A method, apparatus, system and storage medium for generator voltage control

Abstract

Embodiments of the present application provide a generator voltage control method, device, system and storage medium, the method comprising: determining the current driving state of the vehicle according to the vehicle operating parameters of the vehicle where the generator is located; determining the predicted driving state of the vehicle according to the video data collected by the vehicle; determining the target voltage of the generator based on the current driving state and the predicted driving state; and adjusting the voltage of the generator according to the target voltage. With this method, the driving state of the vehicle after the current driving state is predicted based on the collected video data, and then the predicted driving state is corrected based on the current driving state, and then the target voltage of the generator is determined based on the corrected driving state and the state of charge of the vehicle battery and other signals. Compared with the existing generator control scheme, the time lag is long, and the voltage curve of the ideal situation and the actual adjustment situation of the generator is fitted through the calibration of the parameters, which greatly reduces the time difference, improves the response ability of the vehicle, and reduces the energy consumption.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of generator technology, and in particular to a generator voltage control method, device, system and storage medium. BACKGROUND

[0002] The main function of the automobile generator is to supply power to all electrical equipment except the starter when the engine is running normally, and to charge the battery. The generator is the main power source of the automobile.

[0003] In the prior art, the intelligent control of the generator is usually implemented by the following scheme: first, the signals are collected from the messages sent by various sensors and controllers on the vehicle to the bus, which usually include power mode signals, battery state of charge (SOC) signals, engine speed signals, vehicle speed signals, vehicle acceleration signals, fuel consumption signals, throttle position signals and brake pedal position signals; second, the collected signals are preprocessed, which usually includes filtering, data type conversion, relay mode state distinction, etc.; third, the processed signals are calculated to obtain the current vehicle driving state; fourth, the target value of the generator output voltage is determined according to the current vehicle driving state combined with the vehicle battery SOC signal and other fault signals.

[0004] However, this scheme has a long time lag, and the actual adjustment of the generator voltage usually lags behind the ideal situation by 1 to 2 seconds, which leads to the waste of part of the energy in the kinetic energy recovery process and the loss of the acceleration of the vehicle in the early stage of the acceleration process. SUMMARY

[0005] The embodiments of the present application provide a generator voltage control method, device, system and storage medium to reduce the time difference between the actual adjustment of the generator and the ideal situation, improve the response ability of the vehicle and reduce the energy consumption.

[0006] In a first aspect, the present application provides a generator voltage control method, which comprises:

[0007] determining the current driving state of the vehicle according to the vehicle operating parameters of the vehicle where the generator is located;

[0008] determining the predicted driving state of the vehicle according to the video data collected by the vehicle;

[0009] determining the target voltage of the generator based on the current driving state and the predicted driving state;

[0010] adjusting the voltage of the generator according to the target voltage.

[0011] In a second aspect, the embodiment provides a generator power generation control device, which comprises:

[0012] a current state determination module configured to determine a current driving state of the vehicle according to a vehicle operation parameter of the vehicle in which the generator is located;

[0013] a predicted state determination module configured to determine a predicted driving state of the vehicle according to video data collected by the vehicle;

[0014] a target voltage determination module configured to determine a target voltage of the generator based on the current driving state and the predicted driving state;

[0015] an adjustment module configured to perform voltage adjustment on the generator according to the target voltage.

[0016] In a third aspect, the embodiment provides a generator voltage control system, which comprises:

[0017] at least one processor; and

[0018] a memory connected to the at least one processor in communication; wherein

[0019] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the generator voltage control method according to any of the embodiments.

[0020] In a fourth aspect, the embodiment provides a computer readable storage medium, which stores computer instructions for enabling a valve control device to execute the generator voltage control method according to any of the embodiments.

[0021] The embodiment of the present application provides a generator voltage control method, device, system and storage medium, and the method comprises: determining a current driving state of a vehicle where a generator is located according to a vehicle operation parameter of the vehicle; determining a predicted driving state of the vehicle according to video data collected by the vehicle; determining a target voltage of the generator based on the current driving state and the predicted driving state; and adjusting the voltage of the generator according to the target voltage. According to the above technical solution, the driving state of the vehicle after the vehicle is predicted according to the collected video data, then the predicted driving state is corrected based on the current driving state, and finally the target voltage of the generator is determined based on the corrected driving state and a vehicle storage battery charge state lamp signal; compared with the long time lag of the generator control scheme in the prior art, the present technical solution greatly reduces the time difference by calibrating the parameters, fitting the voltage curve of the ideal situation of the generator and the actual adjustment situation, improves the vehicle response ability, and reduces the energy waste.

[0022] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0024] Figure 1 A flowchart of a generator voltage control method provided for the first embodiment of the present application is shown in the figure;

[0025] Figure 2 A flowchart of a generator voltage control method provided for the second embodiment of the present application is shown in the figure;

[0026] Figure 3 A structural schematic diagram of a generator voltage control device provided for the third embodiment of the present application is shown in the figure;

[0027] Figure 4 A structural block diagram of a generator voltage control system provided for the fourth embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0028] In the following, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative effort should belong to the protection scope of the present application.

[0029] It should be noted that the terms "original", "target" and the like in the description, claims, and drawings of the present application are used to distinguish similar objects, and are not necessarily used to describe a particular chronological or sequential order. It should be understood that the data thus used can be interchanged, so that the embodiments of the present application described herein can be carried out in sequences other than those illustrated or described herein. Moreover, the terms "comprise" and "have", and any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product, or apparatus that comprises a list of steps or units is not necessarily limited to those steps or units that are clearly listed, but can include other steps or units that are not clearly listed or inherent to such processes, methods, products, or apparatus.

[0030] Embodiment One

[0031] Figure 1 A flowchart of a generator voltage control method provided by the embodiment one of the present application is shown in the figure. The method can be applied to the case of accurate control of the generator voltage. The method can be executed by a generator voltage control device, which can be realized in the form of hardware and / or software. The device can be configured in a generator voltage control system.

[0032] As shown in the figure, the generator voltage control method provided by the embodiment one can specifically include the following steps: Figure 1

[0033] S101, determining a current driving state of the vehicle according to a vehicle operating parameter of the vehicle in which the generator is located.

[0034] The vehicle operating parameter includes a power supply mode signal, a battery SOC signal, an engine speed signal, a vehicle speed signal, a vehicle acceleration signal, a throttle position signal, a brake pedal position signal, and a main brake state signal, etc. The current driving state of the vehicle includes that the vehicle engine is off or in a starting process, the vehicle is in an idle state, the vehicle is in an acceleration state, the vehicle is in a constant speed state, and the vehicle is in a deceleration state.

[0035] ​The signals used in the messages sent from various sensors and controllers on the bus in this embodiment are collected to determine the vehicle operating parameters. Each vehicle operating parameter plays a different role and represents a different state of the vehicle and related equipment. Among them, the power mode signal is used to determine whether the vehicle is in the power-on state, and when the vehicle is in the power-off state, the generator should not generate power. The battery SOC signal is used to check the current battery power, when the battery power is very high, the generator should not generate power, and vice versa when the battery power is low, the generator should output at a higher voltage. The engine speed signal can determine whether the vehicle is in the starting process, if in the starting process, the generator should not generate power. The vehicle speed signal, vehicle acceleration signal, fuel consumption signal, throttle position signal and brake pedal position signal can be used to determine the vehicle driving state. Under normal circumstances, when the vehicle decelerates, the generator output voltage is increased for energy recovery and increased braking, and when the vehicle accelerates, the generator output voltage is reduced to improve vehicle acceleration performance. When the active brake signal is triggered, the generator reaches the maximum output voltage at the fastest speed (with the largest adjustment step) to enhance the braking ability.

[0036] In this step, the collected vehicle operating parameters are preprocessed, including filtering, data type conversion, relay mode, and state processing. Then the processed vehicle operating parameters are calculated to obtain the current vehicle driving state. For example, the power mode signal is Boolean processed to determine the vehicle ignition switch state value; the battery state of charge is processed according to the relay mode to determine the battery state value; the vehicle speed signal is processed according to the relay mode to determine the vehicle speed state value; the engine speed signal is processed according to the relay mode to determine the engine speed state value; and the vehicle acceleration signal is processed according to the relay mode to determine the vehicle acceleration state value.

[0037] For example, the power mode signal is 0, indicating that the ignition switch is in IG OFF, that is, all the vehicle except the normal fire is not powered; the value is 4, indicating that the ignition switch is in IG ON, that is, the automobile ignition gear. The ignition switch state value IG_state is obtained by Boolean processing of the power mode signal, that is, IG_state=0 indicates IG OFF, and IG_state=1 indicates IG ON. In this embodiment, the vehicle acceleration signal needs to be processed by low-pass filtering to remove noise to obtain the vehicle acceleration signal aVeh. When the vehicle acceleration aVeh is higher than the limit value 0.35 m / s2, the acceleration acceleration state value aVeh_Acc used to judge the acceleration state is 1, and the vehicle is in the acceleration state; otherwise, if it is lower than the limit value 0.25 m / s2, aVeh_Acc=0, and the vehicle is not in the acceleration state, wherein 0.25 m / s2 to 0.35 m / s2 is the relay area; when the vehicle acceleration aVeh is lower than the limit value -0.35 m / s2, the acceleration deceleration state value aVeh_Dec used to judge the acceleration state is 1, and the vehicle is in the deceleration state; otherwise, if it is higher than the limit value -0.25 m / s2, aVeh_Dec=0, and the vehicle is not in the deceleration state, wherein -0.25 m / s2 to -0.35 m / s2 is the relay area. Similarly, the operating parameters can be processed according to the corresponding rules of the parameters to obtain the engine speed state value, the vehicle speed state value, and the acceleration state value.

[0038] Further, according to the state value of each operating parameter, the current state of the vehicle is determined in combination with the preset judgment rule. For example, when the engine speed first state value nEng_idl=1 and the vehicle speed state value vVeh_idl=1 and the acceleration deceleration state value aVeh_Dec=1 and the brake state value brake_state=1, Drv_state=4 is determined, and the vehicle is in the deceleration.

[0039] S102, according to the video data collected by the vehicle, the predicted driving state of the vehicle is determined.

[0040] Among them, the predicted driving state of the vehicle includes acceleration driving, constant speed driving and deceleration driving. In this embodiment, the video data collected by the vehicle needs to be analyzed to determine the predicted driving state of the vehicle.

[0041] In this embodiment, the video data of the environment around the vehicle is collected in real time by the radar camera device installed on the vehicle. The collected video data is frame-sampled. The sampling method can be to cut the video data by time period, and to filter out the most representative image in each time period by using the mature deep convolutional network (CNN) model trained. For example, the time period can be set to 0.1 second, that is, 10 images can be extracted from the video in 1 second.

[0042] In this step, the two-dimensional matrix data obtained by pre-processing the video data is operated by a deep neural network model to obtain the conclusion of the above three predicted driving states of the vehicle. The deep neural network model includes a convolution module and a recurrent module. The two-dimensional matrix data is input into the convolution module, which is used to extract convolution features from the two-dimensional matrix data and calculate a vector. The vector obtained after calculation is connected to obtain the output result of the convolution module.

[0043] The output result of the convolution module is input into the recurrent module as input data for calculation, and finally a vector is obtained. Each bit in the vector represents the probability of the vehicle being in the acceleration driving, constant speed driving and deceleration driving states after 2 seconds. The maximum value of the three probabilities in the output vector is the final result. Considering that the actual adjustment of the generator voltage in the prior art usually has a 1-2 second lag compared to the ideal situation, in this step, the predicted driving state of the vehicle after 2 seconds is obtained by pre-processing and calculating the video data. For example, the final result is (0.2, 0.25, 0.55), which represents that the probability of the vehicle being in the acceleration driving state is 20%, the probability of being in the constant speed driving state is 25%, and the probability of being in the deceleration driving state is 55%. Therefore, the prediction result of the driving state of the vehicle after 2 seconds is deceleration driving.

[0044] S103, determining the target voltage of the generator based on the current driving state and the predicted driving state.

[0045] In this step, the predicted driving state is corrected using the current driving state, and the corrected driving state is used in combination with the vehicle battery SOC signal and other fault signals to determine the target value of the generator output voltage. By predicting the driving state of the vehicle in the near future (i.e. the lag time of the existing scheme), the existing scheme is compensated, thereby reducing the time lag of the existing scheme.

[0046] Specifically, if the current driving state is that the vehicle engine is off or in the starting process, the target driving state is that the vehicle engine is off or in the starting process. If the current driving state is that the vehicle is in the idle state, the target driving state is that the vehicle is in the idle state. If the current driving state is the acceleration state or the uniform speed state or the deceleration state, the target driving state needs to be determined in combination with the predicted driving state. For example, if the current driving state of the vehicle cannot be consistent with the predicted driving state within a set time (such as 5 seconds), it indicates that the predicted driving state is inaccurate, and the current driving state of the vehicle needs to be determined as the target driving state.

[0047] If the current driving state of the vehicle can be consistent with the predicted driving state within a set time (such as 5 seconds), it indicates that the predicted driving state is accurate, and the next step needs to be entered for judgment. Continue to judge which one of the current driving state and the predicted driving state changes first. If the current driving state changes first, the current driving state is the target driving state. If the predicted driving state changes first, the target driving state is maintained at the current driving state for 0.5 seconds. If the predicted driving state does not change within 0.5 seconds, the target driving state is maintained at the predicted driving state for 1.5 seconds. Then, it is judged whether there is an equal time within 0.5 seconds after the current driving state and the predicted driving state and the equal time accounts for 85%. If yes, the target driving state is maintained at the predicted driving state. Otherwise, the target driving state is maintained at the current driving state.

[0048] S104, according to the target voltage, voltage regulation is performed on the generator.

[0049] Among them, different generator voltage control modes are pre-associated with corresponding target voltages, so that the corresponding target voltage is determined through the generator voltage control mode, and is used for voltage regulation on the generator, thereby realizing dynamic voltage regulation on the generator under different generator control modes.

[0050] This invention provides a generator voltage control method, which includes: determining the current driving state of the vehicle based on its operating parameters; determining the predicted driving state of the vehicle based on video data collected from the vehicle; determining the target voltage of the generator based on the current driving state and the predicted driving state; and adjusting the generator voltage according to the target voltage. Using this method, the vehicle's future driving state is predicted based on the collected video data, then the predicted driving state is corrected based on the current driving state, and finally the target voltage of the generator is determined based on the corrected driving state combined with the vehicle's battery status charge indicator signal. Compared to existing generator control schemes with long time lags that lead to energy waste during kinetic energy recovery and loss of acceleration in the initial stage of vehicle acceleration, this technical solution significantly reduces the time lag by calibrating parameters and fitting the voltage curve between the ideal and actual generator conditions, thereby improving vehicle responsiveness and reducing energy waste.

[0051] Example 2

[0052] Figure 2 This is a flowchart of a generator voltage control method provided in Embodiment 2 of the present invention. This embodiment is a further optimization of the above embodiment. In this embodiment, the definition of "determining the current driving state of the vehicle based on the vehicle operating parameters of the vehicle where the generator is located" is further optimized, the definition of "determining the predicted driving state of the vehicle based on the video data collected by the vehicle" is optimized, and the definition of "determining the target voltage of the generator based on the current driving state and the predicted driving state" is optimized.

[0053] like Figure 2 As shown in the figure, this embodiment 2 provides a generator voltage control method, which specifically includes the following steps:

[0054] S201. Collect the operating parameters of each type of vehicle in the vehicle where the generator is located.

[0055] Specifically, various vehicle operating parameters of different types are collected from messages sent to the bus by various sensors and controllers on the vehicle. These vehicle operating parameters include power mode signals, battery SOC signals, engine speed signals, vehicle speed signals, vehicle acceleration signals, throttle position signals, brake pedal position signals, and active braking status signals, etc.

[0056] S202. For each set type of vehicle operating parameter, process the vehicle operating parameter according to the corresponding preset processing rules to obtain the processed parameter status value.

[0057] The preset processing rules include filtering processing, data type conversion processing, relay mode processing, and other processing modes to distinguish the device state or vehicle driving state. For different setting types of vehicle parameters, the corresponding preset processing rules are used to process the vehicle operating parameters to determine the parameter state value. For example, the power mode signal is subjected to Boolean processing to determine the vehicle ignition switch state value. The SOC state value is determined by the relay mode for the battery SOC signal. For the engine speed signal, the vehicle speed signal, and the vehicle acceleration signal, low-pass filtering processing is required to remove noise, and then the corresponding parameter state value is determined based on the relay mode.

[0058] Further, for each setting type of vehicle operating parameter, the vehicle operating parameter is processed according to the corresponding preset processing rule of the vehicle operating parameter to obtain the processed parameter state value, including:

[0059] a1) For the power mode signal in the vehicle operating parameter, the power mode signal is subjected to Boolean processing to determine the vehicle ignition switch state value.

[0060] Specifically, considering that the power mode signal value of 0 indicates that the ignition switch is in IG OFF, i.e., all vehicles except the constant fire are not powered; the value of 4 indicates that the ignition switch is in IG ON, i.e., the automobile ignition gear. The ignition switch state value IG_state is obtained by Boolean processing of the power mode signal, i.e., IG_state=0 indicates IG OFF, and IG_state=1 indicates IG ON.

[0061] b1) For the battery state of charge signal in the vehicle operating parameter, the battery state of charge is processed according to the relay mode to determine the battery state value.

[0062] The battery SOC signal is converted into the SOC state value SOC_state by the relay mode. It has four state values: when SOC≤70%, SOC_state=0, the battery is depleted. When 65%≤SOC≤80%, SOC_state=1, the battery is slightly depleted. When 75%≤SOC≤95%, SOC_state=2, the battery is normally charged. When 90%≤SOC, SOC_state=3, the battery is overcharged. 65% to 70%, 75% to 80%, and 90% to 95% are the relay zones before the four states.

[0063] c1) For the vehicle speed signal in the vehicle operating parameter, the vehicle speed signal is processed according to the relay mode to determine the vehicle speed state value.

[0064] In this embodiment, the vehicle speed signal needs to be processed by low-pass filtering to remove noise and obtain the vehicle speed value vVeh. When the vehicle speed value vVeh is higher than the limit value 3 km / h, the vehicle speed state value vVeh_idl for judging the idling state is 1, and at this time the vehicle is in a running state; otherwise, if it is lower than the limit value 2 km / h, vVeh_idl = 0, and at this time the vehicle is in a stopped state, wherein 2 km / h to 3 km / h is a relay zone.

[0065] d1) For the engine speed signal in the vehicle operating parameter, the engine speed signal is processed in a relay manner to determine the engine speed state value.

[0066] In this embodiment, the engine speed signal needs to be processed by low-pass filtering to remove noise and obtain the engine speed nGen. When the engine speed nGen is higher than the limit value 1000 rpm, the engine speed first state value nEng_idl = 1, and at this time the vehicle is in a possible running state; otherwise, if it is lower than the limit value 900 rpm, nEng_idl = 0, and at this time the vehicle is in an idling or stopped state, wherein 900 rpm to 1000 rpm is a relay zone.

[0067] When the engine speed nGen is higher than the limit value 600 rpm, the engine speed second state value nEng_run = 1, and at this time the vehicle is in a possible starting state; otherwise, if it is lower than the limit value 500 rpm, nEng_run = 0, and at this time the vehicle is in a stopped state, wherein 500 rpm to 600 rpm is a relay zone.

[0068] e1) For the vehicle acceleration signal in the vehicle operating parameter, the vehicle acceleration signal is processed in a relay manner to determine the vehicle acceleration state value.

[0069] In this embodiment, the vehicle acceleration signal needs to be processed by low-pass filtering to remove noise and obtain the vehicle acceleration signal aVeh. When the vehicle acceleration aVeh is higher than the limit value 0.35 m / s 2 , the acceleration acceleration state value aVeh_Acc for judging the acceleration state is 1, and at this time the vehicle is in an acceleration state; otherwise, if it is lower than the limit value 0.25 m / s 2 , aVeh_Acc = 0, and at this time the vehicle is not in an acceleration state, wherein 0.25 m / s 2 to 0.35 m / s 2 is a relay zone.

[0070] When the vehicle acceleration aVeh is lower than the limit value -0.35 m / s 2 , the acceleration deceleration state value aVeh_Dec for judging the acceleration state is 1, and at this time the vehicle is in a deceleration state; otherwise, if it is higher than the limit value -0.25 m / s2 aVeh_Dec = 0, when the vehicle is not in deceleration, wherein -0.25 m / s 2 to -0.35 m / s 2 is the relay zone.

[0071] f1) For the accelerator position signal in the vehicle running parameters, the accelerator position signal is processed in a relay manner to determine the accelerator state value.

[0072] Specifically, the accelerator position signal Acc is converted into the accelerator state value Acc_state in a relay manner. When Acc≤20%, the accelerator state value Acc_state = 0, at this time it is determined that the driver is not stepping on the accelerator and has no active acceleration intention. When 15%≤Acc, Acc_state = 1, at this time it is determined that the driver is stepping on the accelerator and has an active acceleration intention, wherein 15% to 20% is the relay zone.

[0073] g1) For the brake pedal position signal in the vehicle running parameters, the brake pedal position signal is processed in a relay manner to determine the brake state value.

[0074] Specifically, the brake pedal position signal brake is converted into the brake state value brake_state in a relay manner. When brake≤25%, brake_state = 0, at this time it is determined that the driver is not stepping on the brake and has no active deceleration intention; when 20≤brake, brake_state = 1, at this time it is determined that the driver is stepping on the brake and has no active deceleration intention, wherein 20% to 25% is the relay zone.

[0075] h1) Determine the active brake state value in the vehicle running parameters.

[0076] Specifically, the active brake state value Act_brake is a Boolean value and does not need to be processed. The active brake state value Act_brake = 0 is an inactive active brake, and Act_brake = 1 is an active active brake.

[0077] S203, according to the parameter state values, combining the preset determination rule, determining the current driving state of the vehicle.

[0078] Among them, the current driving state includes that the vehicle engine is off or in the starting process, the vehicle is in an idle state, the vehicle is in an acceleration state, the vehicle is in a uniform speed state, and the vehicle is in a deceleration state. Assuming that Drv_state represents the current driving state value of the vehicle, the engine is off or in the starting process, Drv_state = 0; idle driving, Drv_state = 1; acceleration driving, Drv_state = 2; uniform speed driving, Drv_state = 3; deceleration driving, Drv_state = 4.

[0079] The preset determination rule can be expressed as: when the vehicle ignition switch state value IG_state = 0 or the engine speed first state value nEng_run = 0, it is determined that the current driving state value Drv_state = 0, that is, the vehicle engine is off or in the starting process;

[0080] When the vehicle ignition switch state value IG_state = 1 and the engine speed second state value nEng_run = 1 and the engine speed first state value nEng_idl = 0 and the vehicle speed state value vVeh_idl = 0, it is determined that Drv_state = 1, that is, the vehicle is in idle;

[0081] When the engine speed first state value nEng_idl = 1 and the vehicle speed state value vVeh_idl = 1 and the vehicle acceleration acceleration state value aVeh_Acc = 1 and the throttle state value Acc_state = 1, it is determined that Drv_state = 2, that is, the vehicle is accelerating;

[0082] When the engine speed first state value nEng_idl = 1 and the vehicle speed state value vVeh_idl = 1 and the vehicle acceleration acceleration state value aVeh_Acc = 0 and the vehicle acceleration deceleration state value aVeh_Dec = 0, it is determined that Drv_state = 3, that is, the vehicle is in uniform speed.

[0083] When the engine speed first state value nEng_idl = 1 and the vehicle speed state value vVeh_idl = 1 and the acceleration deceleration state value aVeh_Dec = 1 and the brake state value brake_state = 1, it is determined that Drv_state = 4, that is, the vehicle is in deceleration.

[0084] S204, input the video data collected by the vehicle into the pre-trained deep convolutional network model to obtain a target image.

[0085] In this step, the vehicle surrounding environment data is collected by radar camera and other devices, which is used to predict the future driving state of the vehicle. The collected video data is frame-sampled, and the sampling method is to cut the video by time period, and the most representative frame image in each time period is selected by the mature CNN model trained to be the target image. For example, the time period can be set to 0.1 seconds, that is, 10 images can be extracted from the video in 1 second.

[0086] S205, processing the target image according to the set image processing rule to obtain the target matrix data corresponding to the target image.

[0087] The setting image processing rule includes color adjustment, white balance, contrast balance, image correction, normalization and the like. Specifically, each target image is subjected to color adjustment, white balance, contrast balance, image correction and normalization to obtain two-dimensional matrix data for calculation.

[0088] S206, input the target matrix data into the deep neural network model to determine the predicted driving state of the vehicle.

[0089] The deep neural network model includes a convolution module and a recurrent module. The predicted driving state of the vehicle includes acceleration, uniform speed and deceleration. Assuming that the predicted driving state value of the vehicle is Drv_st_pre, it can be divided into the following states: acceleration, Drv_st_pre = 0; uniform speed, Drv_st_pre = 1; deceleration, Drv_st_pre = 2.

[0090] It should be noted that the deep neural network model can be used in conjunction with real vehicle testing to obtain a large number of training samples for training the deep neural network model, which can make the model overfit to a certain extent, thereby improving the operation accuracy in the common mode.

[0091] In this step, the target matrix is operated through the deep neural network model to obtain the above three states. The two-dimensional matrix data is input into the convolution module, which is used to extract the convolution features of the two-dimensional matrix data and calculate the vector. The vector obtained after calculation is connected to obtain the output result of the convolution module.

[0092] The output result of the convolution module is input into the recurrent module as input data for calculation, and finally a vector is obtained, each bit of which represents the probability of the vehicle being in acceleration, uniform speed and deceleration after 2 seconds. The maximum value of the three probabilities in the output vector is the final result. It should be noted that the time delay is 2 seconds in the prior art, so the predicted driving state of the vehicle in this step is the driving state after 2 seconds.

[0093] Further, the target matrix data is input into the deep neural network model to determine the predicted driving state of the vehicle, including:

[0094] a2) input the target matrix data into the convolution module to obtain the output result, and the convolution module is composed of a set number of deep convolutional neural networks.

[0095] The convolution module is composed of 20 small deep convolutional neural networks (CNNs), each of which is responsible for extracting features from a two-dimensional matrix and calculating a 3xN length vector. The sigmoid function is used as the activation function for each bit of the vector, and the expression is: This function can make the calculation result between 0 and 100%, representing the probability of being in a certain state. The 3 in the vector represents the three states of accelerating, constant speed, and decelerating, and N represents the depth of the convolution kernel in the convolution network. The vectors obtained after the calculation of the 20 convolution networks are connected to obtain a two-dimensional matrix with a size of [3xN, 20], which is the output result of the convolution module.

[0096] b2) The output result is input into the cycle module to obtain the probability of the vehicle being in accelerating, constant speed, and decelerating states, respectively. The cycle module is composed of a set of deep recurrent neural networks (RNNs) with a specified length and two fully connected neural networks.

[0097] In this step, the output result of the convolution module is input into the cycle module for calculation, and a 3-length vector is finally obtained, each bit of which represents the probability of the vehicle being in accelerating, constant speed, and decelerating states after 2 seconds. The cycle module is composed of a set of deep recurrent neural networks (RNNs) with a length of 20 and two fully connected neural networks. The last fully connected network has three neurons with sigmoid activation functions. The RNN is responsible for calculating the internal relationship and trend of the input information of 20 3xN features, and the fully connected network is responsible for calculating the probabilities of the vehicle being in the three forms of states after 2 seconds based on the trend from the present moment to the previous 2 seconds.

[0098] c2) The driving state corresponding to the maximum value of the probability of the vehicle being in each driving state is determined as the predicted driving state of the vehicle.

[0099] Specifically, the above steps have determined the probabilities of the vehicle being in each driving state and output them in vector form. The maximum value of the three probabilities in the output vector is the final result. For example, the final result (0.2, 0.25, 0.55) represents that the probability of the vehicle being in accelerating state is 20%, the probability of being in constant speed is 25%, and the probability of being in decelerating state is 55%. Therefore, the predicted result of the vehicle driving state after 2 seconds is decelerating, and Drv_st_pre is determined as 2.

[0100] S207, based on the current driving state and the predicted driving state, determine the target driving state.

[0101] In this step, the predicted driving state is corrected according to the current driving state, and the corrected predicted driving state is taken as the target driving state. The target driving state includes: the engine is off or in the starting process, and the driving state value can be represented as Drv_st=0; idle driving, and the driving state value can be represented as Drv_st=1; acceleration driving, and the driving state value can be represented as Drv_st=2; constant speed driving, and the driving state value can be represented as Drv_st=3; deceleration driving, and the driving state value can be represented as Drv_st=4.

[0102] Further, the target driving state is determined based on the current driving state and the predicted driving state, including:

[0103] a3) If the current driving state is that the vehicle engine is off or in the starting process, the target driving state is that the vehicle engine is off or in the starting process.

[0104] Specifically, when Drv_state=0, Drv_st=0.

[0105] b3) If the current driving state is that the vehicle is in an idle state, the target driving state is that the vehicle is in an idle state.

[0106] Specifically, when Drv_state=1, Drv_st=1.

[0107] c3) If the current driving state is an acceleration state or a constant speed state or a deceleration state, the target driving state is determined in combination with the predicted driving state.

[0108] Specifically, when the current driving state Drv_state=2 or Drv_state=3 or Drv_state=4, the predicted driving state Drv_st_pre signal needs to be comprehensively judged, and the judgment rule is: if the current driving state Drv_state and the predicted driving state Drv_st_pre cannot be consistent for 5 seconds, the target driving state Drv_st=the current driving state Drv_state; if Drv_state and Drv_st_pre can be consistent for 5 seconds, the next step is entered for judgment.

[0109] Preferably, if the current driving state is an acceleration state or a constant speed state or a deceleration state, the target driving state is determined in combination with the predicted driving state, including:

[0110] c31) determining whether the current driving state is consistent with the predicted driving state within a first set time.

[0111] The first set time can be determined according to actual experience value. For example, assuming that the first set time is set to 5 seconds, it is determined whether the current driving state is consistent with the predicted driving state within 5 seconds, that is, whether the values of Drv_stat and Drv_st_pre are the same. That is, it is determined whether the predicted driving state is accurate according to the actual current driving state.

[0112] c32) If the consistency is not maintained, the current driving state is determined as the target driving state.

[0113] Specifically, if the current driving state and the predicted driving state cannot be consistent within the first set time, it indicates that the predicted driving state is inaccurate, and the current driving state is determined as the target driving state. For example, if the current driving state value Drv_state and the predicted driving state value Drv_st_pre cannot be consistent within 5 seconds, the target driving state value Drv_st is equal to the current driving state value Drv_stat.

[0114] c33) If the consistency is maintained, the target driving state is determined based on the change of the current driving state and the target driving state within the next first set time.

[0115] Specifically, if the current driving state and the predicted driving state can be consistent within the first set time, the next step is continued to determine. When the next step is,

[0116] For example, if Drv_stat and Drv_st_pre can be consistent within 5 seconds, the next step is entered. If the current driving state value Drv_stat changes first, the target driving state value Drv_st is equal to the current driving state value Drv_stat; if the predicted driving state value Drv_st_pre changes first, the target driving state value Drv_st is equal to the current driving state value Drv_stat for 0.5 seconds, and if the predicted driving state value Drv_st_pre remains unchanged within the 0.5 seconds, the target driving state value Drv_st is equal to the predicted driving state value Drv_st_pre for 1.5 seconds; then it is determined whether there is an equal moment within the next 0.5 seconds and the equal moment accounts for 85% of the total time. If yes, Drv_st is equal to Drv_st_pre. Otherwise, Drv_st is equal to Drv_stat.

[0117] S208, determining the generator control mode based on the processed parameter state value and / or the target driving state.

[0118] The generator voltage control mode includes: start mode, emergency stop mode, fast charging mode, deceleration charging mode, accelerator pedal mode, battery discharge mode and battery SOC mode.

[0119] Specifically, when the ignition switch state value IG_state = 1 and the target driving state value Drv_st = 0 or within five seconds of changing to 1, the start mode is adopted; when the ignition switch IG_state = 1 and the active brake state value Act_brake = 1, the emergency stop mode is adopted; when the battery state value SOC_state = 0, the fast charging mode is adopted;

[0120] When Drv_st = 4 and the battery state value SOC_state = 1 or 2, the deceleration charging mode is adopted;

[0121] When Acc_state = 1 and the ignition switch IG_state = 1, the accelerator pedal mode is adopted, and the voltage cannot be too low when the oil pump is at high speed;

[0122] When Drv_st = 2 and the SOC state value SOC_state = 1 or 2, or when Drv_st = 4 and SOC_state = 3, the battery discharge mode is adopted;

[0123] When Drv_st = 3, if SOC_state = 0, if SOC_state = 1, if SOC_state = 2, if SOC_state = 3, the battery SOC mode is adopted, and the four states correspond to different target voltages.

[0124] S209, according to the generator control mode, combined with the pre-stored mode voltage relationship table, determine the target voltage associated with the generator control mode.

[0125] Wherein, the mode voltage relationship table stores the relationship between the generator control mode and the target voltage corresponding to the mode. It can be understood that when the generator control mode is determined, the target voltage associated with the generator mode can be determined by querying the pre-stored mode voltage relationship table. Wherein, the starting mode (when IG_state = 1 and Drv_st = 0 or within five seconds of suddenly changing to 1), the target voltage uGen = 10.6V; the emergency stop mode (when IG_state = 1 and Act_brake = 1), the target voltage uGen = 15.5V; the fast charging mode (when SOC_state = 0), the target voltage uGen = 15.5V; the deceleration charging mode (when Drv_st = 4 and SOC_state = 1 or 2), the target voltage uGen = 15.5V; the accelerator pedal mode (when Acc_state = 1 and IG_state = 1), the target voltage uGen = 14V, the voltage cannot be too low when the oil pump is at high speed; the battery discharging mode (when Drv_st = 2 and SOC_state = 1 or 2, or when Drv_st = 4 and SOC_state = 3), the target voltage uGen = 12V; the battery SOC-based mode: when Drv_st = 3, if SOC_state = 0, the target voltage uGen = 15.5V. If SOC_state = 1, the target voltage uGen = 14.5V. If SOC_state = 2, the target voltage uGen = 13V. If SOC_state = 3, the target voltage uGen = 12V.

[0126] S210, according to the target voltage, the voltage of the generator is adjusted.

[0127] The embodiment refines the steps of determining the current driving state of the vehicle according to the vehicle operating parameters of the vehicle where the generator is located, determining the predicted driving state of the vehicle according to the video data collected by the vehicle, and determining the target voltage of the generator based on the current driving state and the predicted driving state based on the above-mentioned embodiment one. By using this method, signals that can be used to predict the driving state of the vehicle after 2 seconds to a certain extent are introduced. Using the deep neural network, the predicted driving state of the vehicle after 2 seconds is obtained by preprocessing and calculating these signals, and the current vehicle driving state is corrected, and then the corrected driving state is used in combination with the vehicle battery SOC signal and other fault signals to determine the target value of the generator output voltage. The generator improves the electric energy value of the useless kinetic energy recovery during the deceleration process of the vehicle, thereby reducing fuel consumption; the braking ability during the emergency deceleration process of the vehicle is improved; the response ability during the acceleration process of the vehicle is improved.

[0128] Embodiment three

[0129] Figure 3 This is a schematic diagram of a generator voltage control device provided in Embodiment 3 of the present invention. It is applicable to situations requiring precise control of generator voltage. The device can be implemented in hardware and / or software and is generally integrated into the generator voltage control system. For example... Figure 3 As shown, the device includes: a current state determination module 31, a predicted state determination module 32, a target voltage determination module 33, and an adjustment module 34, wherein...

[0130] The current status determination module 31 is used to determine the current driving status of the vehicle based on the vehicle operating parameters of the vehicle where the generator is located.

[0131] The predicted state determination module 32 is used to determine the predicted driving state of the vehicle based on the video data collected by the vehicle.

[0132] The target voltage determination module 33 is used to determine the target voltage of the generator based on the current driving state and the predicted driving state;

[0133] The regulating module 34 is used to regulate the voltage of the generator according to the target voltage.

[0134] Optionally, the current state determination module 31 includes:

[0135] The parameter acquisition unit is used to collect the operating parameters of various vehicle types in the vehicle where the generator is located.

[0136] The parameter status value determination unit is used to process the vehicle operating parameters according to the corresponding preset processing rules for each set type of vehicle operating parameter, and obtain the processed parameter status value.

[0137] The current driving state determination unit is used to determine the current driving state of the vehicle based on the status values ​​of various parameters and in combination with preset judgment rules. The current driving state includes the vehicle engine being off or in the process of starting, the vehicle being in an idling state, the vehicle being in an acceleration state, the vehicle being in a constant speed state, and the vehicle being in a deceleration state.

[0138] Furthermore, the parameter state value determination unit is specifically used for:

[0139] For the power mode signal in the vehicle operating parameters, the power mode signal is Booleanized to determine the vehicle ignition switch status value.

[0140] For the battery state of charge signal in the vehicle operating parameters, the battery state of charge is processed in a relay manner to determine the battery state value.

[0141] For the vehicle speed signal in the vehicle operating parameters, the vehicle speed signal is processed in a relay manner to determine the vehicle status value;

[0142] For the vehicle speed and engine speed signals in the vehicle operating parameters, the engine speed signal is processed in a relay manner to determine the vehicle status value.

[0143] For the vehicle acceleration signal in the vehicle operating parameters, the vehicle acceleration signal is processed in a relay manner to determine the vehicle state value;

[0144] For the throttle position signal in the vehicle operating parameters, the throttle position signal is processed in a relay manner to determine the throttle status value;

[0145] For the brake pedal position signal in the vehicle operating parameters, the brake pedal position signal is processed in a relay mode to determine the braking status value.

[0146] Determine the status value of the active braking status signal in the vehicle operating parameters.

[0147] Optionally, the predicted state determination module 32 includes:

[0148] The target image determination unit is used to input the video data collected by the vehicle into a pre-trained deep convolutional network model to obtain the target image;

[0149] The matrix data determination unit is used to process the target image according to the set image processing rules to obtain the target matrix data corresponding to the target image;

[0150] The driving state prediction unit is used to input the target matrix data into the deep neural network model to determine the predicted driving state of the vehicle. The deep neural network model includes a convolutional module and a recurrent module. The predicted driving state includes the vehicle being in an acceleration state, the vehicle being in a constant speed state, and the vehicle being in a deceleration state.

[0151] The driving state prediction unit is specifically used for:

[0152] The target matrix data is input into the convolution module to obtain the output result. The convolution module consists of a set number of deep convolutional neural networks.

[0153] The output is input into the loop module to obtain the probability that the vehicle is in the state of accelerating, constant speed and decelerating respectively. The loop module consists of a set of deep recurrent neural networks of a set length and two fully connected neural networks.

[0154] The driving state corresponding to the maximum probability among all driving states is determined as the predicted driving state of the vehicle.

[0155] Optionally, the target voltage determination module 33 includes:

[0156] The target driving state determination unit is used to determine the target driving state based on the current driving state and the predicted driving state.

[0157] The control mode determination unit is used to determine the generator control mode based on the processed parameter status values ​​and / or the target driving state.

[0158] The target voltage determination unit is used to determine the target voltage associated with the generator control mode based on the generator control mode and a pre-stored mode voltage relationship table.

[0159] Furthermore, the target driving state determination unit is specifically used for:

[0160] If the current driving status is that the vehicle engine is off or in the process of starting, then the target driving status is that the vehicle engine is off or in the process of starting.

[0161] If the current driving state is that the vehicle is idling, then the target driving state is that the vehicle is idling.

[0162] If the current driving state is accelerating, moving at a constant speed, or decelerating, then the target driving state is determined by combining the predicted driving state.

[0163] Furthermore, the target driving state determination unit uses the predicted driving state to determine the target driving state in the following steps:

[0164] Determine whether the current driving state is consistent with the predicted driving state within a first set time period;

[0165] If they are not consistent, then the current driving state is determined to be the target driving state;

[0166] If they remain consistent, the target driving state will be determined based on the changes in the current driving state and the target driving state within the next first set time period.

[0167] The generator voltage control device provided in the embodiments of the present invention can execute the generator voltage control method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0168] Example 4

[0169] Figure 4 This is a structural block diagram of a generator voltage control system provided in Embodiment 4 of the present invention, as shown below. Figure 4 As shown, the generator voltage control system includes a processor 41, a memory 42, an input device 43, and an output device 44; the number of processors 41 in the generator voltage control system can be one or more. Figure 4Taking a processor 41 as an example; the processor 41, memory 42, input device 43, and output device 44 in the generator voltage control system can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.

[0170] The memory 42, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the modules corresponding to the generator voltage control method in this embodiment of the invention (e.g., the current state determination module 31, the predicted state determination module 32, the target voltage determination module 33, and the adjustment module 34 in the generator voltage control device). The processor 41 executes various functional applications and data processing of the generator voltage control system by running the software programs, instructions, and modules stored in the memory 42, thereby realizing the aforementioned generator voltage control method.

[0171] The memory 42 may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function; the data storage area may store data created based on terminal usage. Furthermore, the memory 42 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory, or other non-volatile solid-state storage device. In some instances, the memory 42 may further include memory remotely located relative to the processor 41, which can be connected to the generator voltage control system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0172] Input device 43 can be used to receive input digital or character information, and to generate key signal inputs related to user settings and function control of the generator voltage control system. Output device 44 may include display devices such as a display screen.

[0173] Example 5

[0174] Embodiment 5 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a generator voltage control method, the method comprising:

[0175] The current driving status of the vehicle is determined based on the vehicle operating parameters of the vehicle where the generator is located.

[0176] Based on the video data collected from the vehicle, the predicted driving status of the vehicle is determined;

[0177] Based on the current driving state and the predicted driving state, the target voltage of the generator is determined;

[0178] The generator is regulated according to the target voltage.

[0179] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0180] It is worth noting that in the embodiments of the generator voltage control device described above, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0181] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method of voltage control of a generator, characterized by, The method comprises the following steps: determining a current driving state of the vehicle according to vehicle operating parameters of the vehicle in which the generator is located; determining a predicted driving state of the vehicle according to video data collected by the vehicle, the predicted driving state being a driving state of the vehicle at a predicted lag time, the lag time being a time during which an actual adjustment of the engine voltage lags behind an ideal adjustment; determining a target voltage of the generator based on the current driving state and the predicted driving state; adjusting the voltage of the generator according to the target voltage; wherein the step of determining the target voltage of the generator based on the current driving state and the predicted driving state comprises: determining a target driving state based on the current driving state and the predicted driving state; determining a generator control mode based on the processed parameter state value and / or the target driving state; determining a target voltage associated with the generator control mode according to the generator control mode and a pre-stored mode-voltage relationship table, the generator control mode including: a start-up mode, an emergency stop mode, a fast charging mode, a deceleration charging mode, a pedal depression mode, a battery discharging mode, and a battery SOC-based mode; wherein the step of determining the target driving state based on the current driving state and the predicted driving state comprises: if the current driving state is an acceleration state or a constant speed state or a deceleration state, determining the target driving state in combination with the predicted driving state; wherein the step of determining the target driving state in combination with the predicted driving state if the current driving state is an acceleration state or a constant speed state or a deceleration state comprises: determining whether the current driving state is consistent with the predicted driving state within a first set time; if not consistent, determining the current driving state as the target driving state; if consistent, continuing to determine the target driving state based on changes in the current driving state and the target driving state within a next first set time; wherein the step of continuing to determine the target driving state based on changes in the current driving state and the target driving state within a next first set time comprises: if the current driving state value changes first, taking the current driving state value as the target driving state value; if the predicted driving state value changes first, maintaining the current driving state value as the target driving state value for 0.5 seconds, if the predicted driving state value remains unchanged within 0.5 seconds, maintaining the predicted driving state value as the target driving state value for 1.5 seconds; then determining whether there is an equal moment within the next 0.5 seconds and whether the equal moment accounts for 85% of the time, if so, maintaining the predicted driving state value as the target driving state value, otherwise, taking the current driving state value as the target driving state value.

2. The method of claim 1, wherein, The step of determining the current driving state of the vehicle according to vehicle operating parameters of the vehicle in which the generator is located comprises: collecting vehicle operating parameters of each set type in the vehicle in which the generator is located; According to a preset processing rule corresponding to each set type of vehicle operation parameter, the vehicle operation parameter is processed to obtain a processed parameter state value; According to the parameter state values, a preset determination rule is combined to determine a current driving state of the vehicle, wherein the current driving state includes a vehicle engine off or in a starting process, a vehicle in an idle state, a vehicle in an acceleration state, a vehicle in a uniform speed state, and a vehicle in a deceleration state.

3. The method of claim 2, wherein, According to a preset processing rule corresponding to each set type of vehicle operation parameter, the vehicle operation parameter is processed to obtain a processed parameter state value, including: For the power mode signal in the vehicle operation parameter, the power mode signal is processed by Boolean to determine a vehicle ignition switch state value; For the battery state of charge signal in the vehicle operation parameter, the battery state of charge is processed in a relay mode to determine a battery state value; For the vehicle speed signal in the vehicle operation parameter, the vehicle speed signal is processed in a relay mode to determine a vehicle speed state value; For the engine speed signal in the vehicle operation parameter, the engine speed signal is processed in a relay mode to determine an engine speed state value; For the vehicle acceleration signal in the vehicle operation parameter, the vehicle acceleration signal is processed in a relay mode to determine a vehicle acceleration state value; For the throttle position signal in the vehicle operation parameter, the throttle position signal is processed in a relay mode to determine a throttle state value; For the brake pedal position signal in the vehicle operation parameter, the brake pedal position signal is processed in a relay mode to determine a brake state value; The active brake state value in the vehicle operation parameter is determined.

4. The method of claim 1, wherein, According to the video data collected by the vehicle, the predicted driving state of the vehicle is determined, including: The video data collected by the vehicle is input into a pre-trained deep convolutional network model to obtain a target image; According to a set image processing rule, the target image is processed to obtain target matrix data corresponding to the target image; The target matrix data is input into a deep neural network model to determine the predicted driving state of the vehicle, wherein the deep neural network model includes a convolution module and a recurrent module, and the predicted driving state includes a vehicle in an acceleration state, a vehicle in a uniform speed state, and a vehicle in a deceleration state.

5. The method of claim 4, wherein, The target matrix data is input into the deep neural network model to determine the predicted driving state of the vehicle, including: The target matrix data is input into the convolution module to obtain an output result, wherein the convolution module is composed of a set number of deep convolutional neural networks; The output result is input into the recurrent module to obtain the probability of the vehicle in an acceleration driving state, a uniform speed driving state, and a deceleration driving state, wherein the recurrent module is composed of a group of deep recurrent neural networks with a set length and two layers of fully connected neural networks; The driving state corresponding to the maximum value of the probability of the vehicle in each driving state is determined as the predicted driving state of the vehicle.

6. The method of claim 1, wherein, The target driving state is determined based on the current driving state and the predicted driving state, and the target driving state comprises: If the current driving state is that the vehicle engine is off or in the starting process, the target driving state is that the vehicle engine is off or in the starting process; If the current driving state is that the vehicle is in the idle state, the target driving state is that the vehicle is in the idle state.

7. A generator power generation control device characterized by comprising: Comprise: A current state determination module configured to determine a current driving state of a vehicle according to a vehicle operating parameter of the vehicle in which a generator is located; A predicted state determination module configured to determine a predicted driving state of the vehicle according to video data collected by the vehicle, the predicted driving state being a driving state of the vehicle at a predicted lag time, the lag time being a time during which an actual adjustment of an engine voltage lags behind an ideal situation; A target voltage determination module configured to determine a target voltage of the generator based on the current driving state and the predicted driving state; An adjustment module configured to adjust the voltage of the generator according to the target voltage; A target driving state determination unit configured to determine a target driving state based on the current driving state and the predicted driving state; A control mode determination unit configured to determine a generator control mode based on the processed parameter state value and / or the target driving state; A target voltage determination unit configured to determine a target voltage associated with the generator control mode according to the generator control mode and in combination with a pre-stored mode voltage relationship table; The target driving state determination unit is specifically configured to: If the current driving state is an acceleration state or a constant speed state or a deceleration state, determine the target driving state in combination with the predicted driving state; The target driving state determination unit is configured to determine the target driving state in combination with the predicted driving state, and the step comprises: determining whether the current driving state is consistent with the predicted driving state within a first set time; If not, determining that the current driving state is the target driving state; If consistent, continuing to determine the target driving state based on changes in the current driving state and the target driving state within a next first set time; The step of continuing to determine the target driving state based on changes in the current driving state and the target driving state within a next first set time comprises: If the current driving state value changes first, the current driving state value is taken as the target driving state value; If the predicted driving state value changes first, the current driving state value is taken as the target driving state value for 0.5 seconds, if the predicted driving state value remains unchanged within 0.5 seconds, the predicted driving state value is taken as the target driving state value for 1.5 seconds; then determining whether there is an equal moment within the next 0.5 seconds and the equal moment accounts for 85% of the total time, if yes, maintaining the predicted driving state value as the target driving state value, otherwise, taking the current driving state value as the target driving state value.

8. A generator voltage control system characterized by, Comprise: At least one processor; And A memory in communication connection with the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the generator voltage control method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing a processor to implement the generator voltage control method of any one of claims 1-6 when executed.

Images