Electric vehicle composite energy supply system based on solar energy and kinetic energy recovery and control method

Abstract

The invention relates to an electric vehicle composite energy supply system based on solar energy and kinetic energy recovery and a control method. The system comprises a comprehensive data fusion module used for fusing a vehicle motion state, battery and photovoltaic electrical parameters and environmental data to generate a comprehensive data set; the regenerative braking power withdrawing pre-judgment module is used for carrying out trend prediction according to the data set to obtain a pre-judgment signal containing a load sudden change mark and a change amplitude; the reference voltage prediction and adjustment module is used for outputting a corrected voltage reference instruction according to the pre-judgment signal, the photovoltaic parameter and the voltage reference value; the tracking mode setting module is used for determining a control mode state according to the correction instruction and the pre-judgment signal; and the photovoltaic control instruction dynamic calculation module is used for calculating and outputting a final photovoltaic control instruction according to the control mode state, the bus voltage ripple and the photovoltaic power trend. By adopting the system, photovoltaic output can be adjusted in advance before sudden change of a regenerative braking working condition through prospective cooperative control, so that voltage fluctuation of a direct-current bus is effectively stabilized, and the stability performance and the comprehensive energy efficiency of the system are improved.

Description

Technical Field

[0001] This invention belongs to the field of energy utilization technology, and in particular relates to a hybrid energy supply system and control method for electric vehicles based on solar energy and kinetic energy recovery. Background Technology

[0002] With the deep integration of new energy vehicles and renewable energy technologies, the "photovoltaic + electric vehicle" model, which combines solar power generation with electric vehicles, has become an important research direction for improving the sustainability of transportation energy. This type of system aims to create a composite energy supply system through onboard photovoltaic power generation and vehicle kinetic energy recovery, thereby reducing dependence on the external power grid and improving energy efficiency.

[0003] In this technological field, traditional or existing typical solutions usually employ a strategy of independent control. On the one hand, for solar photovoltaic subsystems, maximum power point tracking (MPPT) control algorithms, such as improved sliding mode control, are commonly used to capture as much electrical energy as possible from the photovoltaic array. On the other hand, for vehicle regenerative braking subsystems, prediction and torque distribution control are mainly based on vehicle conditions (such as pedal opening and vehicle speed) to recover kinetic energy during braking. The electrical energy generated by these two subsystems is ultimately collected on the vehicle's DC bus to charge the power battery or power the load. To maintain stable bus voltage, when there are sudden changes in load or power supply, energy storage units (such as batteries) are usually required for rapid power compensation and voltage support.

[0004] However, the current discrete control method presents a significant systemic problem when handling the coordinated power supply of two transient energy sources: solar power generation and regenerative braking. Because the maximum power point tracking control of photovoltaic power generation and the power recovery control of regenerative braking have different objectives and lack forward coordination, they are prone to causing drastic fluctuations in the DC bus voltage during sudden changes in load conditions. This threatens system stability and limits overall energy efficiency improvement. Specifically, the regenerative braking process, especially at its start or end, causes a step change in load power. Simultaneously, photovoltaic power generation is affected by variations in sunlight, and its output itself is volatile. When these two dynamic processes are superimposed on the DC bus, traditional responsive control struggles to balance power smoothly and promptly, leading to voltage drops or spikes on the bus. This can not only trigger protection circuits, affecting normal vehicle operation, but also reduce the overall efficiency of photovoltaic power generation and energy recovery due to frequent voltage fluctuations. Summary of the Invention

[0005] Therefore, it is necessary to provide a hybrid energy supply system and control method for electric vehicles based on solar energy and kinetic energy recovery, which can proactively adjust the photovoltaic reference voltage based on the regenerative braking power withdrawal trend to suppress DC bus voltage fluctuations.

[0006] In a first aspect, this application provides a hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery, comprising:

[0007] The integrated data fusion module is used to fuse data based on vehicle motion status data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters to obtain a comprehensive dataset containing multi-dimensional status information;

[0008] The regenerative braking power withdrawal prediction module is used to perform withdrawal trend prediction calculations based on the pedal opening, vehicle speed, motor torque, and vehicle auxiliary load power in the comprehensive data set, and obtain the regenerative braking power withdrawal prediction signal; the regenerative braking power withdrawal prediction signal includes a load change prediction flag and a predicted load change magnitude;

[0009] The reference voltage prediction and adjustment module is used to perform reference voltage prediction and adjustment based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, voltage reference value of high voltage DC bus and predicted load change amplitude, so as to obtain the corrected voltage reference command.

[0010] The tracking mode setting module is used to set the tracking mode according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal, and obtain the control mode status.

[0011] The photovoltaic control command dynamic calculation module is used to dynamically calculate control commands based on the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend, so as to obtain the final photovoltaic control commands.

[0012] In one embodiment, the regenerative braking power withdrawal prediction module includes:

[0013] The braking feature value calculation submodule is used to calculate feature values ​​based on the brake pedal opening and vehicle speed in the comprehensive dataset, and obtain intermediate feature value variables.

[0014] The load mutation prediction flag generation submodule is used to obtain the load mutation prediction flag by comparing the intermediate variable of the feature value with a preset negative threshold.

[0015] The predicted load change calculation submodule is used to calculate the power difference based on the motor torque, motor speed and vehicle auxiliary load power in the comprehensive dataset, and obtain the predicted load change range.

[0016] The data packaging submodule is used to package data based on the load change prediction flag and the predicted load change magnitude to obtain the regenerative braking power withdrawal prediction signal.

[0017] In one embodiment, the braking characteristic value calculation submodule includes:

[0018] The brake pedal change rate calculation unit is used to perform differential calculations based on the brake pedal opening to obtain the brake pedal opening change rate.

[0019] The normalized vehicle speed factor calculation unit is used to perform normalization calculations based on vehicle speed and the vehicle's maximum design speed to obtain the normalized vehicle speed factor.

[0020] The eigenvalue intermediate variable synthesis unit is used to perform a weighted summation calculation based on the brake pedal opening change rate and the normalized vehicle speed factor to obtain the eigenvalue intermediate variable; the eigenvalue intermediate variable is calculated using the following formula:

[0021]

[0022] in, As an intermediate variable for eigenvalues, As the first weighting coefficient, The rate of change of brake pedal opening. This is the second weighting coefficient. For vehicle speed, This indicates the vehicle's maximum design speed.

[0023] In one embodiment, the load change magnitude calculation submodule includes:

[0024] The current regenerative braking power calculation unit is used to calculate the current regenerative braking power by multiplying the motor torque and motor speed with the inverter efficiency.

[0025] The auxiliary power reference value retrieval unit is used to retrieve data based on the vehicle's auxiliary load power to obtain the auxiliary power reference value.

[0026] The load change difference unit is used to calculate the difference between the current regenerative braking power and the auxiliary power reference value to obtain the predicted load change range.

[0027] In one embodiment, the reference voltage prediction adjustment module includes:

[0028] The control state update submodule is used to perform control state update processing based on the regenerative braking power withdrawal prediction signal to obtain updated control state data; the updated control state data is used to pause the standard maximum power point tracking logic.

[0029] The basic maximum power point voltage query submodule is used to query a preset photovoltaic characteristic curve table based on the updated control status data and the ambient light irradiance and photovoltaic panel backsheet temperature in the photovoltaic array electrical parameters to obtain the basic maximum power point voltage.

[0030] The voltage compensation and limiting submodule is used to perform voltage compensation and limiting calculations based on the base maximum power point voltage, the voltage reference value of the high-voltage DC bus, and the predicted load change amplitude, so as to obtain the corrected voltage reference command.

[0031] In one embodiment, the voltage compensation limiting submodule includes:

[0032] The reference offset calculation unit is used to calculate the voltage difference based on the base maximum power point voltage and the voltage reference value of the high voltage DC bus to obtain the reference offset.

[0033] The voltage compensation proportional calculation unit is used to perform proportional calculations based on the predicted load change amplitude and the preset voltage regulation coefficient to obtain the voltage compensation amount.

[0034] The reference voltage superposition unit is used to calculate the corrected reference voltage by superimposing the voltage compensation amount and the reference offset.

[0035] The safety limiting unit is used to perform safety limiting processing based on the corrected reference voltage to obtain the corrected voltage reference command.

[0036] In one embodiment, the voltage compensation ratio calculation unit includes:

[0037] The load change amplitude extraction subunit is used to extract data based on the predicted load change amplitude to obtain the load change amplitude.

[0038] The regulation coefficient configuration subunit is used to configure parameters according to the voltage regulation coefficient to obtain the regulation coefficient value;

[0039] The voltage compensation product subunit is used to calculate the voltage compensation amount by multiplying the load change amplitude and the adjustment coefficient value. The voltage compensation amount is calculated using the following formula:

[0040]

[0041] in, This is the voltage compensation amount. To adjust the coefficient value, This represents the magnitude of the load change.

[0042] In one embodiment, the photovoltaic control command dynamic calculation module includes:

[0043] The ripple convergence flag generation submodule is used to compare the DC bus voltage ripple data with a preset ripple threshold to obtain the ripple convergence flag.

[0044] The mode recovery flag calculation submodule is used to perform AND logic operations based on the ripple convergence flag and the photovoltaic output power change trend to obtain the mode recovery flag.

[0045] The final control command generation submodule is used to calculate control parameters based on the mode recovery flag, control mode status, and corrected voltage reference command to obtain the final photovoltaic control command.

[0046] Secondly, this application also provides a control method for a hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery, including:

[0047] Based on vehicle motion state data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters, data fusion is performed to obtain a comprehensive dataset containing multi-dimensional state information;

[0048] Based on the pedal opening, vehicle speed, motor torque, and vehicle auxiliary load power in the comprehensive dataset, a retraction trend prediction calculation is performed to obtain the regenerative braking power retraction prediction signal; the regenerative braking power retraction prediction signal includes a load change prediction flag and a predicted load change magnitude;

[0049] Based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, high voltage DC bus voltage reference value and predicted load change amplitude, the reference voltage prediction adjustment is performed to obtain the corrected voltage reference command;

[0050] The tracking mode is set according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal to obtain the control mode status.

[0051] Based on the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend, the control commands are dynamically calculated to obtain the final photovoltaic control commands.

[0052] The aforementioned hybrid energy supply system and control method for electric vehicles based on solar and kinetic energy recovery integrates vehicle motion state, battery and photovoltaic electrical parameters, and environmental data into a multi-dimensional state dataset through a comprehensive data fusion module. A regenerative braking power withdrawal prediction module performs trend prediction calculations based on key parameters such as pedal opening and vehicle speed in the dataset, generating a withdrawal prediction signal that includes load change prediction flags and the magnitude of the change. Subsequently, a reference voltage prediction and adjustment module performs forward-looking correction of the voltage command based on this prediction signal, photovoltaic parameters, and bus voltage reference values. A tracking mode setting module switches and determines the system's control mode state based on the corrected voltage command and prediction signal. Finally, a photovoltaic control command dynamic calculation module dynamically calculates the optimal photovoltaic control command by combining the control mode state, real-time bus voltage ripple, and photovoltaic power change trends. This forms a closed-loop collaborative control process from state perception, trend prediction, command pre-adjustment to mode adaptation. This allows the photovoltaic subsystem to adjust its output smoothly and in advance when a sudden change in regenerative braking load is imminent, effectively suppressing drastic fluctuations in DC bus voltage and improving the overall operational stability and dynamic energy management efficiency of the hybrid energy supply system. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A schematic diagram of a hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery provided by the present invention;

[0055] Figure 2 A schematic diagram of the structure of the regenerative braking power withdrawal prediction module in an optional embodiment of the present invention;

[0056] Figure 3 This is a flowchart illustrating a control method for a hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery, provided by the present invention. Detailed Implementation

[0057] 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 merely illustrative and not intended to limit the scope of this application.

[0058] In one embodiment, such as Figure 1As shown, a hybrid energy supply system for electric vehicles based on solar and kinetic energy recovery is provided. This embodiment illustrates the application of this system to a terminal. It is understood that this system can also be applied to a server, or to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. In this embodiment, the system includes the following module 10:

[0059] The integrated data fusion module 11 is used to fuse data based on vehicle motion state data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters to obtain a comprehensive dataset containing multi-dimensional state information.

[0060] Optionally, the core implementation logic of the integrated data fusion module involves time-series synchronization and feature fusion processing of multi-source data. First, it receives vehicle motion state data, power battery electrical parameters, photovoltaic array electrical parameters, and environmental parameters transmitted from onboard sensors and the acquisition unit. The vehicle motion state data includes pedal opening, vehicle speed, and motor torque; the power battery electrical parameters include remaining charge (State of Charge), SOC, terminal voltage, and charging / discharging current; the photovoltaic array electrical parameters include output voltage, output current, and output power; and the environmental parameters include light intensity and ambient temperature. Time-series synchronization of the multi-source data is achieved through timestamp alignment, eliminating time deviations caused by differences in the acquisition cycles of different parameters. Subsequently, the 3σ criterion is used to remove outliers from the synchronized data, filtering out abnormal data caused by sensor noise and transmission interference. Next, feature quantities of each parameter are extracted, such as the voltage change rate of the power battery terminal, the fluctuation range of photovoltaic output power, and the gradient change of light intensity. Then, the Kalman filter algorithm is used to fuse and calculate the multi-source heterogeneous data. This algorithm constructs a data fusion model through state equations and observation equations, and obtains the optimal estimation result based on prior estimation and iterative updates of observation values. Finally, it outputs a multi-dimensional state information comprehensive dataset containing all the original parameters and extracted features.

[0061] The regenerative braking power withdrawal prediction module 12 is used to perform withdrawal trend prediction calculation based on the pedal opening, vehicle speed, motor torque and vehicle auxiliary load power in the comprehensive data set, and obtain the regenerative braking power withdrawal prediction signal; the regenerative braking power withdrawal prediction signal includes a load change prediction flag and a predicted load change magnitude.

[0062] Optionally, the regenerative braking power withdrawal prediction module predicts the withdrawal trend through temporal feature analysis and a pre-trained prediction model. Based on a comprehensive dataset, four core parameters are selected: pedal opening, vehicle speed, motor torque, and vehicle auxiliary load power. First, a temporal sequence is constructed for these four parameters. Continuous data segments are extracted according to fixed time windows, and temporal features are extracted. Specifically, features such as the rate of change of pedal opening, the slope of vehicle speed decay, the amplitude of motor torque change, and the frequency of fluctuation in vehicle auxiliary load power are extracted. The extracted temporal features are input into a pre-trained Long Short-Term Memory (LSTM) model. This model solves the problem of long sequence dependencies through a gating unit structure and has been trained based on a large amount of historical regenerative braking operating data. The training process uses gradient descent to optimize model parameters and establish a mapping relationship between temporal features and power withdrawal trends. The long short-term memory network model is used to calculate the pullback trend and outputs a regenerative braking power pullback prediction signal that includes a load change prediction flag and a predicted load change magnitude. The load change prediction flag is a binary variable, where 0 indicates no change and 1 indicates a change. The predicted load change magnitude is calculated by the difference between the predicted load power output by the model and the current load power, thus enabling early detection of regenerative braking power pullback.

[0063] The reference voltage prediction and adjustment module 13 is used to perform reference voltage prediction and adjustment based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, voltage reference value of high voltage DC bus and predicted load change amplitude, so as to obtain the corrected voltage reference command.

[0064] Optionally, the reference voltage prediction and adjustment module corrects the voltage reference command based on the power balance principle and droop control strategy. First, it analyzes the regenerative braking power withdrawal prediction signal. If the load change prediction flag is 0, indicating no risk of load change, the initial voltage reference value of the high-voltage DC bus remains unchanged; if the flag is 1, the prediction and adjustment process is initiated. It prioritizes extracting the output voltage and output current from the photovoltaic array's electrical parameters to calculate the current maximum output power of the photovoltaic array. Combined with the predicted load change amplitude, it obtains the power balance difference, i.e., the difference between the predicted load change and the maximum output power of the photovoltaic array. A voltage adjustment calculation model is constructed based on the droop control principle. The core of this principle is to balance system power through small voltage changes. A predefined droop coefficient (a fixed constant determined based on the system impedance characteristics) is used to establish a linear relationship between the voltage adjustment and the power balance difference, i.e., voltage adjustment = droop coefficient × power balance difference. Simultaneously, it compensates for the voltage adjustment by considering the trend of light intensity changes in the photovoltaic array's electrical parameters. If the light intensity shows an increasing trend, the predicted photovoltaic output power will increase, and the voltage adjustment is appropriately reduced to avoid over-adjustment; if the light intensity decreases, the adjustment is appropriately increased. Finally, the initial voltage reference value is superimposed with the calculated voltage adjustment amount to obtain the corrected voltage reference command, thus realizing the forward-looking adjustment of the voltage reference value.

[0065] The tracking mode setting module 14 is used to set the tracking mode according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal, and obtain the control mode status.

[0066] Optionally, the tracking mode setting module achieves precise matching of control modes through threshold judgment and state mapping. First, it acquires the corrected voltage reference command and regenerative braking power withdrawal prediction signal, prioritizing the identification of load mutation prediction flags as the basis for mode division. If the load mutation prediction flag is 0, indicating no load mutation, it is directly set to the conventional Maximum Power Point Tracking (MPPT) mode, where the control priority is maximizing photovoltaic output power capture. If the load mutation prediction flag is 1, the predicted load change amplitude is further extracted and compared with a predefined load change amplitude threshold, which is determined based on the system voltage stability margin. If the predicted load change amplitude is greater than the predefined threshold, it indicates that the load mutation has a significant impact on voltage stability, and it is set to a voltage stability-priority tracking mode, where the control priority is maintaining DC bus voltage stability. If the predicted load change amplitude is less than or equal to the predefined threshold, it indicates that the load mutation has a smaller impact, and it is set to a cooperative tracking mode, balancing photovoltaic maximum power capture and voltage stability. Finally, the three modes are mapped to fixed status codes: 0 represents the regular MPPT mode, 1 represents the voltage stability priority mode, and 2 represents the cooperative tracking mode. The output is passed to the subsequent modules as the control mode status to achieve targeted matching of control strategies.

[0067] The photovoltaic control command dynamic calculation module 15 is used to dynamically calculate the control command based on the control mode status, DC bus voltage ripple data and photovoltaic output power change trend, so as to obtain the final photovoltaic control command.

[0068] Optionally, the photovoltaic control command dynamic calculation module dynamically generates control commands based on a mode-adaptive algorithm. First, it receives the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend. Then, it selects the corresponding core control algorithm based on the control mode status. In the conventional MPPT mode, the Perturb and Observe (P&O) method is used to calculate the control command. By slightly perturbing the switching duty cycle of the photovoltaic array converter, the change in photovoltaic output power after the perturbation is collected. If the power increases, the adjustment continues along the current perturbation direction; otherwise, the perturbation is reversed. This process iterates until the duty cycle command corresponding to the maximum power point is found. In the voltage stability priority mode, a Proportional-Integral-Derivative (PID) control algorithm is used. The corrected voltage reference command is used as the target value, and the DC bus voltage ripple data is used as feedback. Voltage ripple is eliminated by adjusting the parameters of the proportional, integral, and derivative components, thus maintaining bus voltage stability. In the cooperative tracking mode, the disturbance observation method and PID control logic are integrated. The trend of photovoltaic output power change is used as the basis for weight adjustment. The power change slope is obtained through linear fitting. When the absolute value of the slope is small, the weight of the disturbance observation method is increased to improve power capture efficiency. When the absolute value of the slope is large, the weight of PID control is increased to enhance voltage stability. At the same time, Fourier Transform (FT) analysis is performed on the DC bus voltage ripple data. If the ripple amplitude exceeds the predefined range, the PID regulation intensity is increased. Finally, the final photovoltaic control command used to adjust the switching devices of the photovoltaic array converter is calculated.

[0069] In the aforementioned hybrid energy supply system for electric vehicles based on solar and kinetic energy recovery, the integrated data fusion module provides comprehensive and accurate basic status data, the regenerative braking power withdrawal prediction module enables early detection of load surges, the reference voltage prediction and adjustment module performs forward-looking correction of the voltage target, the tracking mode setting module matches and adapts the control strategy, and the photovoltaic control command dynamic calculation module outputs precise control commands. This system allows for early coordination between photovoltaic control and regenerative braking power withdrawal, effectively avoiding drastic DC bus voltage fluctuations caused by the superposition of their dynamic processes, ensuring stable system operation, and simultaneously improving the overall efficiency of photovoltaic power generation and energy recovery.

[0070] In an optional embodiment, such as Figure 2 As shown, the regenerative braking power withdrawal prediction module 12 includes:

[0071] The braking feature value calculation submodule 121 is used to calculate feature values ​​based on the brake pedal opening and vehicle speed in the comprehensive dataset, and obtain intermediate feature value variables.

[0072] Optionally, the brake pedal opening and vehicle speed are preprocessed sequentially, and a moving average filtering algorithm is used to eliminate high-frequency noise in the sensor-acquired data. The sliding window length is adaptively matched based on the data acquisition cycle to ensure that the filtered data retains the true trend of change while removing interference signals. Subsequently, an eigenvalue fusion calculation model is constructed based on the principle of braking dynamics. The preprocessed brake pedal opening and vehicle speed are used as inputs, and the intermediate eigenvalue variables are obtained through linear weighted summation. The weight coefficients are determined by least squares fitting using a large amount of braking condition sample data. The fitting objective is to optimize the correlation between the intermediate eigenvalue variables and the actual regenerative braking power change trend. Specifically, the calculation logic of the intermediate eigenvalue variables is the difference between the weighted values ​​of the two, i.e., intermediate eigenvalue variable = brake pedal opening weight × brake pedal opening - vehicle speed weight × vehicle speed. This difference can accurately characterize the transition trend of braking state from strong to weak.

[0073] The load mutation prediction flag generation submodule 122 is used to obtain the load mutation prediction flag by comparing it with a preset negative threshold based on the intermediate variable of the feature value.

[0074] Optionally, the preset negative threshold is a critical characteristic value obtained based on historical regenerative braking power withdrawal statistics. Its determination process involves collecting intermediate characteristic variables corresponding to the power withdrawal start time at different vehicle speeds and braking intensities, and using statistical analysis to take the minimum value of this type of data as the preset negative threshold. This threshold is used to determine whether the regenerative braking power is about to enter the withdrawal phase. The intermediate characteristic variables are continuously received and monitored in real-time. A continuous sampling judgment logic is used. When the intermediate characteristic variables in multiple consecutive collection periods are all less than the preset negative threshold, a load mutation prediction flag of 1 is generated, indicating that a load mutation caused by regenerative braking power withdrawal is about to occur. If the intermediate characteristic variables are greater than or equal to the preset negative threshold or the continuous sampling condition is not met, a load mutation prediction flag of 0 is generated, indicating that there is no risk of load mutation. This continuous sampling judgment logic can effectively avoid misjudgments caused by fluctuations in a single data point and improve the reliability of the prediction flag.

[0075] The load change calculation submodule 123 is used to calculate the power difference based on the motor torque, motor speed and vehicle auxiliary load power in the comprehensive dataset to obtain the predicted load change range.

[0076] Optionally, the current motor output power is first calculated based on the motor dynamics formula: Motor output power = Motor torque × Motor speed ÷ Speed ​​coefficient, where the speed coefficient is a fixed conversion constant for converting motor speed units to angular velocity units. This formula is the standard theoretical formula for motor power calculation. Then, the current motor output power is superimposed with the vehicle's auxiliary load power to obtain the current total load power. Considering that the vehicle's auxiliary load power changes gradually over a short period, it can be considered a stable value and directly used in the calculation. Based on the physical characteristics of regenerative braking power withdrawal, the predicted load power is the superposition of the withdrawn and stabilized motor output power and the vehicle's auxiliary load power. The withdrawn and stabilized motor output power is calculated by extrapolating the withdrawal trend of the motor torque, i.e., based on the motor torque withdrawal curve constructed from historical data, combined with the current motor torque, the predicted withdrawn and stabilized torque is calculated, and then substituted into the motor output power formula to obtain the predicted motor output power. Finally, the difference between the current total load power and the predicted total load power is calculated to obtain the predicted load change magnitude. A positive difference indicates that the load power will decrease, and its absolute value is the magnitude of the change.

[0077] The data packaging submodule 124 is used to package data according to the load change prediction flag and the predicted load change magnitude to obtain the regenerative braking power withdrawal prediction signal.

[0078] Optionally, a structured data packet format is used to integrate the load mutation prediction flag and the predicted load change magnitude. First, a standardized data frame structure is defined, comprising a frame header, data field, check field, and frame trailer. The frame header and trailer are fixed identifier bytes used for subsequent frame synchronization identification at the receiving end. The data field is divided into two subfields, storing the load mutation prediction flag and the predicted load change magnitude respectively. The load mutation prediction flag uses 1-byte binary data storage, while the predicted load change magnitude uses floating-point data format, specifying its precision and range to ensure transmission accuracy. During packetization, data validity verification is performed synchronously by calculating the Cyclic Redundancy Check (CRC) code of the data field and storing it in the check field. This is used by the receiving end to verify whether errors occurred during data transmission. After packetization, a regenerative braking power withdrawal prediction signal conforming to the communication protocol is generated. This signal can be directly transmitted to the subsequent reference voltage prediction and adjustment module for parsing and use.

[0079] In the above embodiments, the braking feature value calculation submodule accurately extracts core features, the load mutation prediction flag generation submodule reliably identifies mutation risks, the load change amplitude prediction calculation submodule obtains quantitative data on the degree of mutation, and the data packaging submodule ensures standardized information transmission. This embodiment can accurately and timely generate regenerative braking power withdrawal prediction signals containing key information on load mutations, providing accurate forward-looking input for reference voltage adjustment and photovoltaic control mode matching, effectively supporting the coordinated regulation of photovoltaic control and regenerative braking, and reducing the risk of DC bus voltage fluctuations.

[0080] In one embodiment, the braking characteristic value calculation submodule includes:

[0081] The brake pedal change rate calculation unit is used to perform differential calculations based on the brake pedal opening to obtain the brake pedal opening change rate.

[0082] Optionally, a first-order backward difference algorithm is used to calculate the differential of the brake pedal opening. This algorithm is based on the principle of numerical differentiation of discrete-time series, and solves for the rate of change by the difference between the brake pedal opening data of two adjacent sampling periods. It has the advantages of low computational load and strong real-time performance, making it suitable for real-time vehicle control scenarios. First, brake pedal opening data for two consecutive sampling periods are read from the comprehensive dataset, and the current period data is denoted as... The data for the previous period was Simultaneously read the preset sampling period This sampling period is consistent with the sampling period for brake pedal opening and is determined by configuration parameters collected by onboard sensors. Then, it is calculated using the formula: The rate of change of brake pedal opening was calculated. The calculation process uses floating-point data format to ensure accuracy, and the calculation results are constrained to eliminate extreme values ​​that exceed the reasonable range of change due to sensor malfunction, so as to ensure that the output brake pedal opening change rate is true and reliable.

[0083] The normalized speed factor calculation unit is used to perform normalization calculations based on vehicle speed and the vehicle's maximum design speed to obtain the normalized speed factor.

[0084] Optionally, the normalization calculation is based on the principle of dimensional uniformity. By comparing the real-time vehicle speed with the vehicle's maximum design speed, the influence of different speed magnitudes on subsequent fusion calculations is eliminated, ensuring a unified comparison benchmark for the results. First, the vehicle's maximum design speed is read from a predefined vehicle parameter configuration library. This parameter is a fixed value determined during the vehicle design phase, stored in a non-volatile memory unit, and can be directly accessed. The real-time vehicle speed is then read from the comprehensive dataset. Before calculation, the real-time vehicle speed is... Validity verification is performed by determining whether the vehicle speed data falls within the range of 0 to the vehicle's maximum design speed. Within a reasonable range, outlier data is removed. After verification, the formula is used... Normalization calculations are performed to obtain the normalized vehicle speed factor. The factor ranges from 0 to 1 and can accurately represent the proportion of the current vehicle speed to the vehicle's maximum driving speed, intuitively reflecting the intensity of the vehicle's motion state.

[0085] The eigenvalue intermediate variable synthesis unit is used to perform a weighted summation calculation based on the brake pedal opening change rate and the normalized vehicle speed factor to obtain the eigenvalue intermediate variable; the eigenvalue intermediate variable is calculated using the following formula:

[0086]

[0087] in, As an intermediate variable for eigenvalues, As the first weighting coefficient, The rate of change of brake pedal opening. This is the second weighting coefficient. For vehicle speed, This indicates the vehicle's maximum design speed.

[0088] Optionally, the brake pedal opening change rate and the normalized vehicle speed factor can be fused using preset weighting coefficients, so that the output feature value intermediate variable can comprehensively reflect the correlation between braking intensity and vehicle motion state. First, the first weighting coefficient is read from the parameter configuration file. Second weighting coefficient The two sets of weighting coefficients were determined by fitting a large amount of braking condition sample data using the least squares method. The fitting process used eigenvalues ​​as intermediate variables. With the objective of maximizing the correlation with the actual regenerative braking power change trend, the optimal weighting coefficients are stored in the parameter configuration file after fitting. Subsequently, the brake pedal opening change rate output by the brake pedal change rate calculation unit is extracted. The normalized vehicle speed factor output by the normalized vehicle speed factor calculation unit The calculation involves a weighted summation based on the given formula for calculating intermediate eigenvalues. Double-precision floating-point data is used during the calculation to avoid accumulated errors, and the final output is the calculated intermediate eigenvalue. It is directly transmitted to the subsequent load mutation prediction flag generation submodule.

[0089] In the above embodiments, the brake pedal change rate calculation unit accurately captures the dynamic change trend of braking operation, the normalized vehicle speed factor calculation unit achieves dimensional unification of vehicle speed characteristics, and the feature value intermediate variable synthesis unit obtains core feature quantities that can comprehensively characterize the braking state and vehicle motion state through weighted fusion. The feature value intermediate variables output by this embodiment can accurately reflect the precursor characteristics of regenerative braking power withdrawal, improve the accuracy and timeliness of prediction, and thus provide reliable advance support for the coordinated control of photovoltaic and regenerative braking, effectively reducing the risk of DC bus voltage fluctuations.

[0090] In one embodiment, the load change magnitude calculation submodule includes:

[0091] The current regenerative braking power calculation unit is used to calculate the current regenerative braking power by multiplying the motor torque and motor speed with the inverter efficiency.

[0092] Optionally, the actual electrical power fed back to the DC bus is obtained through the coupled calculation of the motor's output mechanical quantity and electrical conversion efficiency. First, real-time motor torque and speed are read from the comprehensive dataset. Simultaneously, the inverter efficiency under the current operating condition is retrieved from a predefined inverter parameter configuration library. This inverter efficiency is a measured statistical value based on different motor speeds and torques, calibrated experimentally and stored in the configuration library, and can be accurately matched and retrieved according to the real-time motor status. Then, a standard power calculation model is used for product calculation. The formula is: Current regenerative braking power = Motor torque × Motor speed ÷ Speed ​​coefficient × Inverter efficiency, where the speed coefficient is a fixed conversion constant used to convert the motor speed unit to the angular velocity unit, ensuring dimensional consistency in the formula. During the calculation process, the validity of the input motor torque and motor speed is verified, and abnormal data exceeding the equipment's rated range is removed. A double-precision floating-point data format is used for the calculation to avoid truncation errors. Finally, the verified current regenerative braking power is output.

[0093] The auxiliary power reference value retrieval unit is used to retrieve data based on the auxiliary load power of the whole vehicle to obtain the auxiliary power reference value.

[0094] Optionally, by establishing a correspondence between the vehicle's auxiliary load power and a benchmark value, a stable reference benchmark is provided for subsequent difference calculations. First, an auxiliary power benchmark value mapping table is constructed. This table is based on statistical data of the vehicle's auxiliary load power under different operating conditions, dividing the vehicle's auxiliary load power into multiple continuous intervals. Each interval corresponds to a preset auxiliary power benchmark value, which is the steady-state average of the auxiliary load power within the corresponding interval. This benchmark value is determined through statistical fitting of a large amount of real-vehicle operating condition test data and stored in non-volatile storage. After reading the real-time vehicle auxiliary load power from the comprehensive dataset, the mapping table is retrieved using an interval matching algorithm to determine the interval range to which the real-time vehicle auxiliary load power belongs, and then the corresponding auxiliary power benchmark value is extracted. During the retrieval process, if the real-time power is at the interval boundary, the average value of the benchmark values ​​of adjacent intervals is automatically matched to ensure a smooth transition of the benchmark value.

[0095] The load change difference unit is used to calculate the difference between the current regenerative braking power and the auxiliary power reference value to obtain the predicted load change range.

[0096] Optionally, the load change magnitude caused by the regenerative braking power withdrawal can be characterized by the difference between the current regenerative braking power and the auxiliary power reference value. First, the current regenerative braking power output from the current regenerative braking power calculation unit and the auxiliary power reference value output from the auxiliary power reference value retrieval unit are read synchronously. A data synchronization verification mechanism is used to ensure that the two data are matched data from the same sampling time, avoiding calculation errors caused by timing deviations. Then, the difference is calculated. The calculation logic is: Predicted load change magnitude = Current regenerative braking power - Auxiliary power reference value. This difference directly reflects the load power change before and after the regenerative braking power withdrawal. When the difference is positive, it indicates that the load power will decrease with the regenerative braking withdrawal, and its absolute value is the magnitude of the load change. After the calculation is completed, the results are constrained to remove extreme values ​​that exceed the reasonable range of change due to data anomalies. The final predicted load change magnitude is output in floating-point data format.

[0097] In the above embodiment, the current regenerative braking power calculation unit accurately obtains the real-time energy recovery power during the braking process, the auxiliary power reference value retrieval unit provides a stable auxiliary load power reference, and the load change amplitude difference unit quantifies the degree of load change through difference calculation. This embodiment can accurately obtain the predicted load change amplitude reflecting the load change caused by the regenerative braking power withdrawal, effectively supporting the forward-looking coordination of photovoltaic control and regenerative braking, and reducing the risk of DC bus voltage fluctuations caused by the superposition of load changes and photovoltaic fluctuations.

[0098] In one embodiment, the reference voltage prediction adjustment module includes:

[0099] The control state update submodule is used to perform control state update processing based on the regenerative braking power withdrawal prediction signal to obtain updated control state data; the updated control state data is the pause standard maximum power point tracking logic.

[0100] Optionally, the switching of the corresponding control logic can be triggered by parsing the load mutation prediction flag in the regenerative braking power withdrawal prediction signal. First, the regenerative braking power withdrawal prediction signal is received and its frame parsed to extract the load mutation prediction flag. This parsing process follows a preset signal frame structure, ensuring signal integrity through frame header matching and verification fields to avoid erroneous state updates due to signal transmission errors. When the parsed load mutation prediction flag is 1, the control state update process is initiated, updating the current control state data to pause the standard maximum power point tracking logic. Simultaneously, the timestamp of the state update is recorded for subsequent submodule data timing synchronization. The updated control state data is stored in a structured format, including a state identifier, update timestamp, and a state validity flag. The state validity flag informs subsequent submodules that the current control state has been updated and is usable. If the load mutation prediction flag is 0, the original control state data remains unchanged, and no update operation is triggered.

[0101] The basic maximum power point voltage query submodule is used to query a preset photovoltaic characteristic curve table based on the updated control status data and the ambient light irradiance and photovoltaic panel backsheet temperature in the photovoltaic array electrical parameters to obtain the basic maximum power point voltage.

[0102] Optionally, the accurate base reference voltage can be obtained by looking up a table, utilizing the correspondence between the electrical characteristics of the photovoltaic array and environmental parameters. First, a pre-defined photovoltaic characteristic curve table is constructed. This table is a two-dimensional mapping table, with ambient light irradiance and photovoltaic panel backsheet temperature as input dimensions and the corresponding maximum power point voltage as the output value. The data source is the experimental calibration results of the photovoltaic array. By testing the output characteristics of the photovoltaic array under different light and temperature conditions, the maximum power point voltage under each condition is extracted and stored in a non-volatile memory unit after data smoothing. When updated control status data is received and the status validity flag is valid, the ambient light irradiance and photovoltaic panel backsheet temperature are extracted from the comprehensive dataset's electrical parameters of the photovoltaic array. The validity of these two parameters is verified, and abnormal data exceeding the reasonable measurement range is removed. Subsequently, a bilinear interpolation algorithm is used to query the photovoltaic characteristic curve table. If the extracted ambient irradiance and photovoltaic panel backsheet temperature happen to match the discrete data points in the photovoltaic characteristic curve table, the corresponding maximum power point voltage is directly extracted as the base maximum power point voltage. If they are between the data points, a smooth base maximum power point voltage is calculated through bilinear interpolation to ensure the continuity and accuracy of the query results.

[0103] The voltage compensation and limiting submodule is used to perform voltage compensation and limiting calculations based on the base maximum power point voltage, the voltage reference value of the high-voltage DC bus, and the predicted load change amplitude, so as to obtain the corrected voltage reference command.

[0104] Optionally, the base maximum power point voltage is dynamically adjusted based on the load change amplitude, while ensuring that the output voltage reference command remains within a safe range. First, the base maximum power point voltage, the voltage reference value of the high-voltage DC bus, and the predicted load change amplitude are read synchronously. Timing stamp comparison ensures that these three are matched data under the same operating condition, avoiding compensation errors caused by timing deviations. A voltage compensation calculation model is constructed based on the power balance principle. A corresponding compensation coefficient is matched according to the predicted load change amplitude. This compensation coefficient is a predefined piecewise coefficient. The optimal compensation coefficient under different load change amplitudes is calibrated experimentally and stored in the parameter configuration file. The larger the predicted load change amplitude, the larger the compensation coefficient value, to achieve more thorough voltage pre-adjustment. The voltage compensation calculation logic is: Compensated voltage = Base maximum power point voltage × Compensation coefficient × (Predicted load change amplitude / High-voltage DC bus voltage reference value). Then, the compensated voltage is superimposed with the high-voltage DC bus voltage reference value to obtain the initial corrected voltage. To ensure system safety, the initial correction voltage needs to be limited. The limiting range is the safe operating voltage range of the high-voltage DC bus. This range is determined by the rated parameters of the bus capacitor and power devices and is pre-stored in the system configuration library. If the initial correction voltage is within the safe range, it is directly output as the corrected voltage reference command. If it exceeds the safe range, it is clamped to the nearest safe boundary value, and finally the corrected voltage reference command after limiting is output.

[0105] In the above embodiments, the control state update submodule implements control logic switching based on load change prediction, the basic maximum power point voltage query submodule accurately obtains the basic reference voltage of the photovoltaic array, and the voltage compensation and limiting submodule dynamically adjusts and ensures voltage safety in conjunction with load changes. This embodiment can generate corrected voltage reference commands adapted to regenerative braking power withdrawal conditions in advance, enabling photovoltaic control to respond proactively to load changes, effectively balancing the superimposed effects of photovoltaic output fluctuations and load changes, reducing the amplitude of high-voltage DC bus voltage fluctuations, ensuring stable system operation, and improving overall energy efficiency.

[0106] In one embodiment, the voltage compensation limiting submodule includes:

[0107] The reference offset calculation unit is used to calculate the voltage difference based on the base maximum power point voltage and the voltage reference value of the high voltage DC bus to obtain the reference offset.

[0108] Optionally, the inherent deviation between the photovoltaic base reference voltage and the target bus voltage can be determined by the difference between the base maximum power point voltage and the reference value of the high-voltage DC bus voltage. First, the base maximum power point voltage output from the base maximum power point voltage query submodule and the predefined reference value of the high-voltage DC bus voltage are read synchronously. A timestamp comparison mechanism is used to ensure that the two data are matched data under the same operating cycle, avoiding deviation calculation errors caused by timing misalignment. Then, the voltage difference is calculated. The calculation logic is: Base offset = Base maximum power point voltage - High-voltage DC bus voltage reference value. This difference directly represents the degree of deviation between the voltage corresponding to the photovoltaic array's maximum power point and the expected bus voltage. During the calculation, a double-precision floating-point data format is used to ensure computational accuracy. Simultaneously, the validity of the two input voltage parameters is verified, and abnormal data exceeding the equipment's rated voltage range is eliminated to ensure that the output base offset accurately reflects the inherent deviation of the voltage reference.

[0109] The voltage compensation proportional calculation unit is used to perform proportional calculations based on the predicted load change amplitude and the preset voltage regulation coefficient to obtain the voltage compensation amount.

[0110] Optionally, the predicted load change amplitude is converted into a corresponding voltage compensation amplitude using a preset voltage regulation coefficient to adapt to the power balance requirements caused by load abrupt changes. First, the preset voltage regulation coefficient is read from the parameter configuration library. This coefficient is a fixed value calibrated based on numerous load abrupt change experiments. The calibration process aims at the optimal matching relationship between the load change amplitude and the voltage adjustment amount. After statistical fitting, it is stored in a non-volatile memory unit and can be directly accessed. Then, the predicted load change amplitude is extracted from the regenerative braking power withdrawal prediction signal and its validity is verified, eliminating extreme values ​​exceeding the reasonable range due to prediction errors. After successful verification, a proportional calculation is performed. The calculation logic is: Voltage compensation amount = Predicted load change amplitude × Preset voltage regulation coefficient. This calculation process converts the load power change into a voltage adjustment amount, achieving a precise match between load abrupt changes and voltage compensation, ensuring that the compensation amount can effectively balance the power fluctuations caused by load abrupt changes.

[0111] The reference voltage superposition unit is used to perform superposition calculations based on the voltage compensation amount and the reference offset to obtain the corrected reference voltage.

[0112] Optionally, the inherent deviation correction corresponding to the reference offset is combined with the load surge adaptation correction corresponding to the voltage compensation to obtain a preliminary voltage reference value that takes into account both photovoltaic characteristics and load requirements. The reference offset output from the reference offset calculation unit and the voltage compensation output from the voltage compensation ratio calculation unit are read synchronously. Data synchronization verification ensures the timing consistency of the two data sets, avoiding superposition errors caused by transmission delays. Then, superposition calculation is performed. The calculation logic is: corrected reference voltage = reference offset + voltage compensation. This superposition process organically integrates the correction of photovoltaic base voltage deviation with the compensation requirements for load surges, ensuring that the initially obtained reference voltage can both adapt to the power output characteristics of the photovoltaic array and proactively address load surges caused by regenerative braking power withdrawal. After the calculation is completed, the superposition results are initially screened to remove outliers that significantly exceed the bus voltage regulation range, providing high-quality input data for subsequent safety limiting processing.

[0113] The safety limiting unit is used to perform safety limiting processing based on the corrected reference voltage to obtain the corrected voltage reference command.

[0114] Optionally, the initially corrected reference voltage is clamped within the safe operating range of the high-voltage DC bus to ensure the safe operation of power devices and bus capacitors. First, the safe operating voltage range of the high-voltage DC bus is read from the system safety configuration library. This range is determined by the rated parameters of core components such as bus capacitors and power devices, and includes an upper and lower voltage limit. This range is pre-stored in the configuration library and can be updated based on device parameters. Then, the corrected reference voltage output from the reference voltage superposition unit is compared with the safe operating voltage range. If the corrected reference voltage is within this safe range, it is directly output as the corrected voltage reference command. If the corrected reference voltage is higher than the upper safe voltage limit, it is clamped to the upper safe voltage limit; if it is lower than the lower safe voltage limit, it is clamped to the lower safe voltage limit. After the limiting process is completed, the final corrected voltage reference command is output, along with a limiting status indicator to inform subsequent modules whether the current voltage command has been adjusted for limiting, ensuring the continuity and safety of the control logic.

[0115] In the above embodiment, the reference offset calculation unit clarifies the inherent deviation between the photovoltaic and bus voltages, the voltage compensation ratio calculation unit adapts to the compensation requirements of load mutations, the reference voltage superposition unit integrates the two types of correction requirements to obtain a preliminary reference voltage, and the safety limiting unit ensures the safety of the voltage command. This embodiment can accurately generate a corrected voltage reference command that takes into account photovoltaic characteristics, load mutation adaptation, and system safety, providing reliable input for photovoltaic control mode switching and control command calculation, effectively supporting forward-looking coordination between photovoltaic control and regenerative braking, reducing the risk of DC bus voltage fluctuations, and ensuring stable system operation.

[0116] In one embodiment, the voltage compensation ratio calculation unit includes:

[0117] The load change amplitude extraction subunit is used to extract data based on the predicted load change amplitude to obtain the load change amplitude.

[0118] Optionally, directional information is extracted from the predicted load change amplitude, retaining only the magnitude of the change for subsequent compensation calculations. First, raw data is read from the predicted load change amplitude obtained by analyzing the regenerative braking power withdrawal prediction signal. This data is then validated to ensure it falls within a pre-defined reasonable range for load changes, eliminating extreme outliers caused by prediction algorithm deviations or data transmission interference. After successful validation, amplitude extraction is performed. The extraction logic involves taking the absolute value of the predicted load change amplitude to obtain the load change amplitude. This amplitude directly reflects the magnitude of the load change caused by the regenerative braking power withdrawal, regardless of the direction of change. During extraction, a floating-point data format is used to preserve the original data precision, and a data validity identifier is added to the extraction result to inform subsequent sub-units that the current load change amplitude can be used in calculations.

[0119] The regulation coefficient configuration subunit is used to configure parameters according to the voltage regulation coefficient to obtain the regulation coefficient value.

[0120] Optionally, a parameter configuration and verification mechanism is employed to accurately determine the adjustment coefficient value, converting the preset voltage adjustment coefficient into a valid parameter value that can be directly used in the calculation. First, the preset voltage adjustment coefficient is read from the system parameter configuration library. This voltage adjustment coefficient is obtained through experimental calibration based on the optimal compensation effect under different load variation amplitudes. The calibration process aims to accurately balance the corresponding load changes and suppress bus voltage fluctuations. After statistical fitting, it is stored in a non-volatile storage unit. After reading, the voltage adjustment coefficient is verified. The verification includes whether the coefficient is within the preset valid value range and whether the data format meets the calculation requirements. If there are abnormal parameters, the default coefficient call mechanism is triggered, calling the pre-stored default adjustment coefficient. If the parameters are normal, they are directly output as the adjustment coefficient value, and the coefficient configuration status is recorded.

[0121] The voltage compensation product subunit is used to calculate the voltage compensation amount by multiplying the load change amplitude and the adjustment coefficient value. The voltage compensation amount is calculated using the following formula:

[0122]

[0123] in, This is the voltage compensation amount. To adjust the coefficient value, This represents the magnitude of the load change.

[0124] Optionally, the load change magnitude is converted into a corresponding voltage compensation magnitude by multiplying the load change amplitude by the adjustment coefficient value. First, the load change amplitude output from the load change amplitude extraction subunit and the adjustment coefficient value output from the adjustment coefficient configuration subunit are read synchronously. A timestamp comparison mechanism is used for timing synchronization verification to ensure that the two input parameters are matched data under the same operating cycle, avoiding calculation errors caused by timing deviations. After successful verification, the product calculation is strictly performed according to the given voltage compensation calculation formula, where... This is the voltage compensation amount. To adjust the coefficient value, This represents the load variation amplitude. The calculation employs a double-precision floating-point data format to ensure computational accuracy and avoid truncation errors affecting the compensation effect. After calculation, the results are screened to remove outliers exceeding the reasonable compensation range, and the final output is the verified voltage compensation amount.

[0125] In the above embodiments, the load change amplitude extraction subunit clearly identifies the core magnitude characteristics of load abrupt changes, the adjustment coefficient configuration subunit provides suitable compensation ratio parameters, and the voltage compensation product subunit completes the accurate mapping between load changes and voltage compensation through a standardized formula. This embodiment can efficiently generate voltage compensation amounts that accurately match the degree of load abrupt changes, providing high-quality input for subsequent reference offset superposition and safety limiting processing, effectively supporting the forward correction of voltage reference commands, ensuring the coordinated response effect of photovoltaic control and regenerative braking, and further reducing the risk of voltage fluctuations on the high-voltage DC bus.

[0126] In one embodiment, the photovoltaic control command dynamic calculation module includes:

[0127] The ripple convergence flag generation submodule is used to compare the DC bus voltage ripple data with a preset ripple threshold to obtain the ripple convergence flag.

[0128] Optionally, the stability of the bus voltage can be determined by quantifying the fluctuation of the DC bus voltage ripple. First, DC bus voltage ripple data is received. This data is a sequence of differences between real-time sampled values ​​and the mean value of the DC bus voltage. After receiving the data, a moving average filtering algorithm is used to smooth the ripple data, eliminating high-frequency noise interference and ensuring the accuracy of ripple amplitude judgment. Then, a preset ripple threshold is read from the system configuration library. This preset ripple threshold is determined based on the DC bus capacitor tolerance characteristics and the power device operating stability requirements. It is set after experimental calibration of the maximum ripple value under different stable operating conditions and is pre-stored in a non-volatile memory unit. A continuous sampling judgment logic is used to compare the filtered ripple data with the preset ripple threshold. If the ripple data amplitude in multiple consecutive sampling periods is less than or equal to the preset ripple threshold, the bus voltage is determined to be in a stable convergence state, and a ripple convergence flag of 1 is generated. If the ripple data amplitude in any sampling period is greater than the preset ripple threshold, or the continuous sampling condition is not met, a ripple convergence flag of 0 is generated. This continuous judgment mechanism can effectively avoid misjudgments caused by fluctuations in a single data point, ensuring the reliability of the generated flags.

[0129] The mode recovery flag operation submodule is used to perform AND logic operations based on the ripple convergence flag and the photovoltaic output power change trend to obtain the mode recovery flag.

[0130] Optionally, both ripple convergence and photovoltaic output power stability conditions must be satisfied simultaneously to ensure the rationality of the mode recovery timing and avoid premature recovery that could cause secondary voltage fluctuations. First, the input ripple convergence flag and photovoltaic output power change trend are preprocessed, converting the photovoltaic output power change trend into a binary state flag. The conversion logic involves extracting the slope of the photovoltaic output power sequence through linear fitting. If the absolute value of the slope is less than a preset power stability threshold, the photovoltaic output power is determined to be in a stable trend, generating a photovoltaic stability flag of 1; otherwise, it is 0. The power stability threshold is determined experimentally by calibrating the slope range under stable photovoltaic output conditions. Then, an AND operation is performed, with the operation rule being: mode recovery flag = ripple convergence flag & photovoltaic stability flag. The mode recovery flag is only 1 when both the ripple convergence flag and the photovoltaic stability flag are 1, indicating that the control mode recovery conditions are currently met; under any other combination, the mode recovery flag is 0, indicating that the recovery conditions are not yet met. After the operation is completed, the mode recovery flag is output, along with the original state data flag used in the operation, for subsequent traceability and verification.

[0131] The final control command generation submodule is used to calculate control parameters based on the mode recovery flag, control mode status, and corrected voltage reference command to obtain the final photovoltaic control command.

[0132] Optionally, based on the combination of the mode recovery flag and the control mode status, the corresponding control parameter calculation logic is matched to ensure that the control command is accurately adapted to the current operating condition. First, the mode recovery flag, control mode status, and corrected voltage reference command are read synchronously. A timestamp comparison mechanism is used to ensure that the three are matched data under the same operating condition cycle, avoiding command calculation errors caused by timing deviations. The control mode is handled according to the status of the mode recovery flag: If the mode recovery flag is 1, it means that the current bus voltage is stable and the photovoltaic output power is stable. The control mode needs to be restored to the normal maximum power point tracking mode. At this time, the parameter calculation logic corresponding to the disturbance observation method is called. The corrected voltage reference command is used as the target, and the switching duty cycle command of the photovoltaic array converter is calculated in combination with the photovoltaic output power change trend. If the mode recovery flag is 0, the current control mode state is maintained, and the corresponding appropriate control parameter calculation logic is called: If the current control mode state is voltage stability priority mode, the proportional-integral-derivative control algorithm is called. The corrected voltage reference command is used as the target value, and the real-time value of the DC bus voltage is used as the feedback quantity. The switching duty cycle command is calculated by adjusting the parameters of the proportional, integral, and derivative links. If it is cooperative tracking mode, the disturbance observation method and the proportional-integral-derivative control logic are integrated. The weight ratio of the two algorithms is dynamically adjusted according to the photovoltaic output power change trend, and the switching duty cycle command is calculated in combination with the corrected voltage reference command. All parameter calculations in all scenarios use double-precision floating-point data format to ensure accuracy. After the calculation is completed, the instructions are range-constrained to ensure that they are within the safe operating range of the converter switching devices. Finally, the verified photovoltaic control instructions are output.

[0133] In the above embodiment, the ripple convergence flag generation submodule accurately identifies the stable state of the bus voltage, the mode recovery flag calculation submodule rationally selects the timing for control mode recovery, and the final control command generation submodule adaptively matches the control logic calculation commands. This embodiment can dynamically adapt to the control requirements under different operating conditions. It can suppress voltage fluctuations through precise commands during load surges and promptly restore the maximum power capture mode after the operating conditions stabilize, achieving a balance between system stability and energy utilization efficiency. It effectively supports the full-condition coordination of photovoltaic control and regenerative braking, further reducing the risk of DC bus voltage fluctuations and improving overall energy efficiency.

[0134] The aforementioned electric vehicle hybrid energy supply system and control method based on solar and kinetic energy recovery first integrates multi-dimensional data from vehicle motion, power battery, photovoltaic array, and environment using a comprehensive data fusion module to obtain a comprehensive dataset. Then, a regenerative braking power withdrawal prediction module extracts braking characteristics, generates load mutation prediction flags, and predicts the magnitude of load changes, packaging them into a prediction signal. Subsequently, a reference voltage prediction and adjustment module corrects the voltage reference command based on the prediction signal and other parameters. A tracking mode setting module matches the corresponding control mode state. Finally, a photovoltaic control command dynamic calculation module combines ripple data and photovoltaic power change trends to generate the final photovoltaic control command. All sub-modules and units at each level achieve full-process collaboration through precise data processing and logical operations. This technical solution, through forward-looking collaborative regulation of photovoltaic control and regenerative braking, effectively avoids the drastic fluctuations in DC bus voltage caused by the superposition of two transient energy dynamic processes under discrete control, ensuring stable system operation and improving the overall efficiency of photovoltaic power generation and energy recovery.

[0135] 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 of other steps.

[0136] Based on the same inventive concept, this application also provides a control method for implementing the aforementioned hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery. The solution provided by this control method is similar to the implementation scheme described in the above system. Therefore, the specific limitations in one or more embodiments of the control method for hybrid energy supply systems for electric vehicles based on solar energy and kinetic energy recovery provided below can be found in the above-described limitations for hybrid energy supply systems for electric vehicles based on solar energy and kinetic energy recovery, and will not be repeated here.

[0137] In one exemplary embodiment, such as Figure 3 As shown, a control method for an electric vehicle hybrid energy supply system based on solar energy and kinetic energy recovery is provided, including:

[0138] S101. Based on vehicle motion state data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters, data fusion is performed to obtain a comprehensive dataset containing multi-dimensional state information;

[0139] S102. Based on the pedal opening, vehicle speed, motor torque and vehicle auxiliary load power in the comprehensive dataset, perform a retraction trend prediction calculation to obtain the regenerative braking power retraction prediction signal; the regenerative braking power retraction prediction signal includes a load change prediction flag and a predicted load change magnitude.

[0140] S103. Based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, high voltage DC bus voltage reference value and predicted load change amplitude, perform reference voltage prediction adjustment to obtain the corrected voltage reference command.

[0141] S104. Set the tracking mode according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal to obtain the control mode status.

[0142] S105. Based on the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend, the control command is dynamically calculated to obtain the final photovoltaic control command.

[0143] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement a module of an electric vehicle hybrid energy supply system based on solar energy and kinetic energy recovery as described above.

[0144] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements a module of a hybrid energy supply system for electric vehicles based on solar and kinetic energy recovery as described above.

[0145] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0146] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery, characterized in that, The system includes: The integrated data fusion module is used to fuse data based on vehicle motion status data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters to obtain a comprehensive dataset containing multi-dimensional status information; The regenerative braking power withdrawal prediction module is used to perform withdrawal trend prediction calculations based on the pedal opening, vehicle speed, motor torque, and vehicle auxiliary load power in the comprehensive dataset to obtain a regenerative braking power withdrawal prediction signal; the regenerative braking power withdrawal prediction signal includes a load change prediction flag and a predicted load change magnitude; The reference voltage prediction and adjustment module is used to perform reference voltage prediction and adjustment based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, voltage reference value of high voltage DC bus and the predicted load change amplitude, so as to obtain the corrected voltage reference command. The tracking mode setting module is used to set the tracking mode according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal, and obtain the control mode state. The photovoltaic control command dynamic calculation module is used to dynamically calculate the control commands based on the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend, so as to obtain the final photovoltaic control commands.

2. The system according to claim 1, characterized in that, The regenerative braking power withdrawal prediction module includes: The braking feature value calculation submodule is used to calculate feature values ​​based on the brake pedal opening and vehicle speed in the comprehensive dataset, and obtain intermediate feature value variables. The load mutation prediction flag generation submodule is used to obtain the load mutation prediction flag by comparing the intermediate variable of the feature value with a preset negative threshold. The predicted load change calculation submodule is used to calculate the power difference based on the motor torque, motor speed and vehicle auxiliary load power in the comprehensive dataset to obtain the predicted load change range. The data packaging submodule is used to package data according to the load change prediction flag and the predicted load change magnitude to obtain the regenerative braking power withdrawal prediction signal.

3. The system according to claim 2, characterized in that, The braking characteristic value calculation submodule includes: The brake pedal change rate calculation unit is used to perform differential calculations based on the brake pedal opening to obtain the brake pedal opening change rate. The normalized vehicle speed factor calculation unit is used to perform normalization calculations based on the vehicle speed and the vehicle's maximum design speed to obtain the normalized vehicle speed factor. The eigenvalue intermediate variable synthesis unit is used to calculate the eigenvalue intermediate variable by performing a weighted summation based on the brake pedal opening change rate and the normalized vehicle speed factor; the eigenvalue intermediate variable is calculated using the following formula: in, The intermediate variable for the feature value, As the first weighting coefficient, The rate of change of the brake pedal opening. This is the second weighting coefficient. For the vehicle speed, This indicates the vehicle's maximum design speed.

4. The system according to claim 2, characterized in that, The submodule for calculating the predicted load change magnitude includes: The current regenerative braking power calculation unit is used to calculate the current regenerative braking power by multiplying the motor torque and motor speed with the inverter efficiency. An auxiliary power reference value retrieval unit is used to retrieve data based on the vehicle's auxiliary load power to obtain an auxiliary power reference value. The load change amplitude difference unit is used to calculate the difference between the current regenerative braking power and the auxiliary power reference value to obtain the predicted load change amplitude.

5. The system according to claim 1, characterized in that, The reference voltage prediction and adjustment module includes: The control state update submodule is used to perform control state update processing based on the regenerative braking power withdrawal prediction signal to obtain updated control state data; the updated control state data is the pause standard maximum power point tracking logic. The basic maximum power point voltage query submodule is used to query a preset photovoltaic characteristic curve table based on the updated control status data and the ambient light irradiance and photovoltaic panel backsheet temperature in the photovoltaic array electrical parameters to obtain the basic maximum power point voltage. The voltage compensation and limiting submodule is used to perform voltage compensation and limiting calculations based on the base maximum power point voltage, the voltage reference value of the high voltage DC bus, and the predicted load change amplitude, to obtain the corrected voltage reference command.

6. The system according to claim 5, characterized in that, The voltage compensation and limiting submodule includes: The reference offset calculation unit is used to calculate the reference offset by calculating the voltage difference between the base maximum power point voltage and the voltage reference value of the high voltage DC bus. The voltage compensation ratio calculation unit is used to perform a ratio calculation based on the predicted load change amplitude and the preset voltage regulation coefficient to obtain the voltage compensation amount. A reference voltage superposition unit is used to perform superposition calculations based on the voltage compensation amount and the reference offset to obtain the corrected reference voltage; A safety limiting unit is used to perform safety limiting processing based on the corrected reference voltage to obtain the corrected voltage reference command.

7. The system according to claim 6, characterized in that, The voltage compensation ratio calculation unit includes: The load change amplitude extraction subunit is used to extract data based on the predicted load change amplitude to obtain the load change amplitude. The adjustment coefficient configuration subunit is used to configure parameters according to the voltage adjustment coefficient to obtain the adjustment coefficient value; The voltage compensation product subunit is used to calculate the voltage compensation amount by multiplying the load change amplitude and the adjustment coefficient value; the voltage compensation amount is calculated using the following formula: in, This refers to the voltage compensation amount. The adjustment coefficient value is... The magnitude of the load change.

8. The system according to claim 1, characterized in that, The photovoltaic control command dynamic calculation module includes: The ripple convergence flag generation submodule is used to compare the DC bus voltage ripple data with a preset ripple threshold to obtain a ripple convergence flag. The mode recovery flag calculation submodule is used to perform AND logic operations based on the ripple convergence flag and the photovoltaic output power change trend to obtain the mode recovery flag. The final control command generation submodule is used to calculate control parameters based on the mode recovery flag, the control mode state, and the corrected voltage reference command to obtain the final photovoltaic control command.

9. A control method for a hybrid energy supply system for electric vehicles based on solar energy and kinetic energy recovery, characterized in that, The method includes: Based on vehicle motion state data, power battery electrical parameters, photovoltaic array electrical parameters and environmental parameters, data fusion is performed to obtain a comprehensive dataset containing multi-dimensional state information; Based on the pedal opening, vehicle speed, motor torque, and vehicle auxiliary load power in the comprehensive dataset, a retraction trend prediction calculation is performed to obtain a regenerative braking power retraction prediction signal; the regenerative braking power retraction prediction signal includes a load change prediction flag and a predicted load change magnitude; Based on the regenerative braking power withdrawal prediction signal, photovoltaic array electrical parameters, high voltage DC bus voltage reference value, and predicted load change amplitude, the reference voltage prediction adjustment is performed to obtain the corrected voltage reference command; The tracking mode is set according to the corrected voltage reference command and the regenerative braking power withdrawal prediction signal to obtain the control mode state; Based on the control mode status, DC bus voltage ripple data, and photovoltaic output power change trend, the control command is dynamically calculated to obtain the final photovoltaic control command.

Images