of the controller in one embodiment. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0046] The terms "first," "second," etc., used in this application may be used to describe various elements, but these elements are not limited by these terms. These terms are used only to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0047] The range extender control method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, the controller 101 communicates with the range extender 102 via a CAN signal. Specifically, the controller 101 receives real-time driving behavior data of the driver regarding the target vehicle collected by the signal acquisition device. Based on this data, it identifies the driver's real-time attention to different vehicle performance dimensions and the driver's real-time attention weights for each dimension. When the target vehicle meets the preset range extender start-up conditions, the controller 101 starts the range extender 102 and obtains the corresponding target power generation and engine speed based on the real-time attention weights. The controller then uses the target power generation and engine speed to control the range extender 102. The controller 101 can be the vehicle controller for the target vehicle.

[0048] In one embodiment, such as Figure 2 As shown, a range extender control method is provided, which is applied to... Figure 1 Taking controller 101 as an example, the following steps are included:

[0049] Step S201: Obtain real-time driving behavior status data for the target vehicle, and obtain real-time attention data of the driver of the target vehicle to different vehicle performance dimensions based on the real-time driving behavior status data.

[0050] Real-time driving behavior data refers to the driver's real-time driving behavior data while driving the target vehicle, as well as the target vehicle's real-time status data. Vehicle performance dimensions refer to the different dimensions of vehicle performance that the driver focuses on, such as NVH performance, fuel consumption performance, and battery life performance, corresponding to the three core vehicle indicators: NVH performance, fuel economy, and battery life. Real-time attention data mainly describes the driver's real-time attention to different vehicle performance dimensions; the higher the real-time attention data for a certain vehicle performance dimension, the higher the driver's real-time attention to that vehicle performance dimension.

[0051] Specifically, the controller 101 can collect real-time driving behavior status data of the target vehicle, and then identify the real-time attention data of the driver of the target vehicle to different vehicle performance dimensions based on the real-time driving behavior status data.

[0052] Step S202: Based on the real-time attention data, obtain the real-time attention weights of drivers for each vehicle performance dimension, and obtain the combination of real-time attention weights of drivers.

[0053] The real-time attention weight represents the driver's real-time attention preference for each vehicle performance dimension, while the real-time attention weight combination is a weight combination composed of the real-time attention weights of different vehicle performance dimensions. In this embodiment, after obtaining the real-time attention data for each vehicle performance dimension, the driver's real-time attention weight for each vehicle performance dimension can also be calculated to form the real-time attention weight combination.

[0054] Step S203: If the target vehicle meets the preset range extender start conditions, start the range extender 102 of the target vehicle, and obtain the target power generation and target engine speed corresponding to the real-time attention weight combination according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations.

[0055] Step S204: The range extender 102 is controlled using the target power generation and the target engine speed.

[0056] The preset range extender start conditions refer to the pre-set conditions for the target vehicle to start the range extender 102. The first mapping relationship is used to store different attention weight combinations and their corresponding mapping relationship between power generation and engine speed. The target power generation and target engine speed refer to the target values ​​of power generation and engine speed, that is, the control target values ​​of power generation and engine speed.

[0057] Specifically, if the target vehicle meets the preset range extender start conditions, the controller 101 can start the range extender 102 and obtain the target power generation and target engine speed corresponding to the real-time attention weight combination from the first mapping relationship, thereby using the target power generation and target engine speed to complete the control of the range extender 102.

[0058] In the above-described range extender control method, the controller 101 acquires real-time driving behavior state data for the target vehicle, and obtains real-time attention data of the driver of the target vehicle for different vehicle performance dimensions based on the real-time driving behavior state data; based on each real-time attention data, it obtains the real-time attention weight of the driver for each vehicle performance dimension, and obtains the driver's real-time attention weight combination; when the target vehicle meets the preset range extender start-up conditions, it starts the range extender 102 of the target vehicle, and obtains the target power generation and target engine speed corresponding to the real-time attention weight combination according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations; and controls the range extender 102 using the target power generation and target engine speed. This application can determine the driver's real-time attention to different vehicle performance dimensions and the real-time attention weight of each vehicle performance dimension by collecting the real-time driving behavior status of the target vehicle. Then, when the range extender start-up conditions are met, the target power generation and target engine speed can be obtained according to the real-time attention weight and the first mapping relationship. The range extender 102 can then be controlled using the target power generation and target engine speed. This method can determine the importance of the driver's attention to different vehicle performance dimensions by combining the real-time driving behavior status, and thus complete the control of the range extender 102. Therefore, it can cope with instantaneous changes in attention, and the range extender control method can meet the user's real-time needs.

[0059] In one embodiment, such as Figure 3 As shown, step S202 may further include:

[0060] Step S301: If any real-time attention data is greater than or equal to the first attention threshold and any real-time attention data is less than the second attention threshold, obtain the driver's first attention weight for each vehicle performance dimension based on each real-time attention data; the first attention threshold is less than the second attention threshold.

[0061] The first attention threshold can be a pre-set low attention threshold, while the second attention threshold can be a pre-set high attention threshold. For example, the first attention threshold can be set to 0.10~0.20, and the second attention threshold can be set to 0.80~0.90. The first attention weight is the attention weight obtained directly based on real-time attention data without undergoing mixed correction processing.

[0062] In this embodiment, if the real-time attention data of any dimension is greater than or equal to the first attention threshold, and any real-time attention data is less than the second attention threshold, that is, not all real-time attention data of vehicle performance dimensions are less than the first attention threshold, or all real-time attention data of vehicle performance dimensions are greater than the second attention threshold, then the driver's first attention weight for each vehicle performance dimension can be calculated based on the real-time attention data of each vehicle performance dimension.

[0063] Step S302: Obtain the pre-set uniform attention weights for each vehicle performance dimension, and use the pre-set residual mixing coefficient to weight each first attention weight and each uniform attention weight to obtain the driver's second attention weights for each vehicle performance dimension.

[0064] The uniform attention weight is a pre-set uniform residual weight, which can be set according to the number of vehicle performance dimensions. For example, if there are three vehicle performance dimensions, the uniform attention weight can be set to 1 / 3. The residual mixing coefficient refers to the fusion coefficient used to combine the uniform attention weight and the first attention weight. It reflects the minimum degree of retention of low-attention dimensions by the driver and can be set to 0.05~0.15. In this embodiment, the first attention weight and the uniform attention weight can be combined using this fusion coefficient to obtain the driver's second attention weight for each vehicle performance dimension. This second attention weight represents the attention weight obtained after residual uniform mixing processing.

[0065] For example, the second attention weight can be calculated using the following formula:

[0066]

[0067] in, This represents the second attention weight for the i-th vehicle performance dimension. This represents the first attention weight for the i-th vehicle performance dimension. This represents the fusion coefficient, and 1 / 3 represents the uniform attention weight.

[0068] Step S303: Utilize the weights of each second level of attention to obtain the weights of each real-time level of attention.

[0069] Finally, the second attention weights of each vehicle performance dimension can be used to obtain the real-time attention weights of each vehicle performance dimension. For example, the second attention weights can be directly used as the real-time attention weights, or the second attention weights can be further modified to obtain the real-time attention weights.

[0070] In this embodiment, if any real-time attention data is greater than or equal to the first attention threshold and less than the second attention threshold, the first attention weight can be obtained based on the real-time attention data. Then, the first attention weight and the uniform attention weight can be fused using the residual mixing coefficient to obtain the second attention weight. Finally, the real-time attention weight can be obtained. This method can improve the accuracy of obtaining the real-time attention weight.

[0071] Furthermore, step S301 may further include: mapping each real-time attention data using a preset temperature coefficient to obtain real-time attention mapping data corresponding to each vehicle performance dimension; and normalizing each real-time attention mapping data to obtain each first attention weight.

[0072] The temperature coefficient is a pre-set coefficient used to control the sharpness of the weight distribution, for example, it can be set to 0.15~0.30. The real-time attention mapping data is the attention data obtained by mapping using the temperature coefficient. In this embodiment, the controller 101 can first use the temperature coefficient to map each real-time attention data to obtain the real-time attention mapping data corresponding to each vehicle performance dimension. Finally, the controller 101 can normalize each real-time attention mapping data to obtain the first attention weight of each vehicle performance dimension.

[0073] For example, vehicle performance dimensions may include NVH performance, fuel consumption performance, and battery life performance. The weight of the primary focus can then be calculated using the following formula:

[0074]

[0075] in, This represents the first attention weight for the i-th vehicle performance dimension. This represents the real-time attention data for the i-th vehicle performance dimension. This represents the temperature coefficient.

[0076] In this embodiment, real-time attention mapping data corresponding to each vehicle performance dimension can also be obtained through temperature coefficient mapping, and then normalized to obtain the first attention weight. This method can improve the calculation accuracy of the first attention weight.

[0077] Furthermore, the real-time attention weight is the driver's attention weight for different vehicle performance dimensions in the current sampling period; step S304 may further include: obtaining a preset number of historical attention weights; wherein, each historical attention weight is the driver's attention weight for different vehicle performance dimensions in each historical sampling period, and each historical sampling period is a preset number of sampling periods before the current sampling period; using each historical attention weight, smoothing each second attention weight to obtain the real-time attention weight.

[0078] The current sampling period refers to the current period for collecting real-time driving behavior status data. In this embodiment, the real-time driving behavior status data is sampled according to a preset period. The historical attention weight is the attention weight calculated based on the driving behavior status data collected in the historical sampling period. The historical sampling period refers to a preset number of sampling periods before the current sampling period, such as four sampling periods before the current sampling period.

[0079] Specifically, the controller 101 can obtain the historical attention weights of the driver to different vehicle performance dimensions in the four historical sampling periods before the current sampling period, and then use the historical attention weights to smooth the second attention weights to obtain the real-time attention weights. This method can effectively eliminate the impact of single input fluctuations on the weight output.

[0080] For example, the real-time attention weight is calculated using the following formula:

[0081]

[0082] in, Indicates the real-time attention weight. The number of steps for the sliding window can be set. , This indicates the second level of attention weight. This indicates the sampling period interval, which can be set to... .

[0083] In this embodiment, the historical attention weights of a preset number of sampling periods prior to the current sampling period can also be used to perform a time-series moving average on the second attention weight, which can effectively eliminate the impact of single input fluctuations on the weight output and further improve the accuracy of the weight output.

[0084] Additionally, step S202 may further include: when all real-time attention data are less than a first attention threshold, obtaining a first default attention weight pre-set for each vehicle performance dimension, and using each first default attention weight as a real-time attention weight; when all real-time attention data are greater than a second attention threshold, obtaining a second default attention weight pre-set for each vehicle performance dimension, and using each second default attention weight as a real-time attention weight; wherein the second default attention weights pre-set for each vehicle performance dimension are evenly distributed.

[0085] If all real-time attention data are less than the first attention threshold, that is, the maximum value of the real-time attention data is less than the first attention threshold, it can be expressed as: ,in These represent real-time attention data for NVH performance, fuel consumption performance, and battery life performance, respectively. This represents the first attention threshold. At this point, the driver may not express any explicit performance preference throughout the entire process. In this case, the pre-set first default attention weight for each vehicle performance dimension can be used as the real-time attention weight for each vehicle performance dimension. For example, if the default weights for the three vehicle performance dimensions are (0.50, 0.30, 0.20), then these default weights can be used as the real-time attention weights.

[0086] However, if all real-time attention data are greater than the second attention threshold, then the condition is met. At that time, among them The second attention threshold is when the driver has strong and clear demands for NVH, fuel consumption, and battery life. In this case, the second default attention weights that are pre-set for each vehicle performance dimension can be used as the real-time attention weights for each vehicle performance dimension. The second default attention weights can be evenly distributed. For example, if the default weights for the three vehicle performance dimensions are (1 / 3, 1 / 3, 1 / 3), then the above default weights can be used as the real-time attention weights.

[0087] In this embodiment, for cases where all real-time attention data are less than the first attention threshold and all real-time attention data are greater than the second attention threshold, attention weight can be set by setting a default attention weight. This method can further ensure the accuracy of the weight setting.

[0088] In one embodiment, such as Figure 4 As shown, step S201 may further include:

[0089] Step S401: Based on real-time driving behavior status data, obtain real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions.

[0090] In this embodiment, the real-time attention data can consist of two parts: firstly, the impact data on the driver's attention to different vehicle performance dimensions caused by the environment, i.e., real-time environmental impact attention data, which can serve as the foundation of the real-time attention data; and secondly, the impact data on the driver's attention to different vehicle performance dimensions caused by short-term interactive events, i.e., real-time event impact attention data, which can serve as the trigger for the real-time attention data. Specifically, after obtaining the real-time driving behavior state data, the controller 101 can simultaneously calculate the driver's real-time environmental impact attention data for different vehicle performance dimensions, as well as the real-time event impact attention data.

[0091] Step S402: Using a pre-set fusion coefficient, the real-time environmental impact attention data and real-time event impact attention data of each vehicle performance dimension are fused to obtain the driver's initial real-time attention data for different vehicle performance dimensions.

[0092] The initial real-time attention data refers to the attention data obtained by fusing real-time environmental impact attention data and real-time event impact attention data. The fusion coefficient is the coefficient used to fuse real-time environmental impact attention data and real-time event impact attention data. In this embodiment, the controller 101 can fuse the real-time environmental impact attention data and real-time event impact attention data of the corresponding vehicle performance dimension through the pre-set fusion coefficient, thereby obtaining the driver's initial real-time attention data for different vehicle performance dimensions.

[0093] Step S403: Obtain the attention data thresholds that are pre-set for each vehicle performance dimension, and take the smaller value between each attention data threshold and each initial real-time attention data as the driver's real-time attention data for each vehicle performance dimension.

[0094] After obtaining the initial real-time attention data, the controller 101 can compare the initial real-time attention data with the attention data thresholds that are pre-set for each vehicle performance dimension. The attention data thresholds can be set to 1, so that the smaller value between the attention data thresholds and the initial real-time attention data is used as the real-time attention data.

[0095] For example, real-time attention data can be calculated using the following formula:

[0096]

[0097] in, This represents the real-time attention data for the i-th dimension, where 1 represents the attention data threshold. and This represents the fusion coefficient, which can be set to 0.7 and 0.3 respectively. This represents the real-time environmental impact attention data for the i-th dimension, while This represents the real-time event impact and attention data for the i-th dimension.

[0098] In this embodiment, the influence of environmental factors on driver attention and the influence of interactive events on driver attention can be calculated first. Then, the data can be fused using a fusion coefficient and compared with a threshold. The smaller value is taken as the real-time attention data. This method can improve the accuracy of real-time attention data calculation.

[0099] Furthermore, the vehicle performance dimension includes the NVH performance dimension; step S401 may further include: obtaining first environmental impact data associated with the NVH performance dimension based on real-time driving behavior state data; the first environmental impact data includes at least one of the following: vehicle speed, window opening, external microphone decibel value, and music volume; using the first environmental impact data, obtaining real-time environmental impact attention data corresponding to the NVH performance dimension; obtaining each first short-term behavioral event associated with the NVH performance dimension triggered within a preset time window based on the real-time driving behavior state data; the first short-term behavioral events include at least one of the following: sudden increase in music volume, sudden change in accelerator pedal position, driver looking at the central control screen, air conditioning adjustment and telephone call connection, and facial expression and grip change events; using the trigger time difference and event type weight associated with each first short-term behavioral event, obtaining real-time event impact attention data corresponding to the NVH performance dimension.

[0100] In this embodiment, the vehicle performance dimension can be the NVH performance dimension, the first environmental impact data refers to the data of the real-time attention data base item used to calculate the NVH performance dimension, and the first short-term behavioral event refers to the short-term behavioral event related to the calculation of the real-time attention data trigger item of the NVH performance dimension.

[0101] In this embodiment, the first environmental impact data may include the target vehicle's speed, window opening, external microphone decibel level, and music volume. Therefore, the controller 101 can calculate the basic items of real-time attention data for the NVH performance dimension based on the vehicle speed, window opening, external microphone decibel level, and music volume, i.e., the real-time environmental impact attention data corresponding to the NVH performance dimension. The first short-term behavioral events include music volume surge events, accelerator pedal abrupt change events, driver looking at the central control screen events, air conditioning adjustment and telephone call events, and facial expression and grip strength change events. Therefore, if the controller 101 detects the above-mentioned first short-term behavioral events, it can also calculate the real-time attention data trigger items corresponding to the NVH performance dimension based on the first short-term behavioral events, i.e., the real-time event impact attention data.

[0102] Specifically, the real-time environmental impact data corresponding to the NVH performance dimension can be calculated using the following formula:

[0103]

[0104] in, This represents real-time environmental impact data corresponding to NVH performance. It is a saturation function. For vehicle speed, For the opening of the car window, This refers to the decibel value of the external microphone. For music volume, , , , These are the sensitivity coefficients for each parameter. , , , This corresponds to the threshold. A higher baseline value indicates weaker NVH suppression from the current environment and a higher level of driver attention to NVH, such as higher vehicle speed. The smaller the value, the stronger the environmental suppression of NVH, and the lower the driver's attention to NVH.

[0105] The real-time event impact data corresponding to the NVH performance dimension can be calculated using the following formula:

[0106]

[0107] in, This represents the real-time event impact and attention data corresponding to the NVH performance dimension. Let be the time difference between the current time and the i-th first short-term behavioral event, i.e., the triggering time difference of the i-th first short-term behavioral event. This is the decay time constant (e.g., 60s). Weights for event types, In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0108] The event type weights corresponding to each type of first short-term behavioral event are shown in Table 1:

[0109] Table 1. Schematic diagram of event type weights corresponding to the first short-term behavioral event.

[0110]

[0111] In this embodiment, real-time environmental impact attention data for the NVH performance dimension can be obtained based on the target vehicle's speed, window opening, external microphone decibel value, and music volume. Real-time event impact attention data for the NVH performance dimension can also be obtained based on events such as sudden increases in music volume, sudden changes in accelerator pedal position, driver's gaze at the central control screen, air conditioning adjustment and telephone call events, and changes in facial expression and grip strength. This method can improve the accuracy of NVH performance dimension attention data acquisition.

[0112] Meanwhile, the vehicle performance dimension also includes a fuel consumption performance dimension; step S401 may further include: obtaining second environmental impact data associated with the fuel consumption performance dimension based on real-time driving behavior state data; the second environmental impact data includes: the number of times the driver views the fuel consumption details page of the target vehicle within a preset historical time period, the actual fuel consumption data of the target vehicle, and the cumulative mileage data driven by fuel; using the second environmental impact data, obtaining real-time environmental impact attention data corresponding to the fuel consumption performance dimension; obtaining each second short-term behavioral event associated with the fuel consumption performance dimension triggered within a preset time window based on the real-time driving behavior state data; the second short-term behavioral events include: driver viewing fuel consumption events, rapid acceleration events, and emergency braking events; using the trigger time difference and event type weight associated with each second short-term behavioral event, obtaining real-time event impact attention data corresponding to the fuel consumption performance dimension.

[0113] Similarly, vehicle performance dimension can also be fuel consumption performance dimension. The second environmental impact data refers to the data of the real-time attention data base item used to calculate the fuel consumption performance dimension. The second short-term behavioral event refers to the short-term behavioral event related to the calculation of the real-time attention data trigger item of the fuel consumption performance dimension.

[0114] In this embodiment, the second environmental impact data may include the number of times the driver views the target vehicle's fuel consumption details page within a preset historical time period, the target vehicle's actual fuel consumption data, and the cumulative mileage data driven by fuel. Therefore, the controller 101 can calculate the basic item of real-time attention data for the fuel consumption performance dimension based on the number of times the driver views the target vehicle's fuel consumption details page within the preset historical time period, the target vehicle's actual fuel consumption data, and the cumulative mileage data driven by fuel. That is, the real-time environmental impact attention data corresponding to the fuel consumption performance dimension. The second short-term behavioral events include driver fuel consumption viewing events, rapid acceleration events, and sudden braking events. Therefore, if the controller 101 detects the above-mentioned second short-term behavioral events, it can also calculate the real-time attention data trigger item corresponding to the fuel consumption performance dimension based on the second short-term behavioral events. That is, the real-time event impact attention data.

[0115] Specifically, the real-time environmental impact data corresponding to the fuel consumption performance dimension can be calculated using the following formula:

[0116]

[0117]

[0118]

[0119]

[0120] in, This represents real-time environmental impact data corresponding to the fuel consumption performance dimension. It is a saturation function. The frequency with which drivers check their fuel consumption records within 7 days. This indicates the number of times the fuel consumption details page was viewed within the 7-day sliding window. The baseline value can be 3 to 5 times. This represents the target vehicle's actual fuel consumption / baseline fuel consumption. This is the actual fuel consumption data per 100 kilometers. The fuel consumption rating per 100 kilometers, such as 10L, is obtained from the vehicle's factory operating condition test. Indicates the percentage of fuel used for mileage. Cumulative mileage for oil-powered driving Initialize the range for oil-powered driving. To accumulate mileage, This is a pre-defined percentage benchmark; for example, 50% means that 50% of the total mileage was driven by fuel. , , The sensitivity coefficients for each factor parameter were evaluated and calibrated by engineering experts. The three factors were selected as follows: The form (high input = high attention) uses a product expression to represent the AND logic that "multidimensional signals must be significant simultaneously to be considered high attention," meaning that any factor at the baseline level ( When this happens, the overall signal will be suppressed to avoid misjudgment of a single point signal.

[0121] The real-time event impact and attention data corresponding to the fuel consumption performance dimension can be calculated using the following formula:

[0122]

[0123] in, This represents real-time event impact and attention data corresponding to the fuel consumption performance dimension. Let be the time difference between the current time and the i-th second short-term behavioral event, i.e., the triggering time difference of the i-th second short-term behavioral event. This is the decay time constant (e.g., 60s). Weights for event types, In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0124] The event type weights corresponding to each type of second short-term behavioral event are shown in Table 2:

[0125] Table 2. Schematic diagram of event type weights corresponding to the second short-term behavioral events.

[0126]

[0127] In this embodiment, real-time environmental impact attention data for the fuel consumption performance dimension can be obtained based on the number of times the driver views the fuel consumption details page of the target vehicle within a preset historical time period, the actual fuel consumption data of the target vehicle, and the cumulative mileage data driven by fuel. Furthermore, real-time event impact attention data for the fuel consumption performance dimension can be obtained based on the types of events such as fuel consumption events, rapid acceleration events, and emergency braking events viewed by the driver. This method can improve the accuracy of attention data acquisition for the fuel consumption performance dimension.

[0128] In addition, the vehicle performance dimension also includes the battery protection performance dimension; step S401 may further include: obtaining third environmental impact data associated with the battery protection performance dimension based on real-time driving behavior state data; the third environmental impact data includes: the number of times the driver views the target vehicle's state of charge within a preset historical time period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power; using the third environmental impact data, obtaining real-time environmental impact attention data corresponding to the battery protection performance dimension; obtaining each third short-term behavioral event associated with the battery protection performance dimension triggered within a preset time window based on the real-time driving behavior state data; the third short-term behavioral events include: driver viewing state of charge events and continuous high-speed driving events; using the trigger time difference and event type weight associated with each third short-term behavioral event, obtaining real-time event impact attention data corresponding to the battery protection performance dimension.

[0129] Vehicle performance dimension can also be battery protection performance dimension. The third environmental impact data refers to the data of the real-time attention data base item used to calculate the battery protection performance dimension. The third short-term behavioral event refers to the short-term behavioral event related to the calculation of the real-time attention data trigger item of the battery protection performance dimension.

[0130] In this embodiment, the third environmental impact data may include the number of times the driver views the target vehicle's state of charge within a preset historical period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power. Therefore, the controller 101 can calculate the basic items of real-time attention data for the power preservation performance dimension based on the number of times the driver views the target vehicle's state of charge within a preset historical period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power. That is, the real-time environmental impact attention data corresponding to the power preservation performance dimension. The third short-term behavioral events include: driver viewing state of charge events and continuous high-speed driving events. Therefore, if the controller 101 detects the above-mentioned third short-term behavioral events, it can also calculate the real-time attention data trigger items corresponding to the power preservation performance dimension based on the third short-term behavioral events. That is, the real-time event impact attention data.

[0131] Specifically, the real-time environmental impact data corresponding to the power supply performance dimension can be calculated using the following formula:

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138] in, This represents real-time environmental impact data corresponding to the power supply performance dimension. It is a saturation function. The frequency at which the driver checks the State of Charge (SOC) within 7 days. This indicates the number of times the SOC was viewed within a 7-day sliding window. The baseline value can be 3 to 5 times. This indicates the degree to which the state of charge deviates from the power supply target. To ensure the power supply target is, for example, 30%, This represents the current real-time state of charge. To ensure the degree of deviation from the power supply target is within a certain range, such as 0.2~0.3, Indicates the rate of decrease in state of charge. The rate of decrease in state of charge per unit time. The baseline value for the rate of decrease in state of charge per unit time is 5~8% / h. Indicates driving smoothness, This refers to the number of times the vehicle speed changes drastically within a preset distance, such as the number of events where the longitudinal acceleration is greater than 3 m / s² within 100 km. The baseline value for the number of times the vehicle speed changes drastically within a preset distance is, for example, 10-15 times. Indicates air conditioner energy consumption. For air conditioning or heating power, The reference value for air conditioning or heating power is 3~4 kW. , , , , The sensitivity coefficients for each factor parameter are evaluated and calibrated by engineering experts.

[0139] The real-time event impact and attention data corresponding to the power supply performance dimension can be calculated using the following formula:

[0140]

[0141] in, This represents real-time event impact and attention data corresponding to the power supply performance dimension. Let be the time difference between the current time and the i-th third short-term behavioral event, i.e., the triggering time difference of the i-th third short-term behavioral event. This is the decay time constant (e.g., 60s). Weights for event types, In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0142] The event type weights corresponding to each type of third short-term behavioral event are shown in Table 3:

[0143] Table 3. Schematic diagram of event type weights corresponding to the third short-term behavioral event.

[0144]

[0145] In this embodiment, real-time environmental impact attention data for the battery protection performance dimension can be obtained based on the number of times the driver views the target vehicle's state of charge within a preset historical time period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power. Additionally, real-time event impact attention data for the battery protection performance dimension can be obtained based on events such as the driver viewing the state of charge and continuous high-speed driving events. This method can improve the accuracy of attention data acquisition for the battery protection performance dimension.

[0146] In one embodiment, such as Figure 5 As shown, step S203 may further include:

[0147] Step S501: Based on real-time driving behavior status data, obtain the driver's NVH performance attention data during the start-up process of the range extender 102.

[0148] Since the start-up process of the range extender 102 is a step-like process, it is easily perceived by the driver. Therefore, a separate definition for the start-up NVH attention is defined, representing the driver's perception of the start-up process of the range extender 102 before it is started. This perception level can be obtained through NVH performance attention data. Specifically, the controller 101 can also obtain NVH performance attention data based on real-time driving behavior status data.

[0149] In step S502, if the real-time state of charge of the target vehicle is less than the first target state of charge and the NVH performance attention data is less than the preset start attention threshold, the range extender 102 is started.

[0150] The first target state of charge can be a pre-set target value for the starting state of charge of the range extender 102, and the starting attention threshold is a pre-set starting attention threshold for the range extender 102. In this embodiment, the range extender 102 is only started when the real-time state of charge of the target vehicle is less than the first target state of charge and the NVH performance attention data is less than the pre-set starting attention threshold. That is, the range extender 102 is started only when the real-time SOC < target SOC and the NVH attention data < starting threshold are simultaneously satisfied.

[0151] The range extender control method may also include:

[0152] Step S503: If the real-time state of charge of the target vehicle is greater than or equal to the second target state of charge, or if the NVH performance attention data is greater than or equal to the shutdown attention threshold, the range extender 102 is turned off; wherein, the shutdown attention threshold is greater than the startup attention threshold, and the second target state of charge is greater than the first target state of charge.

[0153] The second target state of charge is the pre-set target value for the shutdown state of the range extender 102, while the shutdown concern threshold is the pre-set shutdown concern threshold for the range extender 102. The shutdown concern threshold is greater than the startup concern threshold, and the second target state of charge is greater than the first target state of charge. This method can retain a certain hysteresis range, thereby avoiding repeated start-stop cycles.

[0154] Specifically, after the range extender 102 is started, it can continuously collect the real-time state of charge and NVH performance attention data of the target vehicle. If the real-time state of charge of the target vehicle is greater than or equal to the second target state of charge, or if the NVH performance attention data is greater than or equal to the shutdown attention threshold, the range extender 102 will be turned off.

[0155] In this embodiment, real-time state of charge and NVH performance attention data during the start-up process of the range extender 102 can also be collected simultaneously. That is, the start-up of the range extender 102 is controlled by considering both the real-time state of charge and the driver's NVH performance attention. This method can improve the start-up control accuracy of the range extender 102. At the same time, shutting down the range extender 102 also takes into account the real-time state of charge and the driver's NVH performance attention. Moreover, the set shutdown threshold is greater than the start-up threshold. This method can retain a certain hysteresis range, thereby avoiding repeated start-stop cycles.

[0156] Furthermore, the first target state of charge is obtained through the following steps: obtaining the starting state of charge corresponding to the real-time driving mode of the target vehicle; setting the starting state of charge as the first target state of charge when the driver has not set a destination for the target vehicle; obtaining the remaining distance between the target vehicle and the destination when the driver has set a destination for the target vehicle; and obtaining the first target state of charge based on the distance relationship between the starting state of charge, the pre-set maximum state of charge, the remaining distance, and the nominal pure electric driving distance of the target vehicle.

[0157] The initial state of charge (IBC) refers to the initial value of the first target state of charge. In this embodiment, the first target state of charge can be set based on whether the driver has set a destination for the target vehicle. If the driver has not set a destination for the target vehicle, the IBC can be directly set as the first target state of charge, for example, the IBC is 20%.

[0158] If the driver sets a destination for the target vehicle, the starting charge state needs to be corrected upwards. The first target charge state is obtained by calculating the remaining distance between the target vehicle and the destination, combined with the starting charge state, the preset maximum charge state such as 80%, and the target vehicle's nominal pure electric driving range such as 350km.

[0159] For example, when the driver sets a destination for the target vehicle, the first target state of charge can be calculated using the following formula:

[0160]

[0161] in, Indicates the first target state of charge. Indicates the starting state of charge. Indicates the maximum state of charge. This represents the remaining distance, while This indicates the nominal pure electric driving range.

[0162] In this embodiment, if the driver does not set a destination for the target vehicle, the starting charge state is directly used as the first target charge state. If a destination is set, the starting charge state can be corrected upwards to obtain the first target charge state. This method can improve the intelligence of setting the first target charge state.

[0163] In addition, step S501 may further include: obtaining the target vehicle's speed, window opening, longitudinal acceleration, and driving time based on real-time driving behavior status data; and obtaining NVH performance attention data using the vehicle speed, window opening, longitudinal acceleration, and driving time.

[0164] In this embodiment, NVH performance attention data can be calculated using the target vehicle's speed, window opening, longitudinal acceleration, and driving time. For example, it can be calculated using the following formula:

[0165]

[0166] in, This indicates data on NVH performance concerns. It is a saturation function. For vehicle speed, For the opening of the car window, For longitudinal acceleration, For travel time, , , , These are the sensitivity coefficients for each parameter. , , , This corresponds to the threshold. A higher baseline value indicates weaker NVH suppression during engine start-up, and a higher level of driver attention to NVH, such as when longitudinal acceleration is high. The smaller the value, the weaker the driver's perception of the range extender starting.

[0167] In this embodiment, NVH performance attention data can also be calculated based on the target vehicle's speed, window opening, longitudinal acceleration, and driving time. This method can improve the control accuracy of the range extender 102's start-stop control.

[0168] In one embodiment, vehicle performance dimensions include NVH performance, fuel consumption performance, and battery life performance; such as Figure 6 As shown, the first mapping relationship is constructed through the following steps:

[0169] Step S601: Obtain the first cost value corresponding to the NVH performance dimension, the second cost value corresponding to the fuel consumption performance dimension, and the third cost value corresponding to the power saving performance dimension for each power speed combination; each power speed combination includes power generation power and engine speed.

[0170] The first cost value, the second cost value, and the third cost value refer to the cost values ​​corresponding to different vehicle performance dimensions. The power speed combination can be composed of a power generation power and an engine speed. In this embodiment, after the controller 101 obtains different power generation power and engine speed, it can use them as power speed combinations. Then, it can obtain the cost values ​​of each power speed combination corresponding to the NVH performance dimension, fuel consumption performance dimension, and battery life performance dimension, respectively, as the first cost value, the second cost value, and the third cost value.

[0171] Step S602: Obtain the current attention weight combination, and use the current attention weight combination to weight the first cost value, second cost value and third cost value corresponding to each power speed combination to obtain the cost value of each power speed combination corresponding to the current attention weight combination; the current attention weight combination is any one of the different attention weight combinations.

[0172] The current attention weight combination refers to any one of the multiple attention weight combinations in the first mapping relationship that needs to be constructed. After obtaining the first cost value, the second cost value, and the third cost value for each power speed combination, the current attention weight combination can be used to weight the first cost value, the second cost value, and the third cost value to obtain the cost value of the current attention weight combination corresponding to different power speed combinations.

[0173] For example, the current focus weight combination corresponds to the power speed combination. The formula for calculating the cost value is as follows:

[0174]

[0175] in, , , These are the real-time attention weights corresponding to the NVH performance dimension, the fuel consumption performance dimension, and the battery life performance dimension, respectively, in the current attention weight combination. , , These represent the first cost value, the second cost value, and the third cost value, respectively. This indicates that the current focus weight combination corresponds to the power-speed combination. The cost value.

[0176] Step S603: Take the minimum cost value among all cost values ​​as the target cost value corresponding to the current attention weight combination, and take the power speed combination corresponding to the target cost value as the target power speed combination corresponding to the current attention weight combination.

[0177] Step S604: Construct the first mapping relationship by utilizing the correspondence between each current attention weight combination and the target power speed combination.

[0178] After obtaining the cost values ​​of different power-speed combinations corresponding to the current attention weight combination, the minimum cost value can be used as the target cost value corresponding to the previous attention weight combination, and the power-speed combination corresponding to the target cost value can be used as the target power-speed combination corresponding to the current attention weight combination. In this way, the target power-speed combination corresponding to each current attention weight combination can be obtained. Thus, by utilizing the correspondence between each current attention weight combination and the target power-speed combination, the first mapping relationship can be constructed. The first mapping relationship is shown in Table 4.

[0179] Table 4. Schematic diagram of the first mapping relationship

[0180]

[0181] In this embodiment, the first mapping relationship between the optimal power generation and engine speed corresponding to all possible weight combinations can be generated by performing multi-objective optimization offline. This method can improve the accuracy of the first mapping relationship construction.

[0182] Further, step S601 may further include: for any power-speed combination, obtaining a first cost value corresponding to the NVH performance dimension of the power-speed combination from a pre-constructed second mapping relationship; the second mapping relationship pre-stores the correspondence between different power-speed combinations and the first cost value; obtaining a second cost value corresponding to the fuel consumption performance dimension of the power-speed combination from a pre-constructed third mapping relationship; the third mapping relationship pre-stores the correspondence between different power-speed combinations and the second cost value; obtaining a third cost value corresponding to the power-saving performance dimension of the power-speed combination from a pre-constructed fourth mapping relationship; the fourth mapping relationship pre-stores the correspondence between different power-speed combinations and the third cost value.

[0183] In this embodiment, any power-speed combination corresponds to a first cost value, a second cost value, and a third cost value for different vehicle performance dimensions. These can be obtained through pre-constructed second, third, and fourth mapping relationships, respectively. The second mapping relationship stores the correspondence between different power-speed combinations and the first cost value, the third mapping relationship stores the correspondence between different power-speed combinations and the second cost value, and the fourth mapping relationship stores the correspondence between different power-speed combinations and the third cost value. These mapping relationships can be calibrated through bench testing. Therefore, the first cost value, the second cost value, and the third cost value can be obtained by querying the corresponding mapping relationship.

[0184] For example, the second mapping relationship can be shown in Table 5; the remaining third and fourth mapping relationships can also be stored in the form of other mapping tables.

[0185] Table 5. Schematic diagram of the second mapping relationship

[0186]

[0187] In this embodiment, the first cost value, second cost value, and third cost value corresponding to different vehicle performance dimensions can also be queried by querying a pre-built mapping relationship. This method can improve the efficiency of obtaining the first cost value, second cost value, and third cost value.

[0188] In one embodiment, step S204 may further include: when the target power generation exceeds the maximum power generation matching the vehicle speed, controlling the range extender 102 using the maximum power generation and the target engine speed; when the target power generation does not exceed the maximum power generation, controlling the range extender 102 using the target power generation and the target engine speed.

[0189] The maximum power generation capacity is a pre-set safe power limit, which can be determined based on the vehicle speed, for example, by looking up a table. Specifically, after obtaining the target vehicle's speed, the controller 101 can typically obtain the target vehicle's maximum power generation capacity by looking up a table, as shown in Table 6.

[0190] Table 6 Power Safety Boundary Diagram

[0191]

[0192] After obtaining the maximum power generation, it can be determined whether the current output target power generation exceeds the maximum power generation. If it does not exceed the maximum power generation, the range extender 102 is controlled using the target power generation and the target engine speed. If it exceeds the maximum power generation, the range extender 102 is controlled using the maximum power generation and the target engine speed.

[0193] In this embodiment, the maximum power generation can be set according to the vehicle speed to determine whether the target power generation exceeds the maximum power generation. If it does not exceed the maximum power generation, the range extender 102 is controlled using the target power generation and the target engine speed. If it exceeds the maximum power generation, the range extender 102 is controlled using the maximum power generation and the target engine speed. This method can ensure the safety of the range extender 102 control.

[0194] In one embodiment, the target power generation and the target engine speed are the target power generation value and the target engine speed value corresponding to the current sampling period; step S204 may further include: if the target power generation is greater than the historical power generation, after a preset time delay, controlling the range extender 102 using the target power generation and the target engine speed; the historical power generation is the target power generation value corresponding to the previous sampling period; if the target power generation is not greater than the historical power generation, immediately controlling the range extender 102 using the target power generation and the target engine speed.

[0195] The target power generation and target engine speed refer to the target values ​​for power generation control and engine speed control corresponding to the current sampling period. In this embodiment, anti-jitter processing is required for the target power generation and target engine speed output in step S203. Specifically, when the target power generation is greater than the historical power generation, i.e., when the power generation increases, the response control can be executed after a preset time delay, such as a 3-second delay. However, if the target power generation is not greater than the historical power generation, i.e., when the power generation does not increase or even decreases, an immediate response is required to ensure optimal NVH performance. That is, the range extender 102 is controlled immediately using the target power generation and target engine speed.

[0196] In this embodiment, anti-shake processing can also be performed on the target output power and target engine speed, which can further improve the intelligence of controlling the range extender 102.

[0197] In one embodiment, a multi-objective control method for range extenders based on driver attention perception is also provided, which can be applied to, for example... Figure 7 The system architecture shown includes the following steps:

[0198] Step 1, Attention Detection and Quantification: The intelligent cockpit system collects multi-source behavioral state data of the driver in real time, including driver visual attention data, cockpit interaction operation data, and vehicle status environment data. Based on the multi-source behavioral state data, a pre-built attention quantification model is used to calculate the driver's first attention weight for NVH performance, the second attention weight for fuel consumption performance, and the third attention weight for battery retention. The attention quantification model continuously quantifies the driver's unconscious behavioral characteristics, mapping the driver's subjective perception preferences into a numerical attention weight vector.

[0199] Step two, soft allocation of attention and weight generation: After identifying the driver's attention to the three dimensions of NVH, fuel consumption, and SOC in step one, this is a crucial step in continuously mapping these three attention values ​​proportionally to normalized weighted triples, providing a weighted basis for the start-stop door determination in step three. The core innovation of this step lies in using a soft comparison mechanism to replace the existing "primary and secondary dimensions + table lookup" hard allocation scheme, fundamentally eliminating the weight jump problem caused by discrete table lookup, and achieving a full-space continuous differentiable mapping from attention triples to weighted triples.

[0200] Step 3, Range Extender Control Decision Steps: Based on the dynamic weight vector and combined with the pre-stored range extender performance MAP, the optimal control parameters under the current operating conditions are determined by minimizing a multi-objective cost function. The optimal control parameters include at least: range extender start-stop state, target power generation, target engine speed and torque; among which, the range extender start-stop state is determined by NVH attention weight as the core gating condition: when the real-time SOC is lower than the target SOC start-up threshold and the NVH attention weight is lower than the start-up NVH tolerance threshold, the range extender is started; when the real-time SOC is higher than the target SOC shutdown threshold or the NVH attention weight is greater than the shutdown NVH tolerance threshold, the range extender is shut down.

[0201] Step 4, execution steps: Under the premise of meeting the preset safety boundary conditions, the optimal control parameters are sent to the range extender controller for execution, and the control parameters are de-jittered.

[0202] This method can be applied to range-extended electric vehicles equipped with a driver monitoring camera, central touchscreen, voice interaction module, vehicle speed sensor, air conditioning / audio system, in-vehicle and external microphone system, and smart cockpit system. The specific implementation of the above steps is shown below:

[0203] Step 1: Attention detection and quantification.

[0204] The attention level of each dimension is calculated using the multidimensional information shown in Table 7.

[0205] Table 7. Schematic diagram of attention computation dependency parameters

[0206]

[0207] Calculate the attention level for each dimension:

[0208] (1) NVH dimension:

[0209] NVH (Noise, Vibration, and Harshness) concerns during operation refer to the driver's perception of NVH during the operation of the range extender. This can be categorized into environmental impact and event-related impact, and the calculation formula is as follows:

[0210]

[0211] in , The fusion coefficient is determined by experts based on subjective NVH evaluation data from actual vehicles.

[0212] Basic items This reflects the steady-state suppression of NVH perception by the environment. Since different indicators have different dimensions, normalization is necessary. Furthermore, users perceive NVH as essentially "unnoticeable" below 0.3 and "significantly uncomfortable" above 0.7, with a transition zone in between. The sigmoid function naturally fits this psychophysical curve; therefore, the calculation formula is as follows:

[0213]

[0214] in, This represents real-time environmental impact data corresponding to NVH performance. It is a saturation function. For vehicle speed, For the opening of the car window, This refers to the decibel value of the external microphone. For music volume, , , , These are the sensitivity coefficients for each parameter. , , , This corresponds to the threshold. A higher baseline value indicates weaker NVH suppression from the current environment and a higher level of driver attention to NVH, such as higher vehicle speed. The smaller the value, the stronger the environmental suppression of NVH, and the lower the driver's attention to NVH.

[0215] Trigger item The impulse stimulus reflecting the attention garnered by short-term behavioral events, exponential decay, is the only mathematical shape that simultaneously satisfies the criteria of "starting point 1, ending point 0, continuous process, never exploding, and never returning to zero." This aligns well with the real memory pattern of drivers who "naturally forget but retain a faint impression." Therefore, the calculation formula is as follows:

[0216]

[0217] in, This represents the real-time event impact and attention data corresponding to the NVH performance dimension. Let be the time difference between the current time and the i-th first short-term behavioral event, i.e., the triggering time difference of the i-th first short-term behavioral event. This is the decay time constant (e.g., 60s). The weights for event types are shown in Table 1. In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0218] Meanwhile, since the range extender startup process is a step process that is easily perceived by the driver, a separate startup NVH focus is defined for the startup process. This refers to the driver's perception of the range extender startup process before it has started. This weight is only used for the start / stop door determination in step three and does not participate in the weight vector calculation in step two.

[0219]

[0220] in, This indicates data on NVH performance concerns. It is a saturation function. For vehicle speed, For the opening of the car window, For longitudinal acceleration, For travel time, , , , These are the sensitivity coefficients for each parameter. , , , This corresponds to the threshold. A higher baseline value indicates weaker NVH suppression during engine start-up, and a higher level of driver attention to NVH, such as when longitudinal acceleration is high. The smaller the value, the weaker the driver's perception of the range extender starting.

[0221] (2) Fuel consumption dimension:

[0222] Fuel consumption concern is used to evaluate how much drivers care about the current fuel consumption of the vehicle. Like NVH, it can also be divided into environmental impact and event impact, and the calculation formula is the same as NVH.

[0223]

[0224] in , The fusion coefficient is determined by experts based on subjective evaluation data from actual vehicles.

[0225] Basic items This reflects the driver's long-term sensitivity to fuel consumption.

[0226]

[0227] in, , , The sensitivity coefficients for each factor parameter were evaluated and calibrated by engineering experts. The three factors were selected as follows: The form (high input = high attention) uses a product expression to represent the AND logic that "multidimensional signals must be significant simultaneously to be considered high attention," meaning that any factor at the baseline level ( When this happens, the overall signal will be suppressed to avoid misjudgment of a single point signal.

[0228] The meanings of each factor are shown in Table 8:

[0229] Table 8. Schematic diagram of fuel consumption factor meaning

[0230]

[0231] The specific calculation method is as follows:

[0232] (View frequency):

[0233]

[0234] in, This indicates the number of times the fuel consumption details page was viewed within the 7-day sliding window. The baseline value can be 3 to 5 times.

[0235] (Actual / Baseline Fuel Consumption):

[0236]

[0237] in, This is the actual fuel consumption data per 100 kilometers. The fuel consumption per 100 kilometers is the calibrated value obtained from the factory operating condition test of the vehicle model, such as 10L.

[0238] (Fuel mileage percentage):

[0239]

[0240] in, Cumulative mileage for oil-powered driving Initialize the range for oil-powered driving. To accumulate mileage, This is a pre-defined percentage benchmark; for example, 50% means that 50% of the total mileage is driven by fuel.

[0241] Trigger item The trigger item reflects the pulse stimulus that short-term events cause to fuel consumption awareness—drivers' attention will instantly increase when they see something that just happened (rapid acceleration, sudden braking, sudden start of the range extender, etc.), and then they will press [the button]. Exponential decay:

[0242]

[0243] in, This represents real-time event impact and attention data corresponding to the fuel consumption performance dimension. Let be the time difference between the current time and the i-th second short-term behavioral event, i.e., the triggering time difference of the i-th second short-term behavioral event. This is the decay time constant (e.g., 60s). The weights for event types are shown in Table 2. In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0244] (3) SOC power retention:

[0245]

[0246] in The basic items are derived from vehicle operating conditions and usage habits:

[0247]

[0248] in, , , , , The sensitivity coefficients for each factor parameter are evaluated and calibrated by engineering experts.

[0249] The meanings of each factor are shown in Table 9:

[0250] Table 9. Schematic diagram of the meaning of power preservation factor

[0251]

[0252] The specific calculation methods for each factor are as follows:

[0253] (1) (View frequency):

[0254]

[0255] in, This indicates the number of times the SOC was viewed within a 7-day sliding window. The baseline value can be 3 to 5 times.

[0256] (2) (Power supply deviation):

[0257]

[0258] in, To ensure the power supply target is, for example, 30%, This represents the current real-time state of charge. The deviation from the power supply target should be within the benchmark value, such as 0.2~0.3.

[0259] (3) (SOC decay rate):

[0260]

[0261] in, The rate of decrease in state of charge per unit time. The baseline value for the rate of decrease in charge per unit time is 5~8% / h.

[0262] (4) (Driving smoothness):

[0263]

[0264] in, The baseline value for the number of times the vehicle speed changes drastically within a preset distance is set, such as 10 to 15 times.

[0265] (5) (Air conditioning energy consumption):

[0266]

[0267] in, For air conditioning or heating power, The reference value for air conditioning or heating power is 3~4 kW.

[0268] Trigger item The trigger term reflects the pulse excitation of short-term events on the SOC power preservation attention.

[0269]

[0270] in, This represents real-time event impact and attention data corresponding to the power supply performance dimension. Let be the time difference between the current time and the i-th third short-term behavioral event, i.e., the triggering time difference of the i-th third short-term behavioral event. This is the decay time constant (e.g., 60s). The weights for event types are shown in Table 3. In engineering practice, a sliding window, i.e., a preset time window, is used. Calculation, where , Events that do not occur are not included in the summation.

[0271] Step 2: Soft allocation of attention and generation of weights.

[0272] Step 2 receives the three attention values ​​output from Step 1. It needs to be converted into the corresponding weighted triplet. (Saving that the sum of the three is 1 and each is non-negative), used for dynamic adjustment of control parameters in step three.

[0273] This step uses the softmax function to map attention triples to weight triples.

[0274] (1) Formula for generating core rights reorganization.

[0275]

[0276] in, , Indicates the temperature coefficient. The recommended value for T is 0.15 to 0.30.

[0277] (2) The residuals are uniformly mixed.

[0278] To reflect the lag in the perception of battery life and fuel consumption, this step further performs residual uniform mixing on the softmax output:

[0279]

[0280] in The residual mixing coefficient is... This reflects the minimum level of driver retention for low-attention dimensions. A larger value indicates that more data is retained (more "conservative"), while a smaller value indicates that the focus is more on high-priority dimensions (more "sensitive"). This step is recommended. The value range is 0.05 to 0.15.

[0281] (3) Attention threshold branch processing.

[0282] To cover the two boundary scenarios of "all three are low" and "all three are high", this step introduces an attention threshold branch:

[0283] Low attention as a safety net: When At this time, skip the core weight reorganization generation step and directly output the pre-calibrated neutral default weights. ). Recommended value: 0.10~0.20.

[0284] Full Height Mode: When At that time, the output is uniformly high-weighted ( ). Recommended value: 0.80~0.90.

[0285] (4) Time-series average sliding.

[0286] To eliminate the impact of single input fluctuations on the weighted output, this step applies a moving average to the final weights:

[0287]

[0288] in, Indicates the real-time attention weight. The number of steps for the sliding window can be set. , This indicates the second level of attention weight. This indicates the sampling period interval, which can be set to... .

[0289] Calculation example:

[0290] enter: .

[0291] calculate: , , The total is 71.26.

[0292] ;

[0293] .

[0294] Output: ).

[0295] Step 3: Range extender control decision.

[0296] (1) Start-up and shutdown decision.

[0297] Based on the output of step 1, determine whether to initiate the NVH (Noise, Vibration, and Harshness) concern decision-making process, with the following strategy:

[0298] When real-time SOC < target SOC and NVH concern level < start-up threshold: the decision range extender starts; when real-time SOC > target SOC or NVH concern level > shutdown threshold: the decision range extender shuts down; otherwise: the decision range extender remains at the previous time.

[0299] Wherein: Real-time SOC refers to the current SOC reported by BMS, and target SOC refers to the preset SOC for starting the vehicle mode (e.g., 20%). When the smart cockpit has a set destination, in order to alleviate battery anxiety, the target SOC needs to be adjusted upwards, and the algorithm is as follows:

[0300]

[0301] In the formula, This indicates an initial SOC of 20%. This indicates the preset maximum state of charge, such as 80%. Indicates the navigation distance from the destination in km. This indicates the vehicle's nominal pure electric driving range, such as 350km.

[0302] The shutdown threshold is set higher than the startup threshold. For example, if the startup threshold is 80, then the shutdown threshold is 85, which provides a certain hysteresis range to avoid repeated startup and shutdown.

[0303] (2) Power speed decision.

[0304] The vehicle controller is based on the weights in step two, as follows: Nearest neighbor matching is performed in a pre-stored weight-optimal operating point lookup table. This table is designed to meet the real-time requirements of vehicle control and reduce computational load. Offline multi-objective optimization generates the optimal power generation and engine speed corresponding to all possible weight combinations, as shown in Table 4.

[0305] The optimal combination table of all weight vectors is calculated by using a cost function to select all power-speed combinations that maximize the overall cost. The smallest group.

[0306]

[0307] in, , , The weights for NVH, fuel consumption, and battery life are respectively, ranging from 0 to 1. , , These represent the cost values ​​for NVH, fuel consumption, and battery retention, respectively. This information can be obtained from Table 5. Higher values ​​indicate worse NVH (Noise, Vibration, and Harshness), and the unit is subjective cost points. Fuel consumption cost. and SOC cost Similarly, these results were obtained through bench tests.

[0308] Step 4: Execute.

[0309] Based on the output of step three, the vehicle controller controls the EMS and GCU to complete the start-stop process according to the start decision. Based on the speed decision in step three, it requests the speed of the GCU; based on the power decision in step three, it obtains the power and calculates the torque request from the vehicle controller to the EMS.

[0310] Security boundary settings:

[0311] For startup, set a safety boundary: startup is prohibited when SOC > 85%. Startup is mandatory when SOC < a certain threshold (e.g., 10%).

[0312] For power, a safety boundary is set to ensure that the maximum power does not exceed the tolerance limit for normal driving. The tolerance limit for power is determined by expert evaluation, as shown in Table 6.

[0313] Image stabilization:

[0314] The output from step three requires anti-jitter processing. When the power operating point shifts, if the power increases, the response should be delayed (e.g., a 3-second confirmation period) to the target power to avoid jitter. When the power decreases, an immediate response is required to ensure optimal NVH performance.

[0315] This embodiment proposes a multi-objective adaptive control method for range extenders based on attention perception and behavior prediction. Its core framework includes: an attention-driven multi-objective dynamic weighted control framework, which uses the driver's real-time attention to NVH, fuel consumption, and battery reserve as weights to dynamically adjust the range extender's start-stop timing, power generation, and engine speed and torque, ensuring the control objective is always aligned with "what the user cares about most at this moment," rather than a single engineering optimal solution; an attention perception model, by integrating information from cameras in the smart cockpit, the central control interface, and vehicle sensors, constructs an attention quantification module. By analyzing the driver's unconscious behaviors such as the frequency of checking fuel / battery levels, music volume, window status, and accelerator pedal depth, it quantifies the driver's attention to each dimension in real time, achieving deep integration of the smart cockpit and the power domain; and a prediction-feedback closed-loop mechanism, which uses short-term driving behaviors (such as frequent battery checks and repeated volume adjustments) to predict future attention tendencies and adjust control strategies in advance, achieving a leap from passive response to proactive prediction.

[0316] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0317] Based on the same inventive concept, this application also provides a range extender control device for implementing the range extender control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more range extender control device embodiments provided below can be found in the limitations of the range extender control method described above, and will not be repeated here.

[0318] In one embodiment, such as Figure 8 As shown, a range extender control device is provided, including: a attention data acquisition module 801, an attention weight acquisition module 802, a control target acquisition module 803, and a range extender control module 804, wherein:

[0319] The attention data acquisition module 801 is used to acquire real-time driving behavior status data for the target vehicle, and to acquire real-time attention data of the driver of the target vehicle on different vehicle performance dimensions based on the real-time driving behavior status data.

[0320] The attention weight acquisition module 802 is used to obtain the driver's real-time attention weight for each vehicle performance dimension based on the real-time attention data, and to obtain the driver's real-time attention weight combination.

[0321] The target acquisition module 803 is used to start the range extender of the target vehicle when the target vehicle meets the preset range extender start conditions, and to acquire the target power generation and target engine speed corresponding to the real-time attention weight combination according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations.

[0322] The range extender control module 804 is used to control the range extender using the target power generation and the target engine speed.

[0323] In one embodiment, the attention weight acquisition module 802 is further configured to: acquire the driver's first attention weight for each vehicle performance dimension based on each real-time attention data when any real-time attention data is greater than or equal to a first attention threshold and any real-time attention data is less than a second attention threshold; acquire the uniform attention weight pre-set for each vehicle performance dimension; weight each first attention weight and each uniform attention weight using a pre-set residual mixing coefficient to obtain the driver's second attention weight for each vehicle performance dimension; and obtain each real-time attention weight using each second attention weight.

[0324] In one embodiment, the attention weight acquisition module 802 is further configured to map each real-time attention data using a preset temperature coefficient to obtain real-time attention mapping data corresponding to each vehicle performance dimension; and to normalize each real-time attention mapping data to obtain each first attention weight.

[0325] In one embodiment, the real-time attention weight is the driver's attention weight for different vehicle performance dimensions in the current sampling period; the attention weight acquisition module 802 is further used to acquire a preset number of historical attention weights; wherein, each historical attention weight is the driver's attention weight for different vehicle performance dimensions in each historical sampling period, and each historical sampling period is a preset number of sampling periods before the current sampling period; using each historical attention weight, each second attention weight is smoothed to obtain the real-time attention weight.

[0326] In one embodiment, the attention weight acquisition module 802 is further configured to: when all real-time attention data are less than a first attention threshold, acquire a first default attention weight pre-set for each vehicle performance dimension, and use each first default attention weight as a real-time attention weight; when all real-time attention data are greater than a second attention threshold, acquire a second default attention weight pre-set for each vehicle performance dimension, and use each second default attention weight as a real-time attention weight; wherein the second default attention weights pre-set for each vehicle performance dimension are evenly distributed.

[0327] In one embodiment, the attention data acquisition module 801 is further configured to acquire, based on real-time driving behavior state data, real-time environmental impact attention data and real-time event impact attention data of the driver for different vehicle performance dimensions; use a pre-set fusion coefficient to fuse the real-time environmental impact attention data and real-time event impact attention data of each vehicle performance dimension to obtain the driver's initial real-time attention data for different vehicle performance dimensions; acquire the attention data thresholds pre-set for each vehicle performance dimension, and use the smaller value between each attention data threshold and each initial real-time attention data as the driver's real-time attention data for each vehicle performance dimension.

[0328] In one embodiment, the vehicle performance dimension includes the NVH performance dimension; the attention data acquisition module 801 is further configured to acquire first environmental impact data associated with the NVH performance dimension based on real-time driving behavior state data; the first environmental impact data includes at least one of the following: vehicle speed, window opening, external microphone decibel value, and music volume; using the first environmental impact data, real-time environmental impact attention data corresponding to the NVH performance dimension is obtained; based on the real-time driving behavior state data, each first short-term behavioral event associated with the NVH performance dimension triggered within a preset time window is acquired; the first short-term behavioral events include at least one of the following: sudden increase in music volume, sudden change in accelerator pedal position, driver looking at the central control screen, air conditioning adjustment and telephone call connection, and facial expression and grip change events; using the trigger time difference and event type weight associated with each first short-term behavioral event, real-time event impact attention data corresponding to the NVH performance dimension is obtained.

[0329] In one embodiment, the vehicle performance dimension includes a fuel consumption performance dimension; the attention data acquisition module 801 is further used to acquire second environmental impact data associated with the fuel consumption performance dimension based on real-time driving behavior state data; the second environmental impact data includes: the number of times the driver views the fuel consumption details page of the target vehicle within a preset historical time period, the actual fuel consumption data of the target vehicle, and the cumulative mileage data driven by fuel; using the second environmental impact data, real-time environmental impact attention data corresponding to the fuel consumption performance dimension is obtained; based on the real-time driving behavior state data, each second short-term behavioral event associated with the fuel consumption performance dimension triggered within a preset time window is acquired; the second short-term behavioral events include: driver viewing fuel consumption events, rapid acceleration events, and emergency braking events; using the trigger time difference and event type weight associated with each second short-term behavioral event, real-time event impact attention data corresponding to the fuel consumption performance dimension is obtained.

[0330] In one embodiment, the vehicle performance dimension includes a battery maintenance performance dimension. The attention data acquisition module 801 is further configured to acquire third environmental impact data associated with the battery maintenance performance dimension based on real-time driving behavior state data. The third environmental impact data includes: the number of times the driver views the target vehicle's state of charge within a preset historical time period, the target vehicle's real-time state of charge, the real-time state of charge change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the target vehicle's air conditioning or heating power. Using the third environmental impact data, real-time environmental impact attention data corresponding to the battery maintenance performance dimension is obtained. Based on the real-time driving behavior state data, each third short-term behavioral event associated with the battery maintenance performance dimension triggered within a preset time window is acquired. The third short-term behavioral events include: driver viewing of state of charge events and continuous high-speed driving events. Using the trigger time difference and event type weight associated with each third short-term behavioral event, real-time event impact attention data corresponding to the battery maintenance performance dimension is obtained.

[0331] In one embodiment, the control target acquisition module 803 is further configured to acquire NVH performance attention data of the driver during the range extender startup process based on real-time driving behavior state data; start the range extender when the real-time state of charge of the target vehicle is less than the first target state of charge and the NVH performance attention data is less than a preset startup attention threshold; and shut down the range extender when the real-time state of charge of the target vehicle is greater than or equal to the second target state of charge, or the NVH performance attention data is greater than or equal to the shutdown attention threshold; wherein the shutdown attention threshold is greater than the startup attention threshold, and the second target state of charge is greater than the first target state of charge.

[0332] In one embodiment, the target acquisition module 803 is further configured to acquire the starting charge state corresponding to the real-time driving mode of the target vehicle; if the driver has not set a destination for the target vehicle, the starting charge state is set as the first target charge state; if the driver has set a destination for the target vehicle, the remaining distance between the target vehicle and the destination is acquired; and the first target charge state is obtained based on the distance relationship between the starting charge state, the preset maximum charge state, the remaining distance and the nominal pure electric driving distance of the target vehicle.

[0333] In one embodiment, the target acquisition module 803 is further configured to acquire the vehicle speed, window opening, longitudinal acceleration, and driving time of the target vehicle based on real-time driving behavior status data; and to obtain NVH performance attention data using the vehicle speed, window opening, longitudinal acceleration, and driving time.

[0334] In one embodiment, the vehicle performance dimension includes NVH performance, fuel consumption performance, and battery life performance. The range extender control device further includes: a mapping relationship construction module, used to obtain a first cost value corresponding to the NVH performance dimension, a second cost value corresponding to the fuel consumption performance dimension, and a third cost value corresponding to the battery life performance dimension for each power-speed combination; each power-speed combination includes generator power and engine speed; obtain the current attention weight combination, and use the current attention weight combination to weight the first cost value, second cost value, and third cost value corresponding to each power-speed combination to obtain the cost value of the current attention weight combination corresponding to each power-speed combination; the current attention weight combination is any one of different attention weight combinations; the minimum cost value among the cost values ​​is taken as the target cost value corresponding to the current attention weight combination, and the power-speed combination corresponding to the target cost value is taken as the target power-speed combination corresponding to the current attention weight combination; a first mapping relationship is constructed using the correspondence between each current attention weight combination and the target power-speed combination.

[0335] In one embodiment, the mapping relationship construction module is further configured to, for any power-speed combination, obtain a first cost value corresponding to the NVH performance dimension of the power-speed combination from a pre-constructed second mapping relationship; the second mapping relationship pre-stores the correspondence between different power-speed combinations and the first cost value; obtain a second cost value corresponding to the fuel consumption performance dimension of the power-speed combination from a pre-constructed third mapping relationship; the third mapping relationship pre-stores the correspondence between different power-speed combinations and the second cost value; obtain a third cost value corresponding to the power-saving performance dimension of the power-speed combination from a pre-constructed fourth mapping relationship; the fourth mapping relationship pre-stores the correspondence between different power-speed combinations and the third cost value.

[0336] In one embodiment, the range extender control module 804 is further configured to control the range extender using the maximum power generation and the target engine speed when the target power generation exceeds the maximum power generation matching the vehicle speed; and to control the range extender using the target power generation and the target engine speed when the target power generation does not exceed the maximum power generation.

[0337] In one embodiment, the range extender control module 804 is further configured to, when the target power generation is greater than the historical power generation, delay for a preset period of time and then control the range extender using the target power generation and the target engine speed; the historical power generation is the target power generation value corresponding to the previous sampling period; when the target power generation is not greater than the historical power generation, immediately control the range extender using the target power generation and the target engine speed.

[0338] Each module in the aforementioned range extender control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the controller in hardware form or independent of it, or stored in the memory of the controller in software form, so that the processor can call and execute the corresponding operations of each module.

[0339] In one embodiment, a controller is provided, the internal structure of which can be as follows: Figure 9 As shown. The controller includes a processor and memory. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium storing a computer program. When executed by the processor, the computer program implements a range extender control method.

[0340] Those skilled in the art will understand that Figure 9The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the controller (controller) applied thereto. A specific controller may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0341] In one embodiment, a controller is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps in the above method embodiments.

[0342] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0343] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0344] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0345] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program mentioned can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0346] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0347] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A range extender control method, characterized in that, The method includes: Acquire real-time driving behavior status data for the target vehicle, and obtain real-time attention data of the driver of the target vehicle to different vehicle performance dimensions based on the real-time driving behavior status data. Based on the real-time attention data, the real-time attention weights of the driver for each of the vehicle performance dimensions are obtained, and the real-time attention weight combination of the driver is obtained. When the target vehicle meets the preset range extender start-up conditions, the range extender of the target vehicle is started, and the target power generation and target engine speed corresponding to the real-time attention weight combination are obtained according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations. The range extender is controlled using the target power generation and the target engine speed.

2. The method according to claim 1, characterized in that, The step of obtaining the driver's real-time attention weight for each of the vehicle performance dimensions based on the real-time attention data includes: If any of the real-time attention data is greater than or equal to the first attention threshold, and if any of the real-time attention data is less than the second attention threshold, the driver's first attention weight for each of the vehicle performance dimensions is obtained based on each of the real-time attention data; the first attention threshold is less than the second attention threshold. Obtain the uniform attention weights pre-set for each of the vehicle performance dimensions, and use the pre-set residual mixing coefficient to weight each of the first attention weights and each of the uniform attention weights to obtain the second attention weights of the driver for each of the vehicle performance dimensions. The real-time attention weights are obtained by using the second attention weights.

3. The method according to claim 2, characterized in that, The step of obtaining the driver's first attention weight for each of the vehicle performance dimensions based on the real-time attention data includes: By using a preset temperature coefficient to map each of the real-time attention data, real-time attention mapping data corresponding to each of the vehicle performance dimensions is obtained; The real-time attention mapping data is normalized to obtain the first attention weight.

4. The method according to claim 2, characterized in that, The real-time attention weight is the driver's attention weight for different vehicle performance dimensions in the current sampling period; obtaining each of the real-time attention weights using each of the second attention weights includes: Obtain a preset number of historical attention weights; wherein, each historical attention weight is the driver's attention weight for different vehicle performance dimensions in each historical sampling period, and each historical sampling period is a preset number of sampling periods before the current sampling period; The real-time attention weights are obtained by smoothing each of the historical attention weights.

5. The method according to claim 2, characterized in that, The step of obtaining the driver's real-time attention weight corresponding to each of the vehicle performance dimensions based on the real-time attention data further includes: When all the real-time attention data are less than the first attention threshold, a first default attention weight is obtained for each of the vehicle performance dimensions, and each of the first default attention weights is used as the real-time attention weight. When all the real-time attention data are greater than the second attention threshold, a second default attention weight is obtained that is pre-set for each of the vehicle performance dimensions, and each of the second default attention weights is used as the real-time attention weight; wherein, the second default attention weights pre-set for each of the vehicle performance dimensions are evenly distributed.

6. The method according to claim 1, characterized in that, The step of obtaining real-time attention data of the driver of the target vehicle to different vehicle performance dimensions based on the real-time driving behavior state data includes: Based on the real-time driving behavior status data, obtain the driver's real-time environmental impact attention data for different vehicle performance dimensions, as well as real-time event impact attention data; Using a pre-set fusion coefficient, the real-time environmental impact attention data and real-time event impact attention data of each vehicle performance dimension are fused to obtain the driver's initial real-time attention data for different vehicle performance dimensions. Obtain the attention data thresholds that are pre-set for each of the vehicle performance dimensions, and take the smaller value between each attention data threshold and each of the initial real-time attention data as the driver's real-time attention data for each of the vehicle performance dimensions.

7. The method according to claim 6, characterized in that, The vehicle performance dimensions include NVH performance dimensions; the step of obtaining the driver's real-time environmental impact attention data and real-time event impact attention data for different vehicle performance dimensions based on the real-time driving behavior state data includes: Based on the real-time driving behavior status data, the first environmental impact data associated with the NVH performance dimension is obtained; the first environmental impact data includes at least one of the following: vehicle speed, window opening, external microphone decibel value, and music volume of the target vehicle. Using the first environmental impact data, real-time environmental impact attention data corresponding to the NVH performance dimension is obtained; Based on the real-time driving behavior state data, obtain each first short-term behavioral event associated with the NVH performance dimension triggered within a preset time window; the first short-term behavioral events include at least one of the following: sudden increase in music volume event, sudden change in accelerator pedal event, driver looking at the central control screen event, air conditioning adjustment and telephone call event, and facial expression and grip change event. By utilizing the trigger time difference and event type weight associated with each of the first short-term behavioral events, the real-time event impact attention data corresponding to the NVH performance dimension is obtained.

8. The method according to claim 6, characterized in that, The vehicle performance dimensions include fuel consumption performance; the process of obtaining real-time environmental impact attention data and real-time event impact attention data of the driver regarding different vehicle performance dimensions based on the real-time driving behavior state data includes: Based on the real-time driving behavior status data, second environmental impact data associated with the fuel consumption performance dimension is obtained; the second environmental impact data includes: the number of times the driver views the fuel consumption details page of the target vehicle within a preset historical time period, the actual fuel consumption data of the target vehicle, and the cumulative mileage data driven by fuel. Using the second environmental impact data, real-time environmental impact attention data corresponding to the fuel consumption performance dimension is obtained; Based on the real-time driving behavior status data, obtain each of the second short-term behavioral events associated with the fuel consumption performance dimension that are triggered within a preset time window; the second short-term behavioral events include: driver viewing fuel consumption events, rapid acceleration events, and emergency braking events; By utilizing the trigger time difference and event type weight associated with each of the second short-term behavioral events, real-time event impact attention data corresponding to the fuel consumption performance dimension is obtained.

9. The method according to claim 6, characterized in that, The vehicle performance dimensions include a battery retention performance dimension; the step of obtaining the driver's real-time environmental impact attention data and real-time event impact attention data for different vehicle performance dimensions based on the real-time driving behavior state data includes: Based on the real-time driving behavior status data, obtain the third environmental impact data associated with the power preservation performance dimension; the third environmental impact data includes: the number of times the driver views the charge status of the target vehicle within a preset historical time period, the real-time charge status of the target vehicle, the real-time charge status change rate, the number of times the target vehicle's speed changes drastically within a preset distance, and the air conditioning or heating power of the target vehicle. Using the third environmental impact data, we can obtain the real-time environmental impact attention data corresponding to the power supply performance dimension; Based on the real-time driving behavior status data, obtain each third short-term behavioral event associated with the battery protection performance dimension that is triggered within a preset time window; the third short-term behavioral events include: driver viewing charge status event and continuous high-speed driving event; By utilizing the trigger time difference and event type weight associated with each of the aforementioned third short-term behavioral events, real-time event impact attention data corresponding to the power supply performance dimension is obtained.

10. The method according to claim 1, characterized in that, The step of activating the range extender of the target vehicle when the target vehicle meets the preset range extender activation conditions includes: Based on the real-time driving behavior status data, obtain the driver's NVH performance attention data during the range extender startup process; The range extender is activated when the real-time state of charge of the target vehicle is less than the first target state of charge and the NVH performance attention data is less than the preset start attention threshold. The method further includes: The range extender is shut down when the real-time state of charge of the target vehicle is greater than or equal to the second target state of charge, or when the NVH performance concern data is greater than or equal to the shutdown concern threshold; wherein the shutdown concern threshold is greater than the startup concern threshold, and the second target state of charge is greater than the first target state of charge.

11. The method according to claim 10, characterized in that, The first target state of charge is obtained through the following steps: Obtain the startup charge state corresponding to the real-time driving mode of the target vehicle; If the driver has not set a destination for the target vehicle, the starting charge state is set to the first target charge state; When the driver sets a destination for the target vehicle, the remaining distance between the target vehicle and the destination is obtained; based on the starting state of charge, the preset maximum state of charge, and the distance relationship between the remaining distance and the nominal pure electric driving distance of the target vehicle, the first target state of charge is obtained.

12. The method according to claim 10, characterized in that, The step of obtaining the driver's NVH performance attention data during the range extender startup process based on the real-time driving behavior status data includes: Based on the real-time driving behavior status data, the vehicle speed, window opening, longitudinal acceleration, and driving time of the target vehicle are obtained. The NVH performance attention data is obtained by using the vehicle speed, window opening, longitudinal acceleration, and driving time.

13. The method according to claim 1, characterized in that, The vehicle performance dimensions include NVH performance, fuel consumption performance, and battery life performance; the first mapping relationship is constructed through the following steps: Obtain the first cost value corresponding to the NVH performance dimension, the second cost value corresponding to the fuel consumption performance dimension, and the third cost value corresponding to the power retention performance dimension for each power speed combination; each power speed combination includes power generation and engine speed; Obtain the current attention weight combination, and use the current attention weight combination to weight the first cost value, second cost value, and third cost value corresponding to each power speed combination to obtain the cost value of the current attention weight combination for each power speed combination; the current attention weight combination is any one of the different attention weight combinations; The minimum cost value among all the cost values ​​is taken as the target cost value corresponding to the current attention weight combination, and the power speed combination corresponding to the target cost value is taken as the target power speed combination corresponding to the current attention weight combination. The first mapping relationship is constructed by utilizing the correspondence between each current attention weight combination and the target power speed combination.

14. The method according to claim 13, characterized in that, The acquisition of the first cost value corresponding to the NVH performance dimension, the second cost value corresponding to the fuel consumption performance dimension, and the third cost value corresponding to the battery life performance dimension for each power speed combination includes: For any power-speed combination, a first cost value corresponding to the NVH performance dimension of the power-speed combination is obtained from a pre-constructed second mapping relationship; the second mapping relationship pre-stores the correspondence between different power-speed combinations and the first cost value. From the pre-constructed third mapping relationship, the second cost value corresponding to the fuel consumption performance dimension of the power speed combination is obtained; the third mapping relationship pre-stores the correspondence between different power speed combinations and the second cost value; From the pre-constructed fourth mapping relationship, the third cost value corresponding to the power speed combination in the power-saving performance dimension is obtained; the fourth mapping relationship pre-stores the correspondence between different power speed combinations and the third cost value.

15. The method according to any one of claims 1 to 14, characterized in that, The control of the range extender using the target power generation and the target engine speed includes: If the target power generation exceeds the maximum power generation matching the vehicle speed of the target vehicle, the range extender is controlled using the maximum power generation and the target engine speed. If the target power generation does not exceed the maximum power generation, the range extender is controlled using the target power generation and the target engine speed.

16. The method according to any one of claims 1 to 14, characterized in that, The target power generation and target engine speed are the target power generation value and target engine speed value corresponding to the current sampling period; the control of the range extender using the target power generation and target engine speed includes: If the target power generation is greater than the historical power generation, the range extender is controlled using the target power generation and the target engine speed after a preset delay; the historical power generation is the target power generation value corresponding to the previous sampling period. If the target power generation is not greater than the historical power generation, the range extender is immediately controlled using the target power generation and the target engine speed.

17. A range extender control device, characterized in that, The device includes: The attention data acquisition module is used to acquire real-time driving behavior status data for the target vehicle, and to acquire real-time attention data of the driver of the target vehicle on different vehicle performance dimensions based on the real-time driving behavior status data. The attention weight acquisition module is used to obtain the driver's real-time attention weight for each of the vehicle performance dimensions based on the real-time attention data, and to obtain the driver's real-time attention weight combination. The target acquisition module is used to start the range extender of the target vehicle when the target vehicle meets the preset range extender start conditions, and to obtain the target power generation and target engine speed corresponding to the real-time attention weight combination according to the pre-constructed first mapping relationship; the first mapping relationship pre-stores the power generation and engine speed corresponding to different attention weight combinations. The range extender control module is used to control the range extender using the target power generation and the target engine speed.

18. A controller comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 16.

19. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 16.

20. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 16.

Images