A domain controller-based intelligent dynamic power following control method and system for a range extender power system

Abstract

This invention discloses an intelligent dynamic power following control method and system for a range-extended electric vehicle (REEV) powertrain based on a domain controller. The system includes: a multi-source parameter fusion sensing module; an intent prediction and real-time total demand power prediction module; a basic demand power mapping module; a bus voltage feedback correction and multi-dimensional power deviation correction module; a multi-objective decision fusion module and a thermal safety trimming module; a condition fluctuation filtering and Hold logic module; and a final command synthesis and smooth delivery module. Through accurate power prediction and following, the charging and discharging rate of the energy storage device is maintained within a low-loss range, achieving vehicle-battery lifespan and improving the overall system cycle efficiency. The introduction of Hold logic and variable slope control effectively filters out ineffective adjustments under complex road conditions, addressing the pain point of inconsistent sound speeds due to frequent fluctuations in the range extender's speed. AKF filtering for noise reduction and a three-level fault-tolerant degradation strategy ensure the system's operational safety under extreme conditions such as sensor failure or signal noise interference.

Description

Technical Field

[0001] This invention belongs to the field of power system control technology for new energy vehicles, specifically relating to an intelligent dynamic power following control method and system for range-extended power systems based on a domain controller. Background Technology

[0002] Traditional control strategies for range-extended electric vehicle powertrains, such as start-stop control based on SOC thresholds or fixed-point power control, have the following shortcomings:

[0003] 1. Power mismatch: The output power of the range-extended electric vehicle (REEV) cannot respond in real time to millisecond-level load changes at the drive end, resulting in frequent energy transfer within the battery and significant chemical-to-electrical energy conversion losses. 2. High battery thermal load: Due to the lack of accurate power prediction and tracking, the battery is often in a high-rate charging and discharging state, resulting in severe internal resistance heat generation and making it difficult to achieve "vehicle-battery lifespan in tandem." 3. NVH and response lag: Engine speed adjustment lags behind driver intention, creating an abrupt "inconsistent sound speed." 4. Functional isolation: Existing logics such as warm-up, idling, and regenerative braking are often independent of each other, lacking a unified power decision model. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent dynamic power following control method and system for range-extended electric vehicles based on a domain controller, which can effectively solve the aforementioned problems.

[0005] The technical solution adopted in this invention is as follows:

[0006] This invention provides an intelligent dynamic power follower control method for range-extended electric power systems based on domain controllers, applicable to range-extended electric power systems including a range extender, an energy storage device, and a drive unit; the method includes:

[0007] Step S1: Real-time synchronous acquisition of vehicle and environmental status data, followed by filtering, spatiotemporal alignment and fusion, to generate vehicle and environmental status feature vectors;

[0008] Step S2: Based on the vehicle and environmental state feature vector, the physical demand power is obtained using the vehicle longitudinal dynamics model; based on the physical demand power, the real-time total demand power is calculated using the real-time total demand power calculation model.

[0009] Step S3: Based on the vehicle and environmental state feature vector, identify the current operating condition of the vehicle, including normal operating condition and sensor failure condition; establish a fault-tolerant degradation control strategy with fault classification for the sensor failure condition; predict the driving intention based on the rate of change of the real-time total demand power; and determine the basic demand power of the range extender that matches the driving intention by performing dynamic weight mapping between the preset economic optimal operating curve and the transient response operating curve based on the predicted driving intention.

[0010] Step S4: Introduce a bus voltage sensitivity coefficient into the basic power demand of the range extender. The bus voltage feedback correction and multi-dimensional power deviation correction, after weighted aggregation of multi-dimensional power components and safety constraint limitations, yield the safety clamping power of the range extender; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions.

[0011] Step S5: Based on the safety clamping power, the steady-state target output power of the range extender is obtained by using a dual-criteria steady-state determination and a nonlinear slope limitation based on dynamic operating conditions.

[0012] Step S6: Implement adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output command, and send it to the range extender to adjust the output power of the range extender; when the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling in order to maintain the dynamic balance of the vehicle's energy flow.

[0013] Furthermore, step S1 includes:

[0014] Step S11: Collect raw multi-dimensional vehicle and environmental status data, including environmental status data and vehicle status data;

[0015] Step S12: For the original torque signal of the drive device containing high-frequency interference in the vehicle state data, an adaptive Kalman filter is used for smoothing. The discretized state-space equation is defined as:

[0016] (1)

[0017] in: This is the original torque signal; The torque signal is smoothed using an adaptive Kalman filter. The observation matrix defines the mapping relationship between the actual state of the system and the observed values. This is the state transition matrix, which describes the evolution logic of the system state over time. The torque signal from the previous moment has been smoothed using an adaptive Kalman filter; The Kalman gain is dynamically adjusted according to the signal variance;

[0018] Step S13, Time scale alignment of asynchronous data:

[0019] By introducing a zero-order hold combined with a linear interpolation operator, the sampling signal with a sampling frequency higher than the control period k is mapped to the corresponding control period k;

[0020] Step S14: Thus, at the beginning of each control cycle k, multi-dimensional environmental state data and vehicle state data are obtained synchronously.

[0021] The vehicle status data includes driving speed. acceleration Operating status of the drive unit, energy storage unit, range extender unit, busbar status, and real-time brake pedal opening. ;

[0022] The operating state of the drive device includes the real-time torque of the drive device. Real-time speed of the drive unit The operating state of the energy storage device includes the actual state of charge of the energy storage device. Real-time current of energy storage devices The operating status of the range extender includes the engine cooling temperature of the range extender. The bus condition includes the actual value of the bus voltage. .

[0023] Furthermore, step S2 includes:

[0024] Step S21: Using formula (2), the physical power requirement P is obtained using the longitudinal dynamics model of the whole vehicle. pre_demand :

[0025] (2)

[0026] in: The driving speed; The overall efficiency factor of the transmission system; This is an estimated vehicle weight, taking into account the difference between a fully loaded and unloaded vehicle. For acceleration; θ is the acceleration due to gravity; θ is the road slope. air density; This refers to the air drag coefficient; For windward area; This is the rolling resistance coefficient;

[0027] Step S22: Based on the physical demand power, the real-time total demand power is calculated using the real-time total demand power calculation model of formula (3). :

[0028] (3)

[0029] in: This refers to the real-time torque of the drive unit. This refers to the real-time rotational speed of the drive unit. For high-voltage accessory system power load; This is a compensation item for internal system friction. It is a dynamic weighting coefficient, which is adjusted according to the stability of the vehicle's driving conditions, the severity of the control component's movements, and whether a fault degradation mode has occurred.

[0030] Furthermore, step S3 includes:

[0031] Step S31: Pre-establish the economic optimal curve for the range extender. and transient response curve The economically optimal curve The current driving mode is economy mode, which represents the mapping between the real-time total power demand and the output power of the range extender engine that minimizes fuel consumption; the transient response curve This corresponds to the current driving mode as the power mode, and is used to represent the real-time total power demand. It is a mapping between the range extender engine output power obtained by the principle of the fastest dynamic response of the range extender engine to protect the energy storage device from generating a large current discharge.

[0032] Step S32: Using the dynamic weighting factor control formula of formula (5), a transient power demand change factor is introduced. The basic power requirement is obtained using formula (4). :

[0033] (4)

[0034] (5)

[0035] in: : Through transient response curves Mapped to real-time total power demand The corresponding range extender output power; : Through the economic optimal curve Mapped to real-time total power demand The corresponding range extender output power; : Real-time total demand power change rate; when When it increases, tending towards 1; when When things are stable, It tends towards 0; Sensitivity coefficient, estimated based on the vehicle mass m est Dynamic correction, m est The larger, The smaller the value, the more likely it is to suppress frequent power fluctuations under full load; : Saturation function, which restricts the weights to the range [0, 1].

[0036] Furthermore, step S4 includes:

[0037] Step S41, use formula (6) to calculate the basic power demand. Perform bus voltage feedback correction to obtain the basic power requirement for bus voltage correction. :

[0038] (6)

[0039] in: This is the target value for the bus voltage. This is the actual value of the bus voltage; The bus voltage sensitivity coefficient is used to adjust the intensity of the impact of bus voltage fluctuations on the power compensation of the range extender.

[0040] Step S42: Determine the final correction amount of the power compensation power based on the state of charge of the energy storage device. Based on the vehicle's motion state, predict whether braking energy will be generated soon; if so, determine the final correction amount for the kinetic energy compensation power. ;

[0041] Step S43: Using a weighted aggregation model of multidimensional power components, the initial output power of the range extender is obtained. :

[0042] (7)

[0043] in: : Dynamic weighting factor, dynamically determined based on the current vehicle's operating mode;

[0044] In driving mode, increase Prioritize economic efficiency; under power shortage conditions, increase To force the maintenance of power balance;

[0045] Step S44, Nonlinear safety trimming based on aircraft condition: Introduce a decay function based on the temperature and physical limits of the range extender to determine the power generation limit of the range extender. :

[0046] (8)

[0047] in: This represents the upper limit of the power generation capacity of the range extender. This is the attenuation steepness coefficient; The set alarm threshold for the engine cooling temperature of the range extender; This refers to the engine cooling temperature of the range extender.

[0048] Step S45, Electrical safety constraints on bus voltage and current:

[0049] (9)

[0050] in: For safety clamping power; The real-time maximum allowable charging power of the energy storage device; This refers to the real-time current of the energy storage device. The electrical safety factor is adjusted to balance the power of the vehicle and the protection requirements of the energy storage device.

[0051] Furthermore, the calculation method for the final correction amount of the power compensation power is as follows:

[0052] Step A1, using formula (10), based on the vehicle's speed Dynamically correct the basic balance value of the state of charge of energy storage devices The target value of the state of charge of the energy storage device is obtained. Using formula (11), the real-time state of charge deviation of the energy storage device is obtained. ;

[0053] (10)

[0054] (11)

[0055] in: This is the speed-based feedforward coefficient, which is negative and used to significantly reduce the target value of the state of charge at high vehicle speeds. To make the target value of the state of charge Significantly lower than the basic equilibrium value of the state of charge This reduces the SOC offset correction of the range extender, increases the power supply of the energy storage device, and reserves space for the energy recovered during braking in subsequent control cycles to be stored in the energy storage device. This represents the actual state of charge of the energy storage device.

[0056] A positive value indicates that the energy storage device is short of power, and the power correction caused by the SOC offset needs to be increased so that the output power of the range extender can supply power to the energy storage device, so that the energy storage device's power level approaches the target value of the state of charge. A negative value indicates that the energy storage device has excess power, and the power correction caused by the SOC offset needs to be reduced so that the energy storage device can drive the drive device to discharge in the next control cycle, so that the energy storage device's power level approaches the target value of the state of charge.

[0057] Step A2, set the real-time state of charge deviation value of the energy storage device. By performing a nonlinear gain mapping that introduces dead-zone adjustment, the initial correction amount of the power compensation is obtained. :

[0058] (12)

[0059] Where: δ is the preset dead zone threshold for power correction, which is a positive value; This is the nonlinear gain coefficient for power replenishment when the energy storage device is short of power. This is the nonlinear gain coefficient for load reduction when the energy storage device has excess power. It is a non-linear power exponent; non-linear power exponent The introduction of this makes the correction intensity grow non-linearly with the power deviation; when the actual state of charge of the energy storage device is severely depleted, the correction amount of the basic demand power of the range extender increases in a stepwise manner. By charging the energy storage device through the range extender, the actual state of charge of the energy storage device is forced to return to the normal state, preventing the energy storage device from being over-discharged.

[0060] Step A3, power smoothing based on a first-order inertial element:

[0061] Initial correction amount of power compensation Converted into a power compensation smoothing correction amount with a smooth slope. :

[0062] (13)

[0063] (14)

[0064] in: It is a smoothing factor; To control the scale of the cycle; As a time constant, it is dynamically adjusted according to the current driving mode. In economy mode, it increases. To ensure smoother changes in the operating conditions of the range extender and reduce fuel consumption; in power mode, reduce To accelerate the power follow-up response of the range extender and prevent over-discharge of the energy storage device; To control the cycle Power compensation smoothing correction amount at that time;

[0065] Step A4, power limiting of the coupled battery state of health (SOH):

[0066] Based on the actual charging and discharging capacity of the energy storage device, adjust the power compensation smoothing amount. Power limiting is performed to obtain the final correction amount of the power compensation power. : (15)

[0067] (16)

[0068] (17)

[0069] in: This represents the current actual charging power capacity limit of the energy storage device. This represents the current actual discharge power capacity limit of the energy storage device. This represents the original charging power limit of the energy storage device when it is not in use. This represents the initial discharge power limit of the energy storage device in its unused state. This represents the current health status of the energy storage device; 1.0 indicates that it is not in use, and 0.8 indicates that it is nearing the end of its lifespan. This is the aging sensitivity coefficient; as energy storage devices age, Reduce and automatically compress the maximum power amplitude allowed for the range extender to recharge or discharge power; This is the attenuation scaling factor;

[0070] Step A5: Perform secondary calibration of coupling internal resistance compensation using formula (18):

[0071] (18)

[0072] in: This is the upper limit safety red line value for bus voltage; This is the open-circuit voltage of the energy storage device; This is the actual value of the bus voltage; To take into account the internal resistance of the energy storage device after aging;

[0073] If formula (18) is not satisfied, the calibration result is abnormal, triggering the perception-limited mode in the fault-tolerant degradation control strategy of the fault classification.

[0074] Furthermore, the final correction amount of the kinetic energy compensation power The calculation method is as follows:

[0075] Step B1: Obtain driving speed based on multi-sensor motion state monitoring. acceleration Road gradient θ and real-time brake pedal opening And estimate the estimated value of the vehicle mass m. est ;

[0076] Step B2: Define the regenerative braking potential power based on the vehicle's current kinetic energy change trend. :

[0077] (19)

[0078] in: It is the acceleration due to gravity; The overall efficiency factor for regenerative braking;

[0079] Step B3, Dynamic feedforward compensation based on braking intent:

[0080] Rate of change of brake pedal Identify the driver's intention to decelerate and brake, and generate an initial correction for the kinetic energy compensation power. :

[0081] (20)

[0082] (twenty one)

[0083] in: The acceleration threshold; As a slope predictor; This represents the maximum permissible opening of the brake pedal. This is the braking intensity sensitivity coefficient, or the slope prediction gain;

[0084] Step B4, calibrating the output of kinetic energy compensation:

[0085] (twenty two)

[0086] in: This represents the maximum permissible power reduction slope for the range extender. This is the final correction amount for kinetic energy compensation power.

[0087] Furthermore, step S5 includes:

[0088] Step S51, Dual-criteria steady-state determination:

[0089] (1) Calculate the safety clamping power obtained from the control cycle k The target output power obtained from the previous control cycle k-1 absolute deviation value :

[0090] (twenty three)

[0091] (2) Establish a dual-criteria steady-state determination logic:

[0092] Condition 1, Power Deviation Criterion:

[0093] in: The preset power lockout threshold;

[0094] Condition 2, Time Duration Criterion:

[0095] Where: t stable The cumulative time in the time domain for the current Hold command to enter the steady-state evaluation window; Maximum allowed lock time;

[0096] (3) Logical branch execution path:

[0097] Steady-state maintenance mode, Hold: If both of the above conditions are met, the current target output power is forcibly set to equal the target output power at the previous moment, i.e.: ;

[0098] Steady-state breakout pattern: If ,or Release the Hold lock state and update the target output power: Simultaneously reset the timer. Prepare to enter the next decision loop;

[0099] (4) Anti-frequent jump design based on hysteresis comparator:

[0100] To avoid repeatedly switching between Hold states, a hysteresis coefficient β is introduced. hold The value ranges from 0.8 to 0.95.

[0101] Threshold for exiting the Hold state: ;

[0102] The threshold for re-entering the Hold state: ;

[0103] Step S52, based on nonlinear slope constraint under dynamic operating conditions;

[0104] (twenty four)

[0105] (25)

[0106] (26)

[0107] in: The power-up slope; To reduce the power slope; This refers to the real-time rotational speed of the range extender. This refers to the engine cooling temperature of the range extender. Operating mode; This represents the final output power of the range extender.

[0108] Furthermore, the fault-tolerant degradation control strategy for fault classification includes three levels of fault-tolerant degradation logic:

[0109] ① Level 1 Degradation, Sensing-Limited Mode: When abnormal data from energy storage devices or the environment is detected, the sensing-limited mode is triggered; in the sensing-limited mode, the base demand power is locked to the last base demand power before the sensing limitation; the bus voltage sensitivity coefficient is actively increased. This increases the bus voltage correction term, which in turn ultimately boosts the safety clamping power.

[0110] ② Level 2 Degradation, Aircraft Risk Mode: When the engine cooling temperature of the range extender is detected... When lost, the maintenance risk mode is triggered; in the maintenance risk mode, the maximum safety power limit output strategy is executed to limit the safety clamping power to less than 50% of the rated power of the range extender, so as to prevent mechanical damage to the range extender caused by thermal management runaway.

[0111] ③ Level 3 Degradation, Limp Mode: When all power-related data of the drive unit is detected to be lost, the limp mode is triggered; in the limp mode, the range extender operates smoothly at a fixed optimal efficiency point power, only used to maintain basic power to allow the vehicle to slowly stop in a safe area.

[0112] This invention also provides an intelligent dynamic power following control system for range-extended electric vehicles based on a domain controller, comprising:

[0113] The multi-source parameter fusion sensing module is used to collect vehicle and environmental status data in real time and synchronously. After filtering, spatiotemporal alignment and fusion, it generates vehicle and environmental status feature vectors.

[0114] The intent prediction and real-time total demand power prediction module is used to obtain the physical demand power by using the vehicle longitudinal dynamics model based on the vehicle and environmental state feature vectors; and to calculate the real-time total demand power based on the physical demand power using the real-time total demand power calculation model.

[0115] The basic demand power mapping module is used to identify the current operating conditions of the vehicle, including normal operating conditions and sensor failure conditions, based on the vehicle and environmental state feature vectors. The sensor failure conditions are based on a fault-tolerant degradation control strategy with fault classification. The module predicts the driving intention based on the rate of change of the real-time total demand power. Based on the predicted driving intention, the module performs dynamic weight mapping between a preset economic optimal operating curve and a transient response operating curve to determine the basic demand power of the range extender that matches the driving intention.

[0116] The bus voltage feedback correction and multi-dimensional power deviation correction module incorporates a bus voltage sensitivity coefficient into the basic power requirement of the range extender. Bus voltage feedback correction and multi-dimensional power deviation correction; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions.

[0117] The multi-objective decision fusion module and the thermal safety trimming module are used to obtain the safety clamping power of the range extender after weighted aggregation of multi-dimensional power components and safety constraint limitation;

[0118] The operating condition fluctuation filtering and Hold logic module is used to obtain the steady-state target output power of the range extender based on the safety clamping power, after adopting a dual-criteria steady-state determination and a nonlinear slope limit based on dynamic operating conditions.

[0119] The final instruction synthesis and smooth delivery module performs adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output instruction, which is then sent to the range extender to adjust its output power. When the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling to maintain the dynamic balance of the vehicle's energy flow.

[0120] This invention provides an intelligent dynamic power follower control method and system for range-extended electric vehicles based on a domain controller, which has the following advantages:

[0121] (1) High efficiency and long life: Through accurate power prediction and tracking, the charge and discharge rate of the energy storage device is maintained in the low loss range, realizing "vehicle and battery life" and improving the overall cycle efficiency of the system.

[0122] (2) Excellent NVH performance: The introduction of Hold locking logic and variable slope control effectively filters out invalid adjustments under complex road conditions and solves the pain point of "inconsistent sound speed" caused by frequent fluctuations in the speed of the range extender.

[0123] (3) Extremely high system robustness: Through AKF filtering and noise reduction and a perfect three-level fault tolerance and degradation strategy, the system's operational safety is ensured under extreme conditions such as sensor failure or signal noise interference.

[0124] (4) Highly integrated functional architecture: The functions of Warm-up, SOC balancing, and regenerative braking are coordinated and decided under the DICO domain control platform, eliminating the control conflicts caused by functional isolation in traditional solutions. Attached Figure Description

[0125] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0126] Figure 1 The flowchart shows the intelligent dynamic power following control method for range-extended electric vehicles based on a domain controller provided by this invention. Detailed Implementation

[0127] To make the technical problems solved, the technical solutions, and the beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and are not intended to limit the invention.

[0128] This invention provides an intelligent dynamic power follow-up control method and system for range-extended electric vehicle powertrains based on a domain controller. Specifically, it relates to an energy management and logic control approach operating within a range-extended domain controller (DICO), enabling intelligent dynamic power follow-up (IDPF) control of the powertrain system in range-extended electric vehicles. This invention aims to address issues such as power follow-up lag, high battery thermal load, poor NVH performance, and logic isolation in traditional strategies through a logical architecture. By fusing multi-source sensing and multi-objective decision-making, it achieves high-precision and smooth matching between the range extender's output power and drive requirements, improving system energy efficiency while also protecting battery life and ensuring overall operational safety.

[0129] like Figure 1As shown, this invention provides an intelligent dynamic power following control method for a range-extended powertrain system based on a domain controller, applicable to range-extended powertrain systems including a range extender, an energy storage device, and a drive unit. It should be emphasized that this invention does not limit the specific structural forms of the range extender, energy storage device, and drive unit. The range extender can be a range extender or other types of power sources, such as a diesel engine, a natural gas engine, or a hydrogen fuel cell. The energy storage device can be a power battery or a supercapacitor. The drive unit can be a drive motor used to drive the vehicle wheels to rotate, thereby driving the vehicle's movement.

[0130] The method includes steps S1 to S6:

[0131] Step S1: Real-time synchronous acquisition of vehicle and environmental status data, followed by filtering, spatiotemporal alignment and fusion, to generate vehicle and environmental status feature vectors;

[0132] This step is performed by the "Data Awareness Layer" in the DICO domain controller. Its core objective is to overcome bus transmission latency and sensor noise by constructing a multi-dimensional spatiotemporal feature vector that includes vehicle and environmental states through cross-domain data fusion.

[0133] Step S11, Data perception layer based on cross-domain bus integration: Collect raw multi-dimensional vehicle and environmental status data, including environmental status data and vehicle status data;

[0134] In this embodiment, the DICO domain controller acts as the energy management hub, integrating cross-domain data of the entire vehicle in real time via a standard CAN bus (communication rate of 250K or 500K bps). The system utilizes the efficient interrupt mechanism of the processing chip to synchronize the algorithm execution cycle with the high-frequency frames of the bus, establishing a stable power-following scheduling cycle. The specific logic is as follows:

[0135] ① Periodic Synchronization Extraction Logic:

[0136] The core algorithm control cycle k (i.e. execution cycle) of the system is constant at 10ms. Within each control cycle, the algorithm triggers a global variable update once and synchronously extracts the latest vehicle and environmental status data from the chip communication buffer. That is, one control cycle corresponds to one complete data sampling and extraction, ensuring that the decision command and the highest frequency power frame of the bus are kept in real-time synchronization at a ratio of 1:1.

[0137] ② Data layered asynchronous integration:

[0138] The system captures bus messages at different frequencies through an interrupt mechanism and stores them in categories for algorithm calls:

[0139] 10ms keyframe: Captures the original torque signal and real-time speed of the drive unit. Busbar data, etc.;

[0140] 50ms / 100ms status frames: Acquire quasi-static signals such as SOC and temperature of the energy storage device.

[0141] ③ Calculate load optimization:

[0142] Through the aforementioned 10ms synchronous scheduling mechanism, the system can achieve a dynamic response output as fast as 10ms while keeping the CPU utilization of the processing chip at around 30%, ensuring communication robustness and system operation stability in a low-speed bus environment of 250K / 500K.

[0143] In this embodiment, the system utilizes the hardware interrupt buffer of the processing chip to perform a "synchronous snapshot extraction" of the vehicle data at the beginning of each 10ms constant control cycle. This 1:1 timing synchronization mechanism ensures that the calculation input for the subsequent S2 step is the most real-time physical snapshot on the current bus, eliminating the logic jitter caused by asynchronous sampling from the bottom layer.

[0144] Step S12, signal denoising based on adaptive Kalman filtering (AKF):

[0145] Considering the drastic load fluctuations on the drive end of commercial vehicles and other vehicles, the original torque signal M raw Including high-frequency interference, the original torque signal of the drive device in the vehicle state data containing high-frequency interference is smoothed using an adaptive Kalman filter. Its discretized state-space equation is defined as:

[0146] (1)

[0147] in: The original torque signal, i.e., the original torque signal M raw ; The system state estimate is given at control period k. In this example, it is the torque signal after smoothing using an adaptive Kalman filter. The Kalman gain is dynamically adjusted according to the signal variance; The observation matrix, also known as the measurement matrix, defines the mapping relationship between the actual state of the system and the observed values. This is the state transition matrix, which describes the evolution logic of the system state over time. The torque signal from the previous moment has been smoothed using an adaptive Kalman filter;

[0148] This step effectively filters out signal glitches caused by vibrations in the commercial vehicle's transmission system, while maintaining a rapid response to the driver's true power intentions, thus avoiding misadjustment of the range extender due to "signal distortion".

[0149] Step S13, Time scale alignment of asynchronous data:

[0150] Because the sampling frequencies of each sensor system are different, a zero-order hold (ZOH) is introduced in conjunction with a linear interpolation operator to measure the asynchronously arriving SOC and the range extender engine cooling temperature. Once the signal is mapped into the 10ms core control cycle of DICO:

[0151] In this embodiment, a zero-order hold combined with a linear interpolation operator is introduced to map the sampling signal with a sampling frequency higher than the control period k to the corresponding control period k. The model is as follows:

[0152] ;

[0153] This step ensures that all physical parameters are on the same time base when executing subsequent power fusion logic, eliminating control oscillations caused by phase differences.

[0154] Step S14: Thus, at the beginning of each control cycle k, multi-dimensional environmental state data and vehicle state data are obtained synchronously.

[0155] The vehicle status data includes driving speed. acceleration Operating status of the drive unit, energy storage unit, range extender unit, busbar status, and real-time brake pedal opening. ;

[0156] The operating state of the drive device includes the real-time torque of the drive device. Real-time speed of the drive unit The operating state of the energy storage device includes the actual state of charge of the energy storage device. Real-time current of energy storage devices The operating status of the range extender includes the engine cooling temperature of the range extender. The bus condition includes the actual value of the bus voltage. .

[0157] The above multi-dimensional environmental and vehicle status data are encapsulated into feature vectors. This serves as the global input for subsequent steps.

[0158] Of course, before constructing the feature vector, a signal rationality check and health assessment (PlausibilityCheck) can be added:

[0159] The system performs a confidence assessment on the raw signal. Abnormal sensor data is identified through preset "physical threshold check (Range Check) and rate of change gradient check (Gradient Check)".

[0160] Logical definition: If a certain signal (such as driving speed) Or the actual value of the bus voltage ) N consecutive cycles of data loss (CAN Timeout) or instantaneous jump rate exceeding physical limits If the signal is detected, it is determined that the signal has entered an "unhealthy state".

[0161] Alternative mechanism: Once an abnormal signal is detected, the system automatically activates a virtual estimation based on Analytical Redundancy. For example, if the vehicle speed signal fails, the system will instead perform a reverse calculation using the motor feedback speed and the main reduction ratio to ensure that the perception layer is not interrupted due to a single point of failure.

[0162] Step S2: Based on the vehicle and environmental state feature vector, the physical demand power is obtained using the vehicle longitudinal dynamics model; based on the physical demand power, the real-time total demand power is calculated using the real-time total demand power calculation model.

[0163] Step S21, predict the physical power demand based on the longitudinal dynamics of the vehicle (virtual sensing):

[0164] To compensate for the bus transmission lag, formula (2) is used to obtain the physical power requirement P based on the vehicle's longitudinal dynamics model. pre_demand :

[0165] (2)

[0166] in: The driving speed; The overall efficiency factor of the transmission system; This is an estimated vehicle weight, taking into account the difference between a fully loaded and unloaded vehicle. For acceleration; θ is the acceleration due to gravity; θ is the road slope. air density; This refers to the air drag coefficient; For windward area; This is the rolling resistance coefficient;

[0167] Therefore, the system not only reads the "predetermined power requirements" of the CAN bus, but also estimates the overall vehicle weight. Slope θ and air drag coefficient C d The system autonomously calculates physical requirements based on parameters. This "dual-source verification" mechanism improves the system's robustness under complex mountain roads or full-load acceleration conditions.

[0168] In step S22, the system first receives and parses the instruction stream feature vector output in step S14. The physical demand power P obtained in step S21 pre_demand This serves as the core input source for this step and subsequent allocation logic, providing it with preprocessed, denoised, and physically model-calibrated structured data packets. The system retrieves these packets in real time. The components in the equation serve as the basic parameters for calculating the total load from multiple sources, thereby ensuring that the input of the power calculation has both temporal consistency and physical rationality.

[0169] The DICO domain controller does not directly reference the power of the drive unit, but instead calculates the total real-time power demand of the entire vehicle using Formula 3. .

[0170] Based on the physical demand power, the real-time total demand power calculation model of formula (3) is used to calculate the real-time total demand power. :

[0171] (3)

[0172] in: This refers to the real-time torque of the drive unit. The real-time rotational speed (rpm) of the drive unit. This refers to the power load of high-voltage accessory systems, including the power consumption of low-voltage / high-voltage accessories such as air pumps, air conditioners, and power steering. This is a compensation item for internal system friction. It is a dynamic weighting coefficient, which is adjusted according to the stability of the vehicle's driving conditions, the severity of the control component's movements, and whether a fault degradation mode has occurred.

[0173] Dynamic weighting coefficients This coefficient is not a fixed ratio, but is used for real-time balancing. (Measured power on the bus) and physical power requirement The weight.

[0174] Purpose and Innovation Logic of This Step:

[0175] Eliminating bus latency: Considering the millisecond-level latency of the CAN bus (250K / 500K bps), when the vehicle is under stable operating conditions, The value tends towards 1, and the system primarily relies on the measured torque after AKF noise reduction. ;

[0176] Predictive adjustment: When the system... When the feature vector detects a drastic change in the driver's pedal position but the actual bus measurement value has not yet been updated, the system actively reduces... Increase physical power requirements The weight.

[0177] Fault tolerance correlation: If the signal fails and enters "Level 1 Degradation Mode", then By forcibly setting the value to 0, the system relies entirely on the physical model or voltage feedback for power estimation.

[0178] Technical effect: This "feedforward + feedback" weighted mode ensures that the system's real-time total power demand is within a 10ms control cycle. It has both physical support and can overcome the lag of bus signals to achieve true millisecond-level dynamic balance.

[0179] Step S3: Based on the vehicle and environmental state feature vector, identify the current operating condition of the vehicle, including normal operating condition and sensor failure condition; establish a fault-tolerant degradation control strategy with fault classification for the sensor failure condition; predict the driving intention based on the rate of change of the real-time total demand power; and determine the basic demand power of the range extender that matches the driving intention by performing dynamic weight mapping between the preset economic optimal operating curve and the transient response operating curve based on the predicted driving intention.

[0180] Step S31: To balance economy and dynamic response, the system presets two-dimensional Generator Control Logic (GCL) characteristic curves:

[0181] Pre-establish the economic optimal curve for the range extender. and transient response curve ;

[0182] The economic optimal curve The power mapping, generated based on the principle of minimum engine fuel consumption rate (BSFC lowest), corresponds to the current driving mode as economy mode. It is used to represent the mapping between the real-time total power demand and the output power of the range extender engine with the minimum fuel consumption rate.

[0183] The transient response curve The power mapping generated based on the principle of fastest dynamic response of generator and protection of large current discharge of battery corresponds to the current driving mode as the power mode, and is used to represent the real-time total demand power, and the mapping between the range extender engine output power obtained by the principle of fastest dynamic response of range extender engine to protect energy storage device from large current discharge.

[0184] In this embodiment, the range extender includes an engine and a generator; the engine consumes fuel to drive the generator to generate electricity, which in turn drives the drive unit or charges the energy storage unit.

[0185] Step S32: Using the dynamic weighting factor control formula of formula (5), a transient power demand change factor is introduced. The basic power requirement is obtained using formula (4). :

[0186] (4)

[0187] (5)

[0188] in: : Through transient response curves Mapped to real-time total power demand The corresponding range extender output power; : Through the economic optimal curve Mapped to real-time total power demand The corresponding range extender output power; : Real-time total demand power change rate; when When it increases, tending towards 1; when When things are stable, It tends towards 0; Sensitivity coefficient, estimated based on the vehicle mass m est Dynamic correction, m est The larger, The smaller the value, the more likely it is to suppress frequent power fluctuations under full load; : Saturation function, which restricts the weights to the range [0, 1].

[0189] Step S3 implements the optimal operating point search based on a two-dimensional map. The dynamic weighting factor control formula is based on the proactive prediction idea of ​​the demand power change rate. The system does not adopt a single curve mapping, but introduces a transient demand change factor. , Factor protection protects energy storage devices from instantaneous high current surges.

[0190] The meaning and innovation of formulas (4) and (5) lie in: when the driver accelerates hard, the rate of change of power demand... When it increases, When the value approaches 1, the system switches to the "transient response curve" to protect the battery from generating a sudden surge in current; when the vehicle is traveling at a constant speed, As fuel approaches zero, the system switches to the "economically optimal curve" to save fuel. The specific meaning is as follows:

[0191] ① Transient power demand variation factor and sensitivity coefficient :

[0192] Physical implications: It is an "acceleration sensor" that senses the driver's intention to respond, calculating the rate of change in total power demand in real time. It can identify whether a vehicle is transitioning from a steady-state condition to a transient condition (such as rapid acceleration or hill start) or maintaining a cruise condition.

[0193] Dynamic sensitivity adaptation: Introducing a sensitivity coefficient And based on the estimated vehicle weight m est Real-time correction. For example, under full load, the system automatically reduces... This value is used to increase the damping effect, prevent frequent power fluctuations caused by signal fluctuations, and ensure smooth NVH performance under high mass conditions.

[0194] 2. Transient power demand variation factor Core uses:

[0195] Response path switching: When the driver suddenly presses the accelerator, When the system logic approaches 1, it instantly switches from the "fuel economy curve" to the "transient response curve". This switching allows the range extender to accelerate before the energy storage device's discharge peak, effectively protecting the energy storage device from generating instantaneous super-currents and extending battery life.

[0196] Feedforward compensation mechanism: Essentially, it provides a power feedforward channel. Before the bus voltage fluctuates significantly, the system has already issued a power request in advance through intent recognition, solving the acceleration "sluggishness" problem caused by the lag in traditional feedback control.

[0197] Step S4: Introduce a bus voltage sensitivity coefficient into the basic power demand of the range extender. The bus voltage feedback correction and multi-dimensional power deviation correction, after weighted aggregation of multi-dimensional power components and safety constraint limitations, yield the safety clamping power of the range extender; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions.

[0198] Step S41: Considering that the bus voltage fluctuates greatly when medium and heavy commercial vehicles are going downhill or accelerating under full load, the system introduces a voltage compensation operator to calibrate the mapping results.

[0199] Formula (6) is used to determine the basic power requirement. Perform bus voltage feedback correction to obtain the basic power requirement for bus voltage correction. :

[0200] (6)

[0201] in: This is the target value for the bus voltage. This is the actual value of the bus voltage; The bus voltage sensitivity coefficient is used to adjust the intensity of the impact of bus voltage fluctuations on the power compensation of the range extender.

[0202] This step, bus voltage feedback correction, is based on the passive compensation concept of voltage dips; it is the core of fine-tuning under normal conditions and sensing when limitations are present.

[0203] , is the voltage drop compensation factor, which is the last line of defense for the balance of system energy flow.

[0204] This formula ensures that when the internal resistance of the energy storage device changes or the bus voltage shifts, the generator output power can automatically compensate for the voltage drop, maintaining the stability of the system's power output. Furthermore, this factor has a mode-switching logical attribute: under sensing-limited (first-level degradation) conditions, the system automatically disables the feedforward channel in step S32, increasing the voltage sensitivity coefficient σ in step S41 to the main control weight, achieving seamless logical reconfiguration from load forecasting to voltage feedback.

[0205] Step S42: Determine the final correction amount of the power compensation power based on the state of charge of the energy storage device. Based on the vehicle's motion state, predict whether braking energy will be generated soon; if so, determine the final correction amount for the kinetic energy compensation power. ;

[0206] Final correction amount for power compensation:

[0207] This step is performed by the "SOC Offset Correction Module" in the DICO domain controller. Its core objective is to smoothly adjust the output power of the range extender based on the degree to which the current state of charge (SOC) of the energy storage device deviates from the target value, thereby maintaining the balance of the power supply while minimizing drastic changes in engine operating conditions.

[0208] The method for calculating the final correction amount of the power compensation power is as follows:

[0209] Step A1: First, calculate the real-time state of charge deviation of the energy storage device at the current moment. Unlike traditional fixed target values, this scheme introduces a value based on vehicle speed. The dynamic objective of compensation:

[0210] Using formula (10), based on the vehicle's speed Dynamically correct the basic balance value of the state of charge of energy storage devices The target value of the state of charge of the energy storage device is obtained. Using formula (11), the real-time state of charge deviation of the energy storage device is obtained. ;

[0211] (10)

[0212] (11)

[0213] in: The basic balance value of the state of charge of the energy storage device, such as 50%.

[0214] This is the speed-based feedforward coefficient, which is negative and used to significantly reduce the target value of the state of charge at high vehicle speeds. To make the target value of the state of charge Significantly lower than the basic equilibrium value of the state of charge This reduces the SOC offset correction of the range extender, increases the power supply of the energy storage device, and reserves space for the energy recovered during braking in subsequent control cycles to be stored in the energy storage device. This represents the actual state of charge of the energy storage device.

[0215] : Vehicle speed feedforward term. Utilizing the negative correlation mapping logic between vehicle speed and target SOC, under high vehicle speed conditions, the system proactively and appropriately lowers the target SOC command (e.g., dynamically reducing it from 50% to 45%), thereby reserving a 'buffer' (i.e., buffer capacity) for the high-power kinetic energy recovery generated during subsequent deceleration / braking of commercial vehicles. This strategy of proactively creating space effectively avoids power limiting triggered by instantaneous overcharging of the energy storage device, maximizing recovery efficiency.

[0216] A positive value indicates that the energy storage device is short of power, and the power correction caused by the SOC offset needs to be increased so that the output power of the range extender can supply power to the energy storage device, so that the energy storage device's power level approaches the target value of the state of charge. A negative value indicates that the energy storage device has excess power, and the power correction caused by the SOC offset needs to be reduced so that the energy storage device can drive the drive device to discharge in the next control cycle, so that the energy storage device's power level approaches the target value of the state of charge.

[0217] Step A2 introduces nonlinear gain mapping for dead-zone regulation (the final closed loop for energy storage device lifetime protection and energy flow balance):

[0218] To achieve a smooth correction and prevent power oscillations near the preset equilibrium point of charge state due to sensor noise or transient fluctuations, this step employs a nonlinear mapping function with a dead zone. To calculate the initial correction amount of power compensation .

[0219] Deviation of real-time state of charge of energy storage device By performing a nonlinear gain mapping that introduces dead-zone adjustment, the initial correction amount of the power compensation is obtained. :

[0220] (12)

[0221] in:

[0222] The initial power correction for power compensation is the basic power correction triggered by the power deviation, reflecting the allowable output power of the range extender under the condition of the energy storage device.

[0223] δ is the preset dead zone threshold for energy correction, which is a positive value, such as 2%. The system assumes that the energy storage device's energy is in a safe balance zone and does not trigger the range extender's engine operating condition adjustment to achieve "absolute silence".

[0224] This is the nonlinear gain coefficient for power replenishment when the energy storage device is short of power. This is the nonlinear gain coefficient for load shedding when the energy storage device has excess power; it is typically set as follows: Prioritize ensuring power safety.

[0225] Non-linear power exponent, weighting factor; non-linear power exponent The introduction of this feature causes the correction intensity to increase non-linearly with the power deviation. When the actual state of charge of the energy storage device is severely depleted, the correction amount of the base demand power of the range extender increases dramatically. By charging the energy storage device through the range extender, the actual state of charge of the energy storage device is forced to return to a normal state, preventing over-discharge. This reflects the system principle of prioritizing the safety of the energy storage device. The typical value range is 1.1 ≤ ≤ 2.0.

[0226] Step A3, power smoothing based on a first-order inertial element:

[0227] This step aims to address the initial power correction amount in response to power compensation in commercial vehicle range extenders.

[0228] At the same time, the engine speed jumps drastically due to the step-like step change, and the deterioration of NVH noise, vibration, and acoustic roughness.

[0229] Technical implementation principle: The system introduces a discretized first-order low-pass filter to convert the calculated initial power correction amount for power compensation. Converted into a power compensation smoothing correction amount with a smooth slope. .

[0230] Core mathematical model: Initial correction amount of power compensation Converted into a power compensation smoothing correction amount with a smooth slope. :

[0231] (13)

[0232] (14)

[0233] in: It is a smoothing factor; To control the scale of the period, such as 10ms; This is a time constant, and its value is not fixed; it is dynamically adjusted according to the current driving mode. In economy mode, it increases. (e.g., 2.0s) to make the range extender's operating conditions change smoothly and reduce fuel consumption; in power mode, reduce (e.g., 0.5s) to accelerate the power follow-up response of the range extender and prevent over-discharge of the energy storage device; To control the cycle Power compensation smoothing correction amount at that time;

[0234] This step achieves the following effect:

[0235] Sound follows speed: via The adjustment achieves linear consistency between changes in engine sound and the perceived acceleration of the vehicle, eliminating any abruptness.

[0236] Mechanical protection: avoids the instantaneous large opening and closing of the throttle valve and fuel injection volume of the range extender, reducing carbon deposits caused by incomplete combustion.

[0237] Step A4, power limiting of the coupled battery state of health (SOH):

[0238] This step embodies the core protection logic of the "vehicle and battery lifespan" vision. By introducing battery health (SOH) and physical limit constraints, the final correction amount is non-linearly pruned.

[0239] Dynamic limiting logic definition: final correction amount of power compensation. It must be clamped within the "Peak Power Capability" of the energy storage device at the current moment.

[0240] Core mathematical model: The system calculates the final power by querying a three-dimensional safety matrix with the energy storage device temperature T, real-time SOC, and health status SOH as independent variables.

[0241] Based on the actual charging and discharging capacity of the energy storage device, adjust the power compensation smoothing amount. Power limiting is performed to obtain the final correction amount of the power compensation power. : (15)

[0242] (16)

[0243] (17)

[0244] in: This represents the current actual charging power capacity limit of the energy storage device. This represents the current actual discharge power capacity limit of the energy storage device; it is constrained by temperature (T), energy level (SOC), and state of health (SOH).

[0245] This represents the original charging power limit of the energy storage device when it is not in use. This represents the initial discharge power limit of the energy storage device in an unused state; that is, the initial power limit of the new energy storage device, which is usually obtained by looking up a table in the three-dimensional safety matrix Map.

[0246] This represents the current health status of the energy storage device. 1.0 indicates that it is not in use. For new energy storage devices, 0.8 indicates that it is nearing the end of its lifespan. This is the aging sensitivity coefficient; as energy storage devices age, Reduce and automatically compress the maximum power amplitude allowed for the range extender to recharge or discharge power; This is the attenuation scaling factor;

[0247] Step A5, considering the internal resistance of the energy storage device after aging. Increase To prevent instantaneous voltage drop overload during power replenishment, formula (18) is used for secondary calibration of coupling internal resistance compensation:

[0248] (18)

[0249] in: This is the upper limit safety red line value for bus voltage; This is the open-circuit voltage of the energy storage device; This is the actual value of the bus voltage; To account for the internal resistance of the energy storage device after aging; if formula (18) is not met, the calibration result is abnormal, triggering the fault-tolerant degradation control strategy of the fault classification. Specifically, the system will automatically jump to the first-level degradation mode. In this mode, the system achieves dynamic reconstruction of the control weight by fixing the base power and simultaneously increasing the voltage sensitivity coefficient. Thus, the system changes from active control based on power prediction to passive feedback following based on bus voltage drop compensation factor, ensuring the real-time balance of the vehicle's energy flow under sensor data failure conditions.

[0250] This step achieves the following technical effects:

[0251] Lifecycle closed-loop management: This step ensures that the power system control variables can automatically adapt to the degradation characteristics of the energy storage device throughout its entire life cycle, fundamentally preventing the risk of battery pack thermal runaway caused by "old batteries and high current".

[0252] Safety redundancy: By coupling the power correction power with the physical safety red line (voltage, temperature) in real time, cross-domain safety strategy integration is achieved.

[0253] The final correction amount of the kinetic energy compensation power The calculation method is as follows:

[0254] This step is executed by the "Kinetic Energy Correction Module (GCLcorrection by kinetic energy)" in the DICO domain controller. Its core logic is to predict the regenerative energy to be generated based on the vehicle's current motion state, thereby actively reducing the output power of the range extender at the feedforward end to reserve space for battery absorption of regenerative braking energy. The feedforward compensation amount is provided by the "Kinetic Energy Correction Module" based on the vehicle speed change rate. In conjunction with the estimated vehicle weight, the feedforward compensation amount is dynamically adjusted. This ensures that braking energy is recovered preferentially.

[0255] Step B1: Obtain driving speed based on multi-sensor motion state monitoring. acceleration Road gradient θ and real-time brake pedal opening And estimate the estimated value of the vehicle mass m. est ;

[0256] Specifically, the DICO domain controller collects vehicle speed in real time via the CAN bus. acceleration (or rate of change of vehicle speed) ), slope θ and brake pedal opening Simultaneously, the extended state observer (ESO) or recursive least squares (RLS) method is used to estimate the vehicle mass in real time. The estimated vehicle mass m is... est Consider the difference between full load and no load.

[0257] Step B2, Calculation of braking potential and rate of change of kinetic energy:

[0258] Define the regenerative braking potential power based on the current kinetic energy change trend of the vehicle. :

[0259] (19)

[0260] in: It is the acceleration due to gravity; The overall efficiency factor for regenerative braking is calculated, taking into account motor and inverter losses. The inertial power generated by deceleration. The gravitational potential energy power generated when going downhill.

[0261] Step B3, Dynamic feedforward compensation based on braking intent:

[0262] Rate of change of brake pedal Identify the driver's intention to decelerate and brake, and generate an initial correction for the kinetic energy compensation power. :

[0263] (20)

[0264] (twenty one)

[0265] in: The acceleration threshold; As a slope predictor; This represents the maximum permissible opening of the brake pedal. This is the braking intensity sensitivity coefficient, or the slope prediction gain;

[0266] Meaning: When braking intensity increases or during a long downhill slope, the system will calculate a larger negative power compensation value. This value will directly affect the output of the range extender, preventing energy accumulation at the energy storage device by "reducing power" in advance.

[0267] Step B4, Output calibration of compensation command: To avoid abrupt changes in driving feel due to excessive feedforward compensation, the final compensation amount passes through a dynamic limiter:

[0268] Output calibration of kinetic energy compensation:

[0269] (twenty two)

[0270] in: The maximum allowable power reduction slope for the range extender ensures a smooth transition in engine operating conditions; This is the final correction amount for kinetic energy compensation power.

[0271] Step S43, Multi-objective decision fusion:

[0272] Instead of simply adding the power components algebraically, the system dynamically assigns weighting coefficients based on the current vehicle operating mode, employing a weighted aggregation model of multi-dimensional power components to obtain the initial output power of the range extender. :

[0273] (7)

[0274] in: : Dynamic weighting factor, dynamically determined based on the current vehicle's operating mode;

[0275] In driving mode, increase Prioritize economic efficiency; under power shortage conditions, increase To force the maintenance of power balance;

[0276] Step S44, thermal safety trimming, nonlinear safety trimming based on maintenance status:

[0277] To protect the range extender hardware from damage (engine maintenance), and especially addressing the common problem of "overheating engines" in commercial vehicles, a decay function based on the range extender's temperature and physical limits is introduced to determine the range extender's power generation limit. :

[0278] (8)

[0279] in: This represents the upper limit of the power generation capacity of the range extender. This is the attenuation steepness coefficient; The set alarm threshold for the engine cooling temperature of the range extender; This refers to the engine cooling temperature of the range extender.

[0280] Logical meaning: When the engine cooling temperature exceeds the threshold, the formula uses the exponential decay characteristic to quickly reduce the upper limit of power generation, sacrificing battery power to ensure the thermal safety of the range extender and prevent mechanical fatigue.

[0281] Step S45: Based on the real-time maximum allowable charging power fed back by the energy storage device The power is ultimately clamped to prevent overcharging.

[0282] Electrical safety constraints on bus voltage and current:

[0283] (9)

[0284] in: For safety clamping power; The real-time maximum allowable charging power of the energy storage device; This refers to the real-time current of the energy storage device. The electrical safety factor is adjusted to balance the power of the vehicle and the protection requirements of the energy storage device.

[0285] Steps S43 and S44 are executed by the "Multi-Objective Decision Fusion Module" in the DICO domain controller. Its core logic is to collaboratively optimize fuel economy (GCL), battery life (SOC stability), and mechanical safety (temperature / voltage limitation) while meeting the vehicle's power requirements.

[0286] Step S5, Operating condition fluctuation filtering and Hold logic execution: Based on the safety clamping power, after adopting dual-criteria steady-state determination and nonlinear slope limitation based on dynamic operating conditions, the steady-state target output power of the range extender is obtained;

[0287] This step is performed by the "steady-state scheduling module" of the DICO domain controller. Its core objective is to optimize the final safety clamping power after multi-objective fusion and thermal safety trimming. A smooth lock-in determination is performed. By utilizing the energy buffering characteristics of the energy storage device, ineffective micro-power fluctuations are shielded, ensuring a high degree of stability of the engine operating point in the time domain.

[0288] Step S51: Real-time deviation quantification and standardization processing, dual-criteria steady-state determination:

[0289] (1) Calculate the safety clamping power obtained from the control cycle k The target output power obtained from the previous control cycle k-1 absolute deviation value :

[0290] (twenty three)

[0291] Key logic point: The object of comparison here is the final intended value after "all corrections (SOC, SOH, environment, thermal safety) are superimposed". This ensures that the Hold logic, as the "last line of defense", can intercept low-frequency disturbances from all sources.

[0292] (2) Establish a dual-criteria steady-state determination logic:

[0293] The system establishes a dynamic steady-state evaluation window, and the Hold logic is triggered only when the power deviation and duration simultaneously meet the following conditions:

[0294] Condition 1, Power Deviation Criterion:

[0295] in: The preset power lock-in threshold; recommended range for commercial vehicles: 1kW~20kW.

[0296] Condition 2, Time Duration Criterion:

[0297] Where: t stableThe time-domain cumulative duration for the current Hold lock command to enter the steady-state evaluation window is used to prevent the range extender from frequently changing its operating state due to instantaneous disturbances; The maximum allowed locking time, typically 1s to 10s.

[0298] (3) Logical branch execution path:

[0299] Based on the judgment result, the system executes different instruction update strategies:

[0300] Steady-state maintenance mode (Hold Active), Hold lock: If both of the above conditions are met simultaneously, the current state is determined to be "ineffective load fluctuation," and the current target output power is forcibly set to equal the target output power at the previous moment, i.e.: ;

[0301] Technical effect: At this time, the excess power gap (or surplus) on the drive end is instantly made up or absorbed by the power battery, and the engine speed and fuel injection quantity remain absolutely constant.

[0302] Steady-state breakout mode (Hold Release): If , ,or This indicates that although the fluctuation is small, the deviation is persistent, requiring power correction, release of the Hold lock state, and updating of the target output power. Simultaneously reset the timer. Prepare to enter the next decision loop;

[0303] (4) Anti-frequent jump design based on hysteresis comparator:

[0304] To prevent logical conflicts at threshold edges, i.e., repeated switching of the Hold lock state, the system introduces a hysteresis coefficient β. hold The value ranges from 0.8 to 0.95.

[0305] Threshold for exiting the Hold state: ;

[0306] The threshold for re-entering the Hold state: ;

[0307] Function: By using the cyclic hysteresis logic, the robustness of the control system is ensured, and the oscillation of the algorithm itself is avoided.

[0308] The final output power in this step is the target output power. Since high-frequency components have been filtered out through the previous steps, the output signal has extremely high time-domain stability, laying a logical foundation for the smooth delivery of subsequent execution layers.

[0309] Step S52, final instruction synthesis and smooth delivery:

[0310] This step is performed by the "Command Output Gateway Module" of the DICO domain controller, and its core objective is to increase the target output power. It is converted into a physical signal that can be recognized by the generator controller (GCU) and ensures that the power change process is smooth and safe.

[0311] Mapping and transformation from power to execution layer instructions:

[0312] The system will determine the final output power based on the current optimal efficiency curve (OEA) of the range extender. Further decomposed into target rotational speed n cmd and target torque M cmd The combination of instructions. If the GCU supports direct power control, then signal encapsulation is performed directly.

[0313] Nonlinear slope constraint based on dynamic operating conditions;

[0314] To prevent mechanical shocks or excessive emissions from sudden power changes in the range extender, the system implements adaptive rate limiting on the output commands.

[0315] (twenty four)

[0316] (25)

[0317] (26)

[0318] in: The power-up slope; To reduce the power slope; This refers to the real-time rotational speed of the range extender. This refers to the engine cooling temperature of the range extender. Operating mode; This represents the final output power of the range extender.

[0319] The effect is as follows:

[0320] Low temperature protection: When the engine cooling temperature of the range extender is... At lower levels, ΔR is forcibly reduced to allow the engine to gradually increase power and prevent thermal stress damage.

[0321] NVH optimization: When the speed passes through the resonance range of the range extender, the slope is dynamically reduced to achieve "silent crossing".

[0322] Rapid acceleration response: In Power Mode, the slope limit is appropriately relaxed to ensure immediate power response.

[0323] Step S6: Implement adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output command, and send it to the range extender to adjust the output power of the range extender; when the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling in order to maintain the dynamic balance of the vehicle's energy flow.

[0324] Protocol encapsulation and high-frequency transmission via CAN bus:

[0325] 1. Signal encoding: P out Encoded according to J1939 or a custom protocol, it typically includes power values, mode bits, and an Alive Counter.

[0326] 2. Redundancy check: A checksum mechanism is introduced to ensure that the signal is transmitted without distortion in the complex electromagnetic environment of the dynamic domain.

[0327] 3. Timed transmission: The DICO domain controller sends data to the GCU via the CAN bus at a high frequency of 10ms. Control frames ensure real-time tracking.

[0328] Command feedback monitoring and closed-loop correction (enhancing stability)

[0329] The system monitors the actual output power P fed back by the GCU in real time. act If the deviation between the two continues to exceed the safety threshold ε: ,but:

[0330] 1. Fault self-diagnosis: If the system determines that the response is "lagging" or "hardware is limited", it will immediately trigger the re-arbitration of the operating condition fluctuation filtering and Hold logic execution in step S5, and reduce the target power to prevent energy loss control.

[0331] 2. Integral compensation: If the deviation is within the allowable range, the feedforward term of the next cycle is finely adjusted using the deviation value to eliminate the steady-state error of the system.

[0332] To ensure that the system does not crash under extreme sensor failure conditions, this invention also designs a fault-tolerant strategy, which includes three levels of fault-tolerant logic:

[0333] ① Level 1 Degradation, Perception-Limited Mode:

[0334] When an anomaly is detected in the energy storage device (SOC / SOH) or environmental status data, the system no longer relies on the active correction logic in step S4 and triggers the sensing-limited mode; in the sensing-limited mode, the base demand power is locked to the last base demand power before the sensing-limited event; and the bus voltage sensitivity coefficient is actively increased. This increases the bus voltage correction term, which in turn ultimately boosts the safety clamping power.

[0335] Base power fixed: The system will Locked to the last effective steady-state value before perception is limited (or the preset driving base power).

[0336] Compensation weight shift: The system actively increases the bus voltage sensitivity coefficient. .

[0337] Passive follower logic: In this case, the system fully utilizes the defined voltage sag compensation factor. To capture changes in the overall vehicle load.

[0338] When the motor load increases, it causes the bus voltage to rise. When falling, Instantaneous increase, forced pull .

[0339] In this mode, the algorithm architecture does not need to be changed; it only strengthens the bus voltage sensitivity coefficient. The feedback weights enable the system to switch from "feedforward prediction" to "passive feedback following" based on voltage fluctuations, ensuring the integrity of the control chain.

[0340] ② Level 2 Degradation, Aircraft Risk Mode: When the engine cooling temperature of the range extender is detected... When lost, the maintenance risk mode is triggered; in the maintenance risk mode, the maximum safety power limit output strategy is executed to limit the safety clamping power to less than 50% of the rated power of the range extender, so as to prevent mechanical damage to the range extender caused by thermal management runaway.

[0341] ③ Level 3 Degradation, Limp Home: When all power-related data of the drive unit is detected to be lost, the limp home mode is triggered. In the limp home mode, the range extender operates smoothly at a fixed optimal efficiency point power (e.g., 30kW) and is only used to maintain basic power for the vehicle to slowly stop in a safe area.

[0342] This fault-tolerant logic establishes a closed-loop degradation path from "optimal control in all states" to "minimum functional safety," ensuring the controllability of the range-extended power system under extreme disturbances and reflecting the high reliability requirements of industrial applications.

[0343] To achieve the above objectives, the present invention provides an intelligent dynamic power following control system for range-extended electric vehicles based on a domain controller, comprising:

[0344] The multi-source parameter fusion sensing module is used to collect vehicle and environmental status data in real time and synchronously. After filtering, spatiotemporal alignment and fusion, it generates vehicle and environmental status feature vectors.

[0345] The intent prediction and real-time total demand power prediction module is used to obtain the physical demand power by using the vehicle longitudinal dynamics model based on the vehicle and environmental state feature vectors; and to calculate the real-time total demand power based on the physical demand power using the real-time total demand power calculation model.

[0346] The basic demand power mapping module is used to identify the current operating conditions of the vehicle, including normal operating conditions and sensor failure conditions, based on the vehicle and environmental state feature vectors. The sensor failure conditions are based on a fault-tolerant degradation control strategy with fault classification. The module predicts the driving intention based on the rate of change of the real-time total demand power. Based on the predicted driving intention, the module performs dynamic weight mapping between a preset economic optimal operating curve and a transient response operating curve to determine the basic demand power of the range extender that matches the driving intention.

[0347] The bus voltage feedback correction and multi-dimensional power deviation correction module incorporates a bus voltage sensitivity coefficient into the basic power requirement of the range extender. Bus voltage feedback correction and multi-dimensional power deviation correction; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions.

[0348] The multi-objective decision fusion module and the thermal safety trimming module are used to obtain the safety clamping power of the range extender after weighted aggregation of multi-dimensional power components and safety constraint limitation;

[0349] The operating condition fluctuation filtering and Hold logic module is used to obtain the steady-state target output power of the range extender based on the safety clamping power, after adopting a dual-criteria steady-state determination and a nonlinear slope limit based on dynamic operating conditions.

[0350] The final instruction synthesis and smooth delivery module performs adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output instruction, which is then sent to the range extender to adjust its output power. When the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling to maintain the dynamic balance of the vehicle's energy flow.

[0351] As another embodiment, the present invention provides an intelligent dynamic power following control system for a range-extended electric vehicle based on a domain controller, comprising the following core modules and steps:

[0352] (1) Multi-source parameter fusion perception and intent prediction module:

[0353] The perception layer utilizes a high-speed bus to synchronously collect vehicle and environmental status data in real time. After filtering, spatiotemporal alignment, and fusion, it generates vehicle and environmental status feature vectors. Among these features, a time-scale alignment algorithm is introduced to eliminate phase differences between cross-domain signals. The original torque signal is denoised using an adaptive Kalman filter (AKF). Based on the vehicle and environmental status feature vectors, the physical power requirement is obtained using a longitudinal dynamics model of the entire vehicle.

[0354] (2) Real-time total demand power calculation module:

[0355] Based on the vehicle and environment state feature vectors output by the perception layer, and using the physical demand power, the real-time total demand power calculation model is employed to calculate the real-time total demand power. ;

[0356] (3) Basic Demand Power Mapping Module:

[0357] The driving intention is predicted based on the rate of change of the real-time total demand power; based on the predicted driving intention, a dynamic weight mapping is performed between the preset economic optimal operating curve and the transient response operating curve to determine the base demand power of the range extender that matches the driving intention.

[0358] Specifically, a transient power demand variation factor based on vehicle speed and load change rate is adopted. It switches in real time between economic efficiency (OEA curve) and dynamic tracking performance (transient response curve), outputting the base demand power P. GCL_raw .

[0359] (4) Bus voltage feedback correction and multi-dimensional power deviation correction module:

[0360] Bus voltage feedback correction: Correction is performed using a model that incorporates a bus voltage sensitivity coefficient;

[0361] Final correction to power compensation power, including:

[0362] SOC nonlinear correction: A power function mapping model with dead zone is adopted to dynamically calculate the compensation power based on the power deviation gradient, so as to achieve a smooth regression of power.

[0363] Nonlinear gain mapping: A nonlinear mapping function with a dead zone is used to achieve a "smooth" correction and prevent "power oscillation" caused by sensor noise or transient fluctuations near the equilibrium point.

[0364] Power smoothing based on first-order inertial elements: This solves the problems of drastic engine speed jumps and NVH (noise, vibration, and harshness) deterioration caused by step-like abrupt changes in commercial vehicle range extenders.

[0365] SOH Lifetime Trimming: Real-time dynamic correction of charge and discharge boundaries based on battery health status (internal resistance, aging factor).

[0366] Secondary calibration of coupling internal resistance compensation: to ensure that the control variables of the power system can automatically adapt to the degradation characteristics of the battery throughout its entire life cycle.

[0367] Kinetic energy compensation power correction: By combining the vehicle speed change rate and mass estimation value, the braking feedback power is predicted in advance to reduce unnecessary power output of the range extender.

[0368] (5) Multi-objective decision fusion and thermal safety trimming module:

[0369] Correct the bus voltage to the base demand power The range extender's initial output power is generated by weighting and aggregating the various correction values. The system introduces maintenance safety limit control, performing nonlinear power trimming based on engine exhaust / coolant temperature. It employs electrical safety constraints based on bus voltage and current for forced clamping. Furthermore, it features a three-level fault degradation logic, sequentially entering perception-limited, maintenance risk, and limp-out modes in the event of sensor malfunction or communication loss.

[0370] (6) Operating condition fluctuation filtering and Hold logic module:

[0371] A dual-criteria steady-state determination window based on "power deviation threshold" and "duration threshold" is established. When the fluctuation at the drive end is within the preset range, the current power generation command is locked, and the battery throughput is used to buffer minor disturbances, thereby achieving "silence" of the engine operating condition and optimizing NVH performance.

[0372] (7) Final instruction synthesis and smooth delivery module:

[0373] The aggregated power command is converted into an execution-layer physical signal, and an adaptive rate limit is applied. The slope limit is dynamically adjusted according to the engine's current speed range and warm-up state to ensure that the power boost process avoids the resonance zone and achieves smooth power delivery.

[0374] This invention proposes an Intelligent Dynamic Power Follower (IDPF) system, whose core principle is based on "dynamic reconfiguration of fully balanced flow energy". This system completely overturns the control logic of traditional range-extended power systems that are "battery-driven with range extender passively supplementing power", and instead adopts a balanced control concept of "range extender actively covering power + battery elastically filling gaps".

[0375] In this architecture, the system uses the Domain Controller (DICO) to consistently use the total load of "drive load + accessory load" as the benchmark for real-time power adjustment of the range extender, and redefines the power battery as a "dynamic power reservoir" with peak shaving and valley filling functions. Through this load-feedforward-based active prediction and real-time reconstruction mechanism, the system can precisely control the energy output of the range extender and energy storage devices (such as the power battery) and the range extender (or fuel cell) within an extremely fast response frequency of 10ms, ensuring that the energy flow of the entire vehicle achieves a dual dynamic balance between the physical layer and the algorithm decision layer under various operating conditions at the physical execution layer.

[0376] 1. Acceleration dynamic following principle based on full load perception:

[0377] During vehicle acceleration or high-load operation, this invention breaks away from the traditional limitation of generating electricity solely based on the drive motor's needs, and implements a "full-load integrated compensation" strategy:

[0378] Total demand aggregation: The system collects and processes the instantaneous power demand of the drive motor in real time. It also synchronously couples the real-time power consumption of high-voltage accessory systems (including air conditioning, power steering, brake pump, DC-DC converter, etc.). .

[0379] Energy flow distribution: by formula Determine the total power demand in real time.

[0380] Dual-source collaborative path: The range extender dynamically follows the calculated real-time total power demand, and its output not only supplies the motor but also directly covers the power consumption of accessories. When the real-time total power demand exceeds the range extender's optimal efficiency range, the power battery performs pulsed power compensation through a parallel path to ensure smooth driving and prevent power depletion.

[0381] 2. Predictive kinetic energy recovery and avoidance principle:

[0382] During vehicle deceleration or braking, the system optimizes the energy recovery path through feedforward logic:

[0383] Intent prediction and avoidance: Through the kinetic energy recovery prediction module, the system can predict the amount of regenerative energy at the beginning of braking and proactively reduce the output power of the range extender in advance.

[0384] Multi-path feedback distribution: The feedback current generated by the motor flows preferentially to energy storage devices such as batteries for replenishment.

[0385] Safety dissipation logic: If the battery SOC is high or in a state of low temperature limiting, the system will guide the excess energy to the accessories for consumption or dissipate it through the brake resistor according to the electrical safety constraint logic to ensure the stability of the bus voltage.

[0386] 3. Cross-dimensional logic closed-loop control principle:

[0387] The energy flow of the physical layer is strictly controlled by the six modules of the algorithm layer:

[0388] Perception layer: Generates data containing AKF filtering and a longitudinal dynamics model. The feature vectors enable a forward-looking perception of the total load.

[0389] Decision-making level: Determine the optimal energy allocation ratio using the dynamic weight matrix (DWM) and the SOC / SOH correction factor.

[0390] Constraint and Execution Layer: By using forced electric safety clamping and steady-state window, interference caused by accessory start / stop or minor pedal fluctuations is filtered out, achieving "high smoothness and high stability" of the range extender output, and finally issuing commands through adaptive slope limiting.

[0391] Compared with the prior art, the present invention has the following significant advantages:

[0392] (1) High efficiency and long life: Through accurate power prediction and tracking, the charge and discharge rate of the energy storage device is maintained in the low loss range, realizing "vehicle and battery life" and improving the overall cycle efficiency of the system.

[0393] (2) Excellent NVH performance: The introduction of Hold locking logic and variable slope control effectively filters out invalid adjustments under complex road conditions and solves the pain point of "inconsistent sound speed" caused by frequent fluctuations in the speed of the range extender.

[0394] (3) Extremely high system robustness: Through AKF filtering and noise reduction and a perfect three-level fault tolerance and degradation strategy, the system's operational safety is ensured under extreme conditions such as sensor failure or signal noise interference.

[0395] (4) Highly integrated functional architecture: The functions of Warm-up, SOC balancing, and regenerative braking are coordinated and decided under the DICO domain control platform, eliminating the control conflicts caused by functional isolation in traditional solutions.

[0396] This invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program; when the computer program is executed by the processor, the processor performs the steps of the intelligent dynamic power follower control method for a range-extended power system based on a domain controller as described in the above embodiments.

[0397] This invention also provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of the domain controller-based intelligent dynamic power follower control method for range-extended electric power systems as described in the above embodiments.

[0398] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0399] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0400] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0401] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0402] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A domain controller based intelligent dynamic power following control method for a range extended powertrain system, characterized in that, The method is applied to a range-extended power system including a range extender, an energy storage device, and a drive unit; the method includes: Step S1: Real-time synchronous acquisition of vehicle and environmental status data, followed by filtering, spatiotemporal alignment and fusion, to generate vehicle and environmental status feature vectors; Step S2: Based on the vehicle and environmental state feature vector, the physical demand power is obtained using the vehicle longitudinal dynamics model; based on the physical demand power, the real-time total demand power is calculated using the real-time total demand power calculation model. Step S3: Based on the vehicle and environmental state feature vector, identify the current operating condition of the vehicle, including normal operating condition and sensor failure condition; establish a fault-tolerant degradation control strategy with fault classification for the sensor failure condition; predict the driving intention based on the rate of change of the real-time total demand power; and determine the basic demand power of the range extender that matches the driving intention by performing dynamic weight mapping between the preset economic optimal operating curve and the transient response operating curve based on the predicted driving intention. Step S4: Introduce a bus voltage sensitivity coefficient into the basic power demand of the range extender. The bus voltage feedback correction and multi-dimensional power deviation correction, after weighted aggregation of multi-dimensional power components and safety constraint limitations, yield the safety clamping power of the range extender; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions. Step S5: Based on the safety clamping power, the steady-state target output power of the range extender is obtained by using a dual-criteria steady-state determination and a nonlinear slope limitation based on dynamic operating conditions. Step S6: Implement adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output command, and send it to the range extender to adjust the output power of the range extender; when the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling in order to maintain the dynamic balance of the vehicle's energy flow.

2. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, Step S1 includes: Step S11: Collect raw multi-dimensional vehicle and environmental status data, including environmental status data and vehicle status data; Step S12: For the original torque signal of the drive device containing high-frequency interference in the vehicle state data, an adaptive Kalman filter is used for smoothing. The discretized state-space equation is defined as: （1） in: This is the original torque signal; The torque signal is smoothed using an adaptive Kalman filter. The observation matrix defines the mapping relationship between the actual state of the system and the observed values. This is the state transition matrix, which describes the evolution logic of the system state over time. The torque signal from the previous moment has been smoothed using an adaptive Kalman filter; The Kalman gain is dynamically adjusted according to the signal variance; Step S13, Time scale alignment of asynchronous data: By introducing a zero-order hold combined with a linear interpolation operator, the sampling signal with a sampling frequency higher than the control period k is mapped to the corresponding control period k; Step S14: Thus, at the beginning of each control cycle k, multi-dimensional environmental state data and vehicle state data are obtained synchronously. The vehicle status data includes driving speed. acceleration Operating status of the drive unit, energy storage unit, range extender unit, busbar status, and real-time brake pedal opening. ; The operating state of the drive device includes the real-time torque of the drive device. Real-time speed of the drive unit The operating state of the energy storage device includes the actual state of charge of the energy storage device. Real-time current of energy storage devices The operating status of the range extender includes the engine cooling temperature of the range extender. The bus condition includes the actual value of the bus voltage. .

3. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, Step S2 includes: Step S21, the physical demand power P is obtained by using the formula (2) and using the longitudinal dynamics model of the whole vehicle pre_demand : （2） in: The driving speed; The overall efficiency factor of the transmission system; This is an estimated vehicle weight, taking into account the difference between a fully loaded and unloaded vehicle. For acceleration; θ is the acceleration due to gravity; θ is the road slope. air density; This refers to the air drag coefficient; For windward area; This is the rolling resistance coefficient; Step S22: Based on the physical demand power, the real-time total demand power is calculated using the real-time total demand power calculation model of formula (3). : （3） in: This refers to the real-time torque of the drive unit. This refers to the real-time rotational speed of the drive unit. For high-voltage accessory system power load; This is a compensation item for internal system friction. It is a dynamic weighting coefficient, which is adjusted according to the stability of the vehicle's driving conditions, the severity of the control component's movements, and whether a fault degradation mode has occurred.

4. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, Step S3 includes: Step S31: Pre-establish the economic optimal curve for the range extender. and transient response curve The economically optimal curve The current driving mode is economy mode, which represents the mapping between the real-time total power demand and the output power of the range extender engine that minimizes fuel consumption; the transient response curve This corresponds to the current driving mode as the power mode, and is used to represent the real-time total power demand. It is a mapping between the range extender engine output power obtained by the principle of the fastest dynamic response of the range extender engine to protect the energy storage device from generating a large current discharge. Step S32: Using the dynamic weighting factor control formula of formula (5), a transient power demand change factor is introduced. The basic power requirement is obtained using formula (4). : （4） （5） in: : Through transient response curves Mapped to real-time total power demand The corresponding range extender output power; : Through the economic optimal curve Mapped to real-time total power demand The corresponding range extender output power; : Real-time total demand power change rate; when When it increases, tending towards 1; when When things are stable, It tends towards 0; Sensitivity coefficient, estimated based on the vehicle mass m est Dynamic correction, m est The larger, The smaller the value, the more likely it is to suppress frequent power fluctuations under full load; : Saturation function, which restricts the weights to the range [0, 1].

5. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, Step S4 includes: Step S41, use formula (6) to calculate the basic power demand. Perform bus voltage feedback correction to obtain the basic power requirement for bus voltage correction. : （6） in: This is the target value for the bus voltage; This is the actual value of the bus voltage; The bus voltage sensitivity coefficient is used to adjust the intensity of the impact of bus voltage fluctuations on the power compensation of the range extender. Step S42: Determine the final correction amount of the power compensation power based on the state of charge of the energy storage device. Based on the vehicle's motion state, predict whether braking energy will be generated soon; if so, determine the final correction amount for the kinetic energy compensation power. ; Step S43: Using a weighted aggregation model of multidimensional power components, the initial output power of the range extender is obtained. : （7） in: : Dynamic weighting factor, dynamically determined based on the current vehicle's operating mode; In driving mode, increase Prioritize economic efficiency; under power shortage conditions, increase To force the maintenance of power balance; Step S44, Nonlinear safety trimming based on machine condition: Introduce a decay function based on the temperature and physical limits of the range extender to determine the power generation limit of the range extender. : （8） in: This represents the upper limit of the power generation capacity of the range extender. This is the attenuation steepness coefficient; The set alarm threshold for the engine cooling temperature of the range extender; This refers to the engine cooling temperature of the range extender. Step S45, Electrical safety constraints on bus voltage and current: （9） in: For safety clamping power; The real-time maximum allowable charging power of the energy storage device; This refers to the real-time current of the energy storage device. The electrical safety factor is adjusted to balance the power of the vehicle and the protection requirements of the energy storage device.

6. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 5, characterized in that, The method for calculating the final correction amount of the power compensation power is as follows: Step A1, using formula (10), based on the vehicle's speed Dynamically correct the basic balance value of the state of charge of energy storage devices The target value of the state of charge of the energy storage device is obtained. Using formula (11), the real-time state of charge deviation of the energy storage device is obtained. ; （10） （11） in: This is the speed-based feedforward coefficient, which is negative and used to significantly reduce the target value of the state of charge at high vehicle speeds. To make the target value of the state of charge Significantly lower than the basic equilibrium value of the state of charge This reduces the SOC offset correction of the range extender, increases the power supply of the energy storage device, and reserves space for the energy recovered during braking in subsequent control cycles to be stored in the energy storage device. This represents the actual state of charge of the energy storage device. A positive value indicates that the energy storage device is short of power, and the power correction caused by the SOC offset needs to be increased so that the output power of the range extender can supply power to the energy storage device, so that the energy storage device's power level approaches the target value of the state of charge. A negative value indicates that the energy storage device has excess power, and the power correction caused by the SOC offset needs to be reduced so that the energy storage device can drive the drive device to discharge in the next control cycle, so that the energy storage device's power level approaches the target value of the state of charge. Step A2, set the real-time state of charge deviation value of the energy storage device. By performing a nonlinear gain mapping that introduces dead-zone adjustment, the initial correction amount of the power compensation is obtained. : （12） Where: δ is the preset dead zone threshold for power correction, which is a positive value; This is the nonlinear gain coefficient for power replenishment when the energy storage device is short of power. This is the nonlinear gain coefficient for load reduction when the energy storage device has excess power. It is a non-linear power exponent; non-linear power exponent The introduction of this makes the correction intensity grow non-linearly with the power deviation; when the actual state of charge of the energy storage device is severely depleted, the correction amount of the basic demand power of the range extender increases in a stepwise manner. By charging the energy storage device through the range extender, the actual state of charge of the energy storage device is forced to return to the normal state, preventing the energy storage device from being over-discharged. Step A3, power smoothing based on a first-order inertial element: Initial correction amount of power compensation Converted into a power compensation smoothing correction amount with a smooth slope. : （13） （14） in: It is a smoothing factor; To control the scale of the cycle; As a time constant, it is dynamically adjusted according to the current driving mode. In economy mode, it increases. To ensure smoother changes in the operating conditions of the range extender and reduce fuel consumption; in power mode, reduce To accelerate the power follow-up response of the range extender and prevent over-discharge of the energy storage device; To control the cycle Power compensation smoothing correction amount at that time; Step A4, power limiting of the coupled battery state of health (SOH): Based on the actual charging and discharging capacity of the energy storage device, adjust the power compensation smoothing amount. Power limiting is performed to obtain the final correction amount of the power compensation power. : (15) （16） （17） in: This represents the current actual charging power capacity limit of the energy storage device. This represents the current actual discharge power capacity limit of the energy storage device. This represents the original charging power limit of the energy storage device when it is not in use. This represents the initial discharge power limit of the energy storage device in its unused state. This represents the current health status of the energy storage device; 1.0 indicates that it is not in use, and 0.8 indicates that it is nearing the end of its lifespan. This is the aging sensitivity coefficient; as energy storage devices age, Reduce and automatically compress the maximum power amplitude allowed for the range extender to recharge or discharge power; This is the attenuation scaling factor; Step A5: Perform secondary calibration of coupling internal resistance compensation using formula (18): （18） in: This is the upper limit safety red line value for bus voltage; This is the open-circuit voltage of the energy storage device; This is the actual value of the bus voltage; To take into account the internal resistance of the energy storage device after aging; If formula (18) is not satisfied, the calibration result is abnormal, triggering the perception-limited mode in the fault-tolerant degradation control strategy of the fault classification.

7. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 5, characterized in that, The final correction amount of the kinetic energy compensation power The calculation method is as follows: Step B1: Obtain driving speed based on multi-sensor motion state monitoring. acceleration Road gradient θ and real-time brake pedal opening And estimate the estimated value of the vehicle mass m. est ; Step B2: Define the regenerative braking potential power based on the vehicle's current kinetic energy change trend. : （19） in: It is the acceleration due to gravity; The overall efficiency factor for regenerative braking; Step B3, Dynamic feedforward compensation based on braking intent: Rate of change of brake pedal Identify the driver's intention to decelerate and brake, and generate an initial correction for the kinetic energy compensation power. : （20） （21） in: The acceleration threshold; As a slope predictor; This represents the maximum permissible opening of the brake pedal. This is the braking intensity sensitivity coefficient, or the slope prediction gain; Step B4, calibrating the output of kinetic energy compensation: （22） in: This represents the maximum permissible power reduction slope for the range extender. This is the final correction amount for kinetic energy compensation power.

8. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, Step S5 includes: Step S51, Dual-criteria steady-state determination: (1) Calculate the safety clamping power obtained from the control cycle k The target output power obtained from the previous control cycle k-1 absolute deviation value : （23） (2) Establish a dual-criteria steady-state determination logic: Condition 1, Power Deviation Criterion: ; in: The preset power lockout threshold; Condition 2, Time Duration Criterion: ; Where: t stable The cumulative time in the time domain for the current Hold command to enter the steady-state evaluation window; Maximum allowed lock time; (3) Logical branch execution path: Steady-state maintenance mode, Hold: If both of the above conditions are met, the current target output power is forcibly set to equal the target output power at the previous moment, i.e.: ; Steady-state breakout pattern: If ,or Release the Hold lock state and update the target output power: Simultaneously reset the timer. Prepare to enter the next decision loop; (4) Anti-frequent jump design based on hysteresis comparator: To avoid repeated switching Hold lock state, the introduction of hysteresis coefficient β hold , the value of 0.8~0.95; Threshold for exiting the Hold state: ; The threshold for re-entering the Hold state: ; Step S52, based on nonlinear slope constraint under dynamic operating conditions; （24） （25） （26） in: The power-up slope; To reduce the power slope; This refers to the real-time rotational speed of the range extender. This refers to the engine cooling temperature of the range extender. Operating mode; This represents the final output power of the range extender.

9. The intelligent dynamic power following control method for a range-extended electric vehicle based on a domain controller according to claim 1, characterized in that, The fault-tolerant degradation control strategy for fault classification includes three levels of fault-tolerant degradation logic: ① Level 1 Degradation, Sensing-Limited Mode: When abnormal data from energy storage devices or the environment is detected, the sensing-limited mode is triggered; in the sensing-limited mode, the base demand power is locked to the last base demand power before the sensing limitation; the bus voltage sensitivity coefficient is actively increased. This increases the bus voltage correction term, which in turn ultimately boosts the safety clamping power. ② Level 2 Degradation, Aircraft Risk Mode: When the engine cooling temperature of the range extender is detected... When lost, the aforementioned maintenance risk mode is triggered; In the aforementioned maintenance risk mode, a maximum safety power limiting output strategy is implemented to limit the safety clamping power to below 50% of the range extender's rated power, preventing mechanical damage to the range extender due to thermal management runaway. ③ Level 3 Degradation, Limp Mode: When all power-related data of the drive unit is detected to be lost, the limp mode is triggered; in the limp mode, the range extender operates smoothly at a fixed optimal efficiency point power, only used to maintain basic power to allow the vehicle to slowly stop in a safe area.

10. A smart dynamic power following control system for a range-extended electric vehicle based on a domain controller, characterized in that, include: The multi-source parameter fusion sensing module is used to collect vehicle and environmental status data in real time and synchronously. After filtering, spatiotemporal alignment and fusion, it generates vehicle and environmental status feature vectors. The intent prediction and real-time total demand power prediction module is used to obtain the physical demand power by using the vehicle longitudinal dynamics model based on the vehicle and environmental state feature vectors; and to calculate the real-time total demand power based on the physical demand power using the real-time total demand power calculation model. The basic demand power mapping module is used to identify the current operating conditions of the vehicle, including normal operating conditions and sensor failure conditions, based on the vehicle and environmental state feature vectors. The sensor failure condition is based on a fault-tolerant degradation control strategy with fault classification. The driving intention is predicted based on the rate of change of the real-time total demand power; based on the predicted driving intention, a dynamic weight mapping is performed between the preset economic optimal operating curve and the transient response operating curve to determine the base demand power of the range extender that matches the driving intention. The bus voltage feedback correction and multi-dimensional power deviation correction module incorporates a bus voltage sensitivity coefficient into the basic power requirement of the range extender. Bus voltage feedback correction and multi-dimensional power deviation correction; wherein, the bus voltage sensitivity coefficient Adjustments should be made based on the vehicle's current operating conditions. The multi-objective decision fusion module and the thermal safety trimming module are used to obtain the safety clamping power of the range extender after weighted aggregation of multi-dimensional power components and safety constraint limitation; The operating condition fluctuation filtering and Hold logic module is used to obtain the steady-state target output power of the range extender based on the safety clamping power, after adopting a dual-criteria steady-state determination and a nonlinear slope limit based on dynamic operating conditions. The final instruction synthesis and smooth delivery module performs adaptive rate of change limitation on the steady-state target output power of the range extender to obtain an output instruction, which is then sent to the range extender to adjust its output power. When the output power of the range extender changes, the energy storage device acts as a power buffer to perform peak shaving and valley filling to maintain the dynamic balance of the vehicle's energy flow.

Images