A method and device for joint control of the power system of a new energy vehicle

Abstract

This invention discloses a method and apparatus for joint control of the power system of new energy vehicles, belonging to the field of joint control technology for electric vehicles. It generates a comprehensive state vector by collecting vehicle operating status, driving environment status, and power system status, and generates multiple candidate joint control sequences that have not yet been issued for execution. These candidate joint control sequences are then input into the constraint mirror ports corresponding to the power subsystems for pre-screening, generating pre-screening constraint information, constructing a constraint propagation topology, and calculating constraint propagation costs to determine the target candidate joint control sequence. Finally, the joint control vector of the current control cycle is converted into actual joint control commands and issued for execution. This method can identify constraint propagation risks between multiple power subsystems before actual execution, reducing the probability of multiple power subsystems simultaneously exceeding constraint limits, and improving the safety and smoothness of joint control.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle joint control technology, and more specifically, to a joint control method and device for the power system of a new energy vehicle. Background Technology

[0002] Electric vehicles, especially hybrid, range-extended, or regenerative braking vehicles, typically require the simultaneous coordination of multiple powertrain subsystems, including the battery system, drive motor system, braking system, and range extender generator system. Existing powertrain control schemes often base their decisions on vehicle power demand, battery state of charge (SOC), drive motor efficiency, fuel consumption, regenerative braking recovery capability, or thermal management status. They employ rule-based control, energy management control, or predictive control methods to allocate drive motor output torque, battery charging / discharging power, regenerative braking torque, mechanical braking compensation, and range extender generator power. While these schemes can achieve energy optimization, SOC maintenance, or regenerative braking to some extent, their control focus is usually on ensuring that the current control variables meet the constraints of a single powertrain subsystem or that the overall energy allocation meets a preset optimization objective.

[0003] In actual vehicle operation, there are clear constraint linkages between different power subsystems. When the allowable discharge power of the power battery decreases, the output torque of the drive motor will be limited, and the range extender generator system may need to increase its power generation. When the allowable charging power of the power battery decreases, the regenerative braking torque needs to be reduced, and the mechanical braking compensation needs to be increased. When the output torque is limited due to the increased temperature of the drive motor, the vehicle's power demand will be further transmitted to the power battery system or the range extender generator system. When the braking thermal load increases, leading to a decrease in the mechanical braking compensation capability, the regenerative braking control quantity and the power battery charging constraint will also be affected. Existing control schemes, when generating control sequences, usually only treat the above constraints as boundary conditions or penalty terms for the current power subsystem. They lack a mechanism to pre-examine the real-time constraints of each power subsystem for multiple candidate joint control sequences before actually issuing control commands, and they also lack a structured expression for propagating the constraint exceedance of one power subsystem to other power subsystems.

[0004] Therefore, existing technologies are prone to situations where candidate control sequences, while seemingly meeting power or braking requirements at the vehicle level, are weakened or truncated by real-time constraints of a particular powertrain subsystem during actual execution. For example, after the target torque of the drive motor is reduced by the actual torque limit, the vehicle's power demand cannot be compensated in a timely manner; after the regenerative braking control quantity is limited by the allowable charging power of the power battery, the mechanical braking compensation quantity suddenly increases; after the power change rate of the range extender generator system is limited, the power battery discharge power bears a greater load in a short period of time. These problems can lead to inconsistencies between the actual joint control command and the expected control result, causing constraint conflicts, abrupt changes in control quantities, or frequent recalculations in multiple powertrain subsystems within the same control cycle, thereby affecting vehicle power response, braking smoothness, and powertrain system safety. Therefore, a powertrain joint control method is needed that can identify the constraint propagation risk of candidate joint control sequences before actual execution and select the target control sequence based on the constraint propagation risk. Summary of the Invention

[0005] This invention aims to solve the problem that in the joint control process of existing new energy vehicle power systems, candidate control commands usually only expose real-time constraint conflicts of the power subsystems after they are actually issued. This leads to the difficulty in timely transmitting the constraint effects of a power subsystem to other power subsystems after a constraint exceeds the limit, which in turn causes sudden changes in the control quantities of multiple power subsystems, repeated corrections, or inconsistencies between the actual execution results and the expected control results.

[0006] To address the aforementioned problems, in a first aspect, this invention provides a joint control method for the powertrain system of new energy vehicles. This method generates a comprehensive state vector by collecting data on vehicle operating status, driving environment status, and powertrain system status. Based on this comprehensive state vector, it generates multiple candidate joint control sequences that have not yet been issued for execution. Subsequently, the candidate joint control sequences are input into the constraint mirror ports corresponding to at least two powertrain subsystems, and each joint control vector is pre-screened without outputting execution signals. Then, based on the pre-screening results, pre-screening constraint information is generated, a constraint propagation topology is constructed, and the constraint propagation cost is calculated. Finally, based on the constraint propagation cost, a target candidate joint control sequence is determined, and the joint control vector of the current control cycle in the target candidate joint control sequence is converted into a real joint control command for execution.

[0007] Specifically, the comprehensive state vector is formed by the vehicle operating state, the driving environment state, and the powertrain state. The vehicle operating state may include vehicle speed, accelerator pedal opening, brake pedal opening, and wheel slip ratio; the driving environment state may include road gradient, estimated road adhesion coefficient, and predicted future road gradient; the powertrain state may include the power battery state, drive motor state, braking state, and range extender generator state. Through this state information, the vehicle controller can simultaneously determine the driver's power demand, changes in road load, braking demand, and the real-time availability of each powertrain subsystem.

[0008] Furthermore, the candidate joint control sequence includes multiple joint control vectors within the current control period and subsequent predictive control periods, each joint control vector including control quantities for at least two power subsystems. The power subsystems may include at least two of the following: a power battery system, a drive motor system, a braking system, and a range extender generator system. The joint control vector may include at least two of the following: a target torque for the drive motor, a target charging power for the power battery, a target discharging power for the power battery, a target torque for regenerative braking, a mechanical braking compensation amount, and a target power for the range extender generator. Therefore, the candidate joint control sequence can reflect the trends in power distribution, braking distribution, and energy compensation within the current control period and subsequent predictive control periods in the time dimension.

[0009] The vehicle controller generates a first candidate joint control sequence based on vehicle speed, accelerator pedal opening, and vehicle drive requirements to improve the speed at which the drive motor control quantity follows the vehicle's target power demand. It generates a second candidate joint control sequence based on at least two of the following: power battery status, drive motor status, braking status, driving environment status, and range extender system status. This allows for staggered allocation of critical control quantity peaks in at least two power subsystems to different control cycles. A third candidate joint control sequence is generated based on the following: braking status, driving environment status, and power battery status. This forms a joint control vector including regenerative braking control and power battery charging control during braking or downhill driving. Finally, a fourth candidate joint control sequence is generated based on at least one of the following: power battery status, drive motor status, range extender system status, and braking status. This reduces the power, torque, or braking force variation of the corresponding power subsystem in subsequent predictive control cycles when the temperature of the corresponding power subsystem reaches or exceeds a preset temperature warning threshold, or when the thermal load reaches or exceeds a preset thermal load warning threshold.

[0010] Furthermore, the constraint mirror port is a constraint judgment interface used for pre-screening candidate joint control sequences. The constraint mirror port calls the real-time constraint judgment logic of the corresponding powertrain subsystem, but does not output execution signals to the powertrain subsystem actuators. The vehicle controller compares the joint control vector of each control cycle in the candidate joint control sequence with the allowable control range of the corresponding powertrain subsystem in the corresponding control cycle; when a certain control quantity exceeds the allowable control range of the corresponding powertrain subsystem, it determines the powertrain subsystem identifier corresponding to that powertrain subsystem, and determines the constraint over-limit strength according to the magnitude of the control quantity exceeding the allowable control range; it determines the prediction cycle number of the first constraint over-limit occurrence as the constraint over-limit cycle; and then, according to the preset powertrain subsystem constraint linkage relationship, it determines the constraint propagation object identifier corresponding to the powertrain subsystem where the constraint over-limit occurred.

[0011] Furthermore, the pre-screening constraint information includes a power subsystem identifier, constraint over-limit strength, constraint over-limit period, and constraint propagation object identifier. The power subsystem identifier characterizes the power subsystem where a constraint over-limit occurs; the constraint over-limit strength characterizes the normalization degree to which the corresponding control quantity exceeds the allowable control range; the constraint over-limit period characterizes the predictive control period for the first occurrence of a constraint over-limit in the corresponding power subsystem; and the constraint propagation object identifier characterizes the associated power subsystems affected by the constraint over-limit. Through the above pre-screening constraint information, the real-time constraint states originally scattered across various power subsystems can be organized into a data structure usable for propagation analysis.

[0012] Furthermore, after generating the pre-qualification constraint information, the alternative control range of the corresponding powertrain subsystem can be output through the constraint mirror port. Based on the alternative control range, the vehicle controller performs amplitude limiting and smoothing corrections on the corresponding control quantities in the candidate joint control sequences where constraint limits have been exceeded. Then, based on the constraint propagation object identifier, the compensation portion formed by the amplitude limiting of the corresponding control quantity is allocated to the associated powertrain subsystem with an alternative control range margin. The corrected candidate joint control sequence is re-input into the corresponding constraint mirror port for pre-qualification. The corrected candidate joint control sequence that passes the pre-qualification becomes the updated candidate joint control sequence and participates in subsequent constraint propagation cost calculations.

[0013] Furthermore, the vehicle controller constructs a constraint propagation topology based on the powertrain subsystem identifier and the constraint propagation object identifier. This constraint propagation topology uses each powertrain subsystem as a node, and the relationship from the powertrain subsystem experiencing constraint exceedance to the powertrain subsystem corresponding to the constraint propagation object identifier is used as the directed propagation relationship. Through this constraint propagation topology, constraint influence relationships such as the transmission of power demand to the power battery system or range extender system after the drive motor output torque is limited, and the transmission of braking demand to mechanical braking compensation after the power battery's allowable charging power is limited, can be expressed. Further, the vehicle controller determines the constraint propagation cost corresponding to each candidate joint control sequence based on the constraint exceedance strength, constraint exceedance period, and constraint propagation topology. In one embodiment, the constraint propagation cost is calculated according to the following formula: The constraint propagation cost corresponding to each candidate joint control sequence: ; in, Indicates the first The constraint propagation cost corresponding to each candidate joint control sequence; Indicates the first The set of directed propagation relations in the constraint propagation topology corresponding to each candidate joint control sequence; The constraint is represented by the first The propagation of the first dynamic subsystem to the second One power subsystem; Indicates the first The first dynamic subsystem is related to the first Constraint propagation coupling coefficients of each dynamic subsystem; Indicates by the first The first power subsystem in the first The maximum value of the normalized constraint overlimit strength for each predictive control cycle under each candidate joint control sequence; This indicates that when the current control cycle is the 0th cycle, the 1st cycle... The first power subsystem in the first The prediction period number in which the first constraint exceedance occurs under a candidate joint control sequence.

[0014] When there is no directed propagation relationship between different dynamical subsystems in the constraint propagation topology, but a certain dynamical subsystem itself experiences constraint exceedance, the dynamical subsystem experiencing constraint exceedance can be set as a self-loop computational term. and order The ontological exceedance risk is calculated using the same constraint propagation cost formula. The self-loop calculation term is used to characterize the constraint exceedance risk of the dynamic subsystem itself, but does not characterize the external constraint propagation relationship between different dynamic subsystems.

[0015] Furthermore, the vehicle controller determines the target candidate joint control sequence based on the constraint propagation cost corresponding to each candidate joint control sequence. Specifically, candidate joint control sequences with constraint propagation costs lower than a preset cost threshold can be identified as optional candidate joint control sequences. When multiple optional candidate joint control sequences exist, the target candidate joint control sequence is determined from these sequences based on the vehicle's target power requirements, powertrain subsystem constraint margins, and ride comfort evaluation results. When the constraint propagation costs corresponding to each candidate joint control sequence are not lower than the preset cost threshold, a safety derating control sequence is generated according to a preset safety derating strategy. This safety derating control sequence is then input into the constraint mirror port for pre-screening. If the safety derating control sequence meets the preset safety constraint conditions, it is identified as the target candidate joint control sequence.

[0016] Furthermore, after the target candidate joint control sequence is determined, the vehicle controller converts the joint control vector of the current control cycle in the target candidate joint control sequence into a real joint control command and sends it to the actual execution port of the corresponding powertrain subsystem. The real joint control command may include at least two of the following: drive motor target torque command, power battery target charging power command, power battery target discharging power command, regenerative braking target torque command, mechanical braking compensation command, and range extender generator target power command. Upon receiving the real joint control command, the actual execution port of the corresponding powertrain subsystem controls the power battery system, drive motor system, braking system, or range extender generator system to execute the corresponding control quantity.

[0017] Furthermore, after the actual joint control command is sent to the actual execution port of the corresponding powertrain subsystem, the vehicle controller can also collect actual execution feedback. This feedback characterizes whether the corresponding powertrain subsystem has experienced an actual constraint overrun. When the actual execution feedback indicates that a powertrain subsystem has experienced an actual constraint overrun, and the corresponding pre-approval constraint information does not contain the identifier of that powertrain subsystem, or does not contain an identifier with that powertrain subsystem as the constraint propagation object, the constraint overrun strength correction coefficient for that powertrain subsystem is increased. Conversely, when the corresponding pre-approval constraint information contains the identifier of a powertrain subsystem, and the actual execution feedback indicates that the powertrain subsystem has not experienced an actual constraint overrun, the constraint overrun strength correction coefficient for that powertrain subsystem is decreased. Based on the constraint overrun strength correction coefficient, the vehicle controller adjusts the normalized constraint overrun strength of the corresponding candidate joint control sequence in the next control cycle. Or, the constraint propagation coupling coefficient between corresponding dynamic subsystems. Corrections are then made. As a result, the constraint propagation cost in subsequent control cycles can gradually adapt to the actual operating conditions of the vehicle, the aging state of the powertrain subsystem, and environmental changes.

[0018] Secondly, the present invention also provides a power system joint control device for new energy vehicles. The power system joint control device includes a state acquisition module, a candidate sequence generation module, a constraint pre-examination module, a pre-examination information generation module, a propagation cost determination module, a target sequence determination module, and a real command generation module. The state acquisition module is used to acquire the vehicle operating state, driving environment state, and power system state, and generate a comprehensive state vector; the candidate sequence generation module is used to generate multiple candidate joint control sequences that have not yet been issued for execution based on the comprehensive state vector; the constraint pre-examination module is used to input the candidate joint control sequences into the constraint mirror ports corresponding to at least two power subsystems, and pre-examine each joint control vector without outputting execution signals; the pre-examination information generation module is used to generate pre-examination constraint information based on the pre-examination results; the propagation cost determination module is used to construct a constraint propagation topology based on the power subsystem identifier and the constraint propagation object identifier, and determine the constraint propagation cost; the target sequence determination module is used to determine the target candidate joint control sequence based on each constraint propagation cost; the real command generation module is used to convert the joint control vector of the current control cycle in the target candidate joint control sequence into a real joint control command and send it to the real execution port of the corresponding power subsystem.

[0019] In one implementation, the powertrain joint control device can be implemented by a processor, a memory, and a communication interface in the vehicle controller. The memory stores a program for executing the aforementioned powertrain joint control method; when the processor executes the program, it completes state acquisition, candidate joint control sequence generation, constraint mirror port pre-screening, pre-screening constraint information generation, constraint propagation topology construction, constraint propagation cost calculation, target candidate joint control sequence determination, actual joint control command generation, and actual execution feedback correction; the communication interface is used for data interaction with the power battery controller, drive motor controller, brake controller, and range extender controller.

[0020] Compared with the prior art, the present invention has the following beneficial effects: In this invention, a comprehensive state vector is generated by collecting vehicle operating status, driving environment status, and power system status. Based on this comprehensive state vector, multiple candidate joint control sequences that have not yet been issued for execution are generated. This allows the vehicle controller to simultaneously consider driving demand, braking demand, road environment, and the real-time capabilities of each power system subsystem within the current control cycle and subsequent predictive control cycles. Furthermore, each candidate joint control sequence is input into the constraint mirror ports corresponding to at least two power system subsystems for pre-screening. The constraint mirror ports call the constraint judgment logic of the corresponding power system subsystem without outputting an execution signal. This allows for advance determination of whether each joint control vector will cause constraint overruns in the power battery system, drive motor system, braking system, or range extender system without changing the actual operating state of the vehicle. Pre-screening constraint information, including power system identifier, constraint overrun strength, constraint overrun period, and constraint propagation object identifier, is generated. This processing method allows candidate joint control sequences to undergo constraint checks close to actual execution conditions before actual execution, reducing the risk of weakened power output, sudden braking compensation changes, or control command recalculation caused by discovering constraint overruns only during actual execution.

[0021] This invention also constructs a constraint propagation topology corresponding to candidate joint control sequences based on the powertrain subsystem identifier and constraint propagation object identifier, and determines the constraint propagation cost by combining the constraint over-limit strength and constraint over-limit period, thereby transforming the originally implicit constraint linkage relationship between powertrain subsystems into a calculable and comparable risk indicator. The vehicle controller can determine the target candidate joint control sequence based on the constraint propagation cost, and convert the joint control vector of the current control cycle in the target candidate joint control sequence into a real joint control command and send it to the actual execution port of the corresponding powertrain subsystem. Through this processing chain, candidate joint control sequences with lower constraint over-limit strength, later constraint occurrence time, and smaller constraint propagation impact can be preferentially selected, reducing the risk of multiple powertrain subsystems simultaneously experiencing constraint over-limits in the same control cycle. At the same time, the actual execution feedback can also be used to correct the normalized constraint over-limit strength or constraint propagation coupling coefficient in subsequent control cycles, so that the constraint propagation cost gradually adapts to the actual operating state of the vehicle, thereby improving the safety, stability, and control smoothness of the joint control of the powertrain system of new energy vehicles. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 A flowchart of a joint control method for an electric vehicle powertrain system; Figure 2 Flowchart of pre-approval process for constrained mirror ports; Figure 3To ensure accurate feedback and correction of the sub-flowchart. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0025] like Figure 1-3 As shown, this embodiment provides a joint control method for the power system of a new energy vehicle, applicable to new energy vehicles including at least two power subsystems: a power battery system, a drive motor system, a braking system, and a range extender generator system. The power battery system provides discharge power to the drive motor system and receives charging power during regenerative braking. The drive motor system outputs drive torque according to the vehicle's driving needs and outputs regenerative torque during braking or downhill driving. The braking system coordinates regenerative braking control and mechanical braking compensation based on braking demand intensity. The range extender generator system outputs power generation based on the power battery's SOC, the vehicle's power demand, and the range extender generator system's temperature. The vehicle controller is communicatively connected to the power battery controller, drive motor controller, braking controller, and range extender generator controller. The vehicle controller generates candidate joint control sequences, calls constraint mirror ports for pre-screening, generates pre-screening constraint information, constructs constraint propagation topology, calculates constraint propagation costs, determines target candidate joint control sequences, and converts the joint control vector of the current control cycle into actual joint control commands before sending them to the actual execution ports of the corresponding power subsystems.

[0026] In a specific implementation, the vehicle controller uses 100ms as the current control cycle and the next 10 control cycles as the subsequent predictive control cycles. At the arrival of each control cycle, the vehicle controller collects the vehicle's operating status, driving environment status, and powertrain status, and combines the collected state variables into a comprehensive state vector according to a preset order. The vehicle's operating status includes at least one of the following: vehicle speed, accelerator pedal opening, brake pedal opening, and wheel slip ratio. Vehicle speed is used to determine the vehicle's current speed; accelerator pedal opening represents the driver's acceleration intention; brake pedal opening represents the driver's braking intention; and wheel slip ratio is used to determine whether the drive control or braking control variables are likely to cause tire slippage. The driving environment status includes at least one of the following: road gradient, estimated road adhesion coefficient, and future road gradient prediction information. Road gradient is used to determine changes in vehicle load when going uphill or downhill; the estimated road adhesion coefficient is used to determine the available margin of drive torque or braking torque; and the future road gradient prediction information is used to predict the changing trend of vehicle power demand or braking demand in subsequent predictive control cycles.

[0027] The power system status includes at least one of the following: power battery status, drive motor status, braking status, and range extender system status. Power battery status includes at least one of the following: power battery SOC, power battery temperature, allowable charging power, and allowable discharging power. Power battery SOC is used to determine the remaining charge of the power battery; power battery temperature is used to determine the thermal safety status of the power battery; allowable charging power is used to determine whether the regenerative braking energy can be received by the power battery; and allowable discharging power is used to determine whether the power required by the drive motor system can be provided by the power battery system. Drive motor status includes at least one of the following: drive motor speed, drive motor temperature, allowable output torque, and allowable regenerative torque. Drive motor speed is used to determine the efficiency range and torque output capability of the drive motor; drive motor temperature is used to determine the thermal load of the drive motor; allowable output torque is used to determine the upper limit of the drive torque; and allowable regenerative torque is used to determine the regenerative braking capability. Braking status includes at least one of the following: braking demand intensity, regenerative braking availability, mechanical braking compensation capability, and braking thermal load status. Range extender system status includes at least one of the following: current generating power, range extender system temperature, and allowable generating power.

[0028] When generating the comprehensive state vector, the vehicle controller normalizes state variables of different dimensions, ensuring that vehicle speed, pedal opening, road gradient, battery SOC, battery temperature, drive motor temperature, allowable power, allowable torque, braking thermal load status, and range extender system temperature can all participate in the same control decision. Normalization can be achieved using preset upper and lower limit mapping methods. For example, battery SOC can be mapped from 0 to 100% to 0 to 1, pedal opening from 0 to 100% to 0 to 1, battery temperature from a preset safe temperature range to 0 to 1, and drive motor allowable output torque from peak torque to 0 to 1. After normalization, the vehicle controller can determine vehicle power demand, braking demand, allowable control range of each power subsystem, and constraint margin of each power subsystem based on the same comprehensive state vector.

[0029] The vehicle controller generates multiple candidate joint control sequences that have not yet been executed based on the integrated state vector. Each candidate joint control sequence includes multiple joint control vectors for the current control cycle and subsequent predictive control cycles, and each joint control vector includes control quantities for at least two powertrain subsystems. In this embodiment, a joint control vector includes at least two of the following: target torque of the drive motor, target discharge power of the power battery, target charging power of the power battery, target torque of regenerative braking, mechanical braking compensation, and target power of the range extender generator. When generating candidate joint control sequences, the vehicle controller first determines the target power demand of the vehicle based on vehicle speed, accelerator pedal opening, road gradient, and future road gradient prediction information. Then, it determines the target braking demand based on brake pedal opening, road gradient, wheel slip ratio, and road adhesion coefficient estimates. Simultaneously, it determines the constraint margin of each powertrain subsystem for the current control cycle and subsequent predictive control cycles based on the allowable charging power of the power battery, the allowable discharge power of the power battery, the allowable output torque of the drive motor, the allowable regenerative torque of the drive motor, the mechanical braking compensation capability, and the allowable power generation. The target power demand, target braking demand, and constraint margins of each power subsystem jointly determine the magnitude and trend of the control quantity of each joint control vector in the candidate joint control sequence.

[0030] When the vehicle has an acceleration demand, the vehicle controller generates a first candidate joint control sequence based on the vehicle's operating state and the drive motor's state in the integrated state vector. This first candidate joint control sequence improves the speed at which the drive motor control input follows the vehicle's target power demand, determined by vehicle speed, accelerator pedal opening, and overall vehicle drive requirements. The vehicle controller first determines the driver's desired acceleration based on the accelerator pedal opening and vehicle speed, then determines the vehicle's target power demand by considering the road gradient. This target power demand is then converted into the drive motor's target torque and the battery's target discharge power. If both the battery's allowable discharge power and the drive motor's allowable output torque have sufficient margin, the first candidate joint control sequence ensures that the drive motor's target torque in the current control cycle quickly approaches the target torque corresponding to the vehicle's target power demand, and continues to adjust the drive motor's target torque according to the changing trend of the vehicle's target power demand in subsequent predictive control cycles. The process of changing from the vehicle's target power demand to the first candidate joint control sequence involves converting the driver's acceleration intention and road load into the target torque of the drive motor, then converting the target torque of the drive motor into the target discharge power of the power battery, and finally combining the target torque of the drive motor and the target discharge power of the power battery in the current control cycle and subsequent predicted control cycles into a joint control vector in chronological order.

[0031] When multiple powertrain subsystems of a vehicle are simultaneously approaching the upper limit of their permissible control range or the upper limit of their rate of change, the vehicle controller generates a second candidate joint control sequence based on at least two of the following states in the integrated state vector: battery state, drive motor state, braking state, driving environment state, and range extender system state. This second candidate joint control sequence is used to stagger the peak values ​​of critical control quantities in at least two powertrain subsystems across different control cycles, reducing the risk of multiple powertrain subsystems simultaneously exceeding constraint limits within the same control cycle. The critical control quantity is the control quantity approaching the upper limit of its corresponding permissible control range or the upper limit of its rate of change. For example, when the vehicle is accelerating uphill, the target torque of the drive motor is close to the permissible output torque of the drive motor, the target discharge power of the battery is close to the permissible discharge power of the battery, and the target power of the range extender is close to the permissible power generation power. In the second candidate joint control sequence, the vehicle controller moderately increases the target torque of the drive motor in the current control cycle, gradually increases the target power of the range extender in subsequent predictive control cycles, and correspondingly decreases the target discharge power of the battery after the target power of the range extender increases. The process of changing from multiple critical control variables to the second candidate joint control sequence involves first identifying control variables that are close to the upper limit of the allowable control range or the upper limit of the rate of change, and then allocating the peak values ​​of each control variable to different predictive control cycles based on the constraint margins and response speeds of each dynamic subsystem.

[0032] When the vehicle is braking or descending a slope, the vehicle controller generates a third candidate joint control sequence based on the braking state, driving environment state, and battery state in the integrated state vector. This third candidate joint control sequence is used to generate a joint control vector including regenerative braking control and battery charging control quantities during braking or downhill driving. The vehicle controller first determines the target braking demand based on brake pedal opening, vehicle speed, road gradient, and wheel slip ratio. Then, it determines the available regenerative braking capability based on the battery's allowable charging power, battery SOC, battery temperature, and the drive motor's allowable regenerative torque. If the battery's allowable charging power is high and the drive motor's allowable regenerative torque is sufficient, the vehicle controller increases the regenerative braking target torque in the third candidate joint control sequence and simultaneously sets a battery target charging power that matches the regenerative braking target torque. If the battery SOC is high, the battery temperature is high, or the battery's allowable charging power is low, the vehicle controller decreases the regenerative braking target torque and increases the mechanical braking compensation. The process of changing from braking demand to the third candidate joint control sequence involves converting the driver's braking intention and downhill load into target braking demand, and then allocating the target braking demand into regenerative braking target torque, power battery target charging power, and mechanical braking compensation amount.

[0033] When the temperature of a powertrain subsystem reaches or exceeds a preset temperature warning threshold, or when the thermal load reaches or exceeds a preset thermal load warning threshold, the vehicle controller generates a fourth candidate joint control sequence based on at least one of the following states in the integrated state vector: power battery state, drive motor state, range extender generator state, and braking state. This fourth candidate joint control sequence is used to reduce the magnitude of power variation, torque variation, or braking force variation of the corresponding powertrain subsystem in subsequent predictive control cycles. For example, when the drive motor temperature reaches the preset temperature warning threshold, the vehicle controller reduces the increase in the target torque of the drive motor in subsequent predictive control cycles and maintains stable vehicle power output through coordination between the target discharge power of the power battery and the target power of the range extender generator. When the braking thermal load reaches the preset thermal load warning threshold, the vehicle controller limits the rate of increase in mechanical braking compensation and increases the target regenerative braking torque when the allowable charging power of the power battery and the allowable regenerative torque of the drive motor meet the conditions. The process of changing from temperature or thermal load state to the fourth candidate joint control sequence involves first identifying the power subsystem with thermal risk, then reducing the abrupt change amplitude of the relevant control quantity of the power subsystem in the subsequent predictive control cycle, and using other power subsystems with constraint margins to undertake part of the power or braking demand.

[0034] After generating multiple candidate joint control sequences, the vehicle controller inputs each candidate joint control sequence into the constraint mirror ports corresponding to at least two powertrain subsystems. The constraint mirror ports are located in the software interfaces of the battery controller, drive motor controller, brake controller, and range extender controller. The constraint mirror ports are used to call the constraint judgment logic of the corresponding powertrain subsystem without outputting execution signals. The battery controller, drive motor controller, brake controller, and range extender controller all have constraint mirror ports and actual execution ports. After receiving the candidate joint control sequence, the constraint mirror port only reads the real-time constraints, allowable control range, and rate of change limits of the corresponding powertrain subsystem and outputs the preliminary review results. After receiving the actual joint control command, the actual execution port outputs control signals to the power converter, drive inverter, brake actuator, or range extender actuator. Because the constraint mirror port calls the same constraint judgment logic as the actual execution port, the candidate joint control sequence can undergo constraint checks consistent with the actual execution conditions without changing the actual operating state of the vehicle.

[0035] During pre-screening via the constraint mirror port, the vehicle controller compares the joint control vector of each control cycle in the candidate joint control sequence with the allowable control range of the corresponding powertrain subsystem in the corresponding control cycle. The allowable control range for the power battery system includes the allowable charging power range and the allowable discharging power range; the allowable control range for the drive motor system includes the allowable output torque range and the allowable regenerative torque range; the allowable control range for the braking system includes the allowable regenerative braking power range and the mechanical braking compensation range; and the allowable control range for the range-extended generator system includes the allowable generator power range and the generator power change rate range. When the control quantity corresponding to a certain powertrain subsystem in the joint control vector exceeds the allowable control range of that powertrain subsystem in the corresponding control cycle, the vehicle controller determines the powertrain subsystem identifier corresponding to the constraint exceeding the limit. The powertrain subsystem identifier can be represented by a preset number, for example, B for the power battery system, M for the drive motor system, R for the braking system, and G for the range-extended generator system. The vehicle controller can also use numbers to represent the power subsystem identifiers, for example, the power battery system is identified as 1, the drive motor system as 2, the braking system as 3, and the range extender generator system as 4.

[0036] The vehicle controller determines the constraint over-limit strength based on the magnitude by which the controlled variable exceeds the permissible control range. The constraint over-limit strength is a normalized value used to eliminate dimensional differences between different control variables. The constraint over-limit strength for power-related control variables is determined by the ratio between the excess power and the corresponding permissible power range width; for torque-related control variables, it is determined by the ratio between the excess torque and the corresponding permissible torque range width; and for braking force-related control variables, it is determined by the ratio between the excess braking force and the corresponding permissible braking force range width. The vehicle controller calculates the normalized constraint over-limit strength for the corresponding powertrain subsystem in each predictive control cycle and uses the maximum normalized constraint over-limit strength for the same powertrain subsystem across all predictive control cycles as the constraint over-limit strength in subsequent constraint propagation cost calculations.

[0037] The vehicle controller determines the constraint overshoot period based on the prediction period number of the first constraint overshoot. The current control period number is set to 0, the first prediction control period number after the current control period is set to 1, the second prediction control period number is set to 2, and so on. If the drive motor system experiences its first constraint overshoot in the second prediction control period, then the constraint overshoot period for the drive motor system under the corresponding candidate joint control sequence is 2. The smaller the constraint overshoot period, the earlier the corresponding candidate joint control sequence triggers constraint risk, and the higher the risk weight is assigned in the constraint propagation cost calculation by the vehicle controller.

[0038] The vehicle controller determines the constraint propagation object identifier corresponding to the powertrain subsystem identifier based on preset powertrain subsystem constraint linkage relationships. These powertrain subsystem constraint linkage relationships are stored in the vehicle controller's linkage relationship table, which records which related powertrain subsystems the control demand may propagate to when a powertrain subsystem experiences a constraint exceedance. For example, when the drive motor system's output torque is limited, the vehicle's target power demand will require compensation from the battery system and range extender system; therefore, the constraint propagation object identifier corresponding to the drive motor system identifier M can include B and G. When the battery system's allowable discharge power is limited, the drive motor system's output torque may need to be reduced, and the range extender system's target power may need to be increased; therefore, the constraint propagation object identifier corresponding to the battery system identifier B can include M and G. When the battery system's allowable charging power is limited, the regenerative braking target torque needs to be reduced, and the mechanical braking compensation amount needs to be increased; therefore, the constraint propagation object identifier corresponding to the battery system identifier B can include R. When the braking system's mechanical braking compensation capability is limited, the regenerative braking target torque and the battery's target charging power need to be redistributed; therefore, the constraint propagation object identifier corresponding to the braking system identifier R can include M and B. The constraint propagation object identifier can be stored using the same numbering method as the dynamic subsystem identifier, or it can be stored using bit identifiers. For example, a four-bit binary identifier can be used to correspond to B, M, R, and G respectively, with a value of 1 indicating that the corresponding dynamic subsystem belongs to the constraint propagation object.

[0039] The vehicle controller generates pre-qualification constraint information based on the pre-qualification results. This information includes the powertrain subsystem identifier, constraint over-limit strength, constraint over-limit period, and constraint propagation object identifier. The powertrain subsystem identifier indicates the powertrain subsystem where the constraint over-limit occurred; the constraint over-limit strength indicates the normalization degree to which the corresponding control quantity exceeds the allowable control range; the constraint over-limit period indicates the predicted period number of the first constraint over-limit occurrence for the corresponding powertrain subsystem; and the constraint propagation object identifier indicates the associated powertrain subsystems affected by the constraint over-limit. The pre-qualification constraint information can be stored in the order of candidate joint control sequence number, powertrain subsystem identifier, constraint over-limit strength, constraint over-limit period, and constraint propagation object identifier, enabling the vehicle controller to establish a corresponding constraint propagation topology for each candidate joint control sequence.

[0040] When generating pre-constraint information, the constraint mirror port also outputs the alternative control range of the corresponding power subsystem in the current state. The alternative control range characterizes the range of compensatory control quantities that the corresponding power subsystem can undertake without exceeding its real-time constraints. The alternative control range for the power battery system includes the allowable charging power range and the allowable discharging power range; for the drive motor system, it includes the allowable output torque range and the allowable regenerative torque range; for the braking system, it includes the allowable regenerative braking power range and the mechanical braking compensation range; and for the range-extended generator system, it includes the allowable generator power range. Based on the alternative control range, the vehicle controller performs amplitude limiting and smoothing corrections on the corresponding control quantities in the candidate joint control sequences where constraint limits have been exceeded. Amplitude limiting restricts the corresponding control quantity to the alternative control range, while smoothing correction adjusts the amplitude of control quantity changes between adjacent control cycles according to the rate of change of the corresponding power subsystem.

[0041] When the control quantity of a certain powertrain subsystem is limited, the vehicle controller determines the difference between the control quantity before and after limitation and uses this difference as a compensation portion. The vehicle controller searches for associated powertrain subsystems capable of undertaking this compensation portion based on the constraint propagation object identifier and allocates the compensation portion according to the alternative control range margin of each associated powertrain subsystem. If only one associated powertrain subsystem has an alternative control range margin, the vehicle controller allocates the compensation portion to that subsystem. If multiple associated powertrain subsystems have alternative control range margins, the vehicle controller allocates the compensation portion proportionally according to the size of the alternative control range margin of each subsystem; the larger the alternative control range margin, the more compensation portion is allocated. If the alternative control range margin of a certain associated powertrain subsystem is insufficient to undertake the allocated compensation portion, the vehicle controller continues to allocate the excess portion to other associated powertrain subsystems that still have alternative control range margins. After the compensation portion allocation, the vehicle controller re-enters the corrected candidate joint control sequence into the corresponding constraint mirror port for pre-screening. The revised candidate joint control sequence, after being reviewed again, will be used as the updated candidate joint control sequence in subsequent constraint propagation cost calculations.

[0042] The vehicle controller constructs a constraint propagation topology for each candidate joint control sequence based on the powertrain subsystem identifier and constraint propagation object identifier. The constraint propagation topology is represented by a directed graph, where each node corresponds to a powertrain subsystem, and each directed propagation relationship indicates that a constraint propagates from one powertrain subsystem to another. If the powertrain subsystem identifier in the pre-qualification constraint information is M, and the constraint propagation object identifiers include B and G, the vehicle controller generates a directed propagation relationship from the drive motor system to the power battery system and from the drive motor system to the range extender generator system in the constraint propagation topology of the corresponding candidate joint control sequence. If the powertrain subsystem identifier in the pre-qualification constraint information is B, and the constraint propagation object identifiers include R, the vehicle controller generates a directed propagation relationship from the power battery system to the braking system in the constraint propagation topology of the corresponding candidate joint control sequence. Through the constraint propagation topology, the vehicle controller can transform the constraint exceedance of a single powertrain subsystem into constraint influence relationships between multiple powertrain subsystems.

[0043] The vehicle controller is determined according to the following formula: The constraint propagation cost corresponding to each candidate joint control sequence: ; in, Indicates the first The constraint propagation cost corresponding to each candidate joint control sequence; Indicates the first The set of directed propagation relations in the constraint propagation topology corresponding to each candidate joint control sequence; The constraint is represented by the first The propagation of the first dynamic subsystem to the second One power subsystem; Indicates the first The first dynamic subsystem is related to the first Constraint propagation coupling coefficients of each dynamic subsystem; Indicates by the first The first power subsystem in the first The maximum value of the normalized constraint overlimit strength for each predictive control cycle under each candidate joint control sequence; This indicates that when the current control cycle is the 0th cycle, the 1st cycle... The first power subsystem in the first The prediction period number in which the first constraint exceedance occurs under a candidate joint control sequence.

[0044] In the above formula, It originates from the constraint propagation topology constructed based on the dynamic subsystem identifier and the constraint propagation object identifier; It originates from the pre-calibrated power subsystem constraint linkage relationship, and can also be updated based on the vehicle's historical execution feedback; The results are derived from the preliminary review of candidate joint control sequences by the constrained mirror port. This is derived from the prediction period number of the first constraint exceedance of the corresponding dynamic subsystem. (Constraint propagation coupling coefficient) For non-negative numbers, the constraint exceeds the strength. Non-negative numbers constrain over-limit periods Since the integers are non-negative, all terms on the right-hand side of the formula are non-negative scalars, and the summation result is also a non-negative scalar. This is related to the constraint propagation cost on the left-hand side. Maintain consistency in mathematical dimensions.

[0045] When the When there are directed propagation relationships between different powertrain subsystems in the constraint propagation topology corresponding to several candidate joint control sequences, the vehicle controller calculates the constraint propagation cost according to the above formula. The greater the constraint exceedance intensity, the greater the constraint propagation cost of the corresponding candidate joint control sequence; the greater the constraint propagation coupling coefficient, the greater the constraint propagation cost of the corresponding candidate joint control sequence; the smaller the prediction period number of the first constraint exceedance, the greater the constraint propagation cost of the corresponding candidate joint control sequence. Through this calculation method, the vehicle controller can prioritize the identification of candidate joint control sequences that occur in a short period of time, have high exceedance intensity, and are easy to propagate to other powertrain subsystems.

[0046] When the In the constraint propagation topology corresponding to the candidate joint control sequences, there are no directed propagation relationships between different powertrain subsystems. However, when a powertrain subsystem itself experiences constraint exceedance, the vehicle controller determines the constraint propagation cost of the corresponding candidate joint control sequence according to the intrinsic exceedance cost. The intrinsic exceedance cost is processed according to the same constraint propagation cost formula: the vehicle controller sets the powertrain subsystem experiencing constraint exceedance as a self-loop calculation item. And let the corresponding coupling coefficient The value is 1. The self-loop calculation term is used to calculate the risk of constraint exceedance within the dynamic subsystem itself. The self-loop calculation term does not indicate the existence of external constraint propagation relationships between different dynamic subsystems. Through the self-loop calculation term, even if a candidate joint control sequence does not cause constraint propagation in other dynamic subsystems, as long as there is constraint exceedance within the dynamic subsystem itself, the constraint propagation cost can still reflect the intrinsic exceedance risk. When the... When no constraint over-limit occurs in any of the dynamic subsystems, The corresponding item is not included in the constraint propagation cost.

[0047] The vehicle controller determines the target candidate joint control sequence based on the constraint propagation cost. The vehicle controller first selects candidate joint control sequences with constraint propagation costs below a preset cost threshold as optional candidate joint control sequences. When multiple optional candidate joint control sequences exist, the vehicle controller determines the power demand matching degree, power subsystem constraint margin, and ride comfort evaluation result of each optional candidate joint control sequence based on the integrated state vector and the joint control vector in each optional candidate joint control sequence. The power demand matching degree characterizes whether the target torque of the drive motor, the target discharge power of the power battery, and the target power of the range extender can meet the vehicle's target power demand; the power subsystem constraint margin characterizes the remaining margin between each control quantity and the corresponding allowable control range boundary; the ride comfort evaluation result characterizes whether the changes in the target torque of the drive motor, the target torque of regenerative braking, the mechanical braking compensation, and the target power of the range extender are smooth between adjacent control cycles. The vehicle controller preferentially selects candidate joint control sequences with lower constraint propagation costs, higher power demand matching degrees, larger power subsystem constraint margins, and better ride comfort evaluation results as target candidate joint control sequences.

[0048] When the constraint propagation cost corresponding to each candidate joint control sequence is not lower than a preset cost threshold, the vehicle controller generates a safety derating control sequence according to a preset safety derating strategy and inputs the safety derating control sequence into the constraint mirror port for pre-screening. The preset safety derating strategy includes at least one of limiting the rate of change of drive motor torque, limiting the power surge of the range extender generator system, limiting the surge of regenerative braking, and increasing the smooth compensation of mechanical braking. The preset safety constraints include that the control quantities of each power subsystem are within the corresponding allowable control range, the change amplitude of control quantities between adjacent control cycles is within the corresponding rate of change limit, the temperature of the power battery, the temperature of the drive motor, the braking thermal load state, and the temperature of the range extender generator system do not exceed the corresponding safety limits, and the vehicle's target braking demand can be met by both the regenerative braking control quantity and the mechanical braking compensation quantity. When the safety derating control sequence meets the preset safety constraints, the vehicle controller uses the safety derating control sequence as the target candidate joint control sequence.

[0049] After determining the target candidate joint control sequence, the vehicle controller converts the joint control vector of the current control cycle in the target candidate joint control sequence into actual joint control commands and sends them to the actual execution port of the corresponding powertrain subsystem. The actual joint control commands include at least two of the following: drive motor target torque command, power battery target charging power command, power battery target discharging power command, regenerative braking target torque command, mechanical braking compensation command, and range extender generator target power command. The power battery controller controls the energy flow of the power battery system according to the power battery target charging power command or the power battery target discharging power command; the drive motor controller controls the output torque or regenerative torque of the drive motor system according to the drive motor target torque command; the braking controller controls the execution of regenerative braking and mechanical braking according to the regenerative braking target torque command and the mechanical braking compensation command; and the range extender generator controller controls the power generation output of the range extender generator system according to the range extender generator target power command. Because the actual joint control commands originate from the target candidate joint control sequence that has been pre-screened through constraint mirror ports and filtered by constraint propagation costs, the risk of multiple powertrain subsystems simultaneously exceeding constraint limits during actual execution is reduced.

[0050] After the actual joint control command is sent to the actual execution port of the corresponding powertrain subsystem, the vehicle controller collects the actual execution feedback. The actual execution feedback characterizes whether the corresponding powertrain subsystem has experienced an actual constraint exceedance. The actual execution feedback includes at least one of the following: actual charging power, actual discharging power, power battery temperature, and power limit status from the battery controller; actual output torque, actual regenerative torque, drive motor temperature, and torque limit status from the drive motor controller; actual regenerative braking torque, mechanical braking compensation, and braking thermal load status from the brake controller; and actual power generation power, range extender system temperature, and power limit status from the range extender controller. The vehicle controller compares the actual execution feedback with the pre-qualified constraint information to correct the normalized constraint exceedance strength of the corresponding candidate joint control sequence in the next control cycle. Or the constraint propagation coupling coefficient between corresponding dynamic subsystems .

[0051] When actual execution feedback indicates that a powertrain subsystem has exceeded its constraints, and the corresponding pre-qualification constraint information does not include the identifier of that powertrain subsystem, or does not include an identifier that uses that powertrain subsystem as the constraint propagation object, the vehicle controller increases the constraint exceedance strength correction factor for that powertrain subsystem. After increasing the constraint exceedance strength correction factor, the vehicle controller calculates the normalized constraint exceedance strength for similar candidate joint control sequences in the next control cycle. If the pre-approved normalized constraint over-limit strength is increased, the same candidate joint control sequences will receive a higher cost in constraint propagation cost calculation. If the actual execution feedback indicates that the drive motor system experiences actual torque limitation, and the pre-approved constraint information does not include the drive motor system identifier, the vehicle controller will increase the normalized constraint over-limit strength corresponding to the drive motor system in the next control cycle. This reduces the probability that a high-torque candidate joint control sequence for a drive motor will be identified as the target candidate joint control sequence.

[0052] When the pre-qualification constraint information includes a powertrain subsystem identifier, and the actual execution feedback indicates that the powertrain subsystem has not experienced an actual constraint overrun, the vehicle controller reduces the constraint overrun intensity correction factor corresponding to that powertrain subsystem. After reducing the constraint overrun intensity correction factor, the vehicle controller calculates the normalized constraint overrun intensity for similar candidate joint control sequences in the next control cycle. At that time, the normalized constraint over-limit strength obtained from the pre-audit is lowered, so that the pre-audit results of the constraint mirror port gradually approach the actual execution feedback. Regarding the constraint propagation coupling coefficient... The vehicle controller makes corrections based on the discrepancy between the actual execution feedback and the pre-approval constraint information. If multiple actual execution feedbacks indicate that the first... After the first dynamic subsystem exceeds the constraint limit, the second... Individual power subsystems are also prone to exceeding actual constraints, thus requiring the vehicle controller to improve... If multiple real-world executions show that the feedback indicates the first... After the first dynamic subsystem exceeds the constraint limit, the second... If the individual powertrain subsystems are not significantly affected, the vehicle controller will reduce [the load]. Through the above correction process, the vehicle controller can gradually adapt the constraint propagation cost calculation to the actual operating state of the vehicle, the aging state of the power subsystem, and environmental changes.

[0053] In a specific hardware implementation, the powertrain joint control device for new energy vehicles includes a state acquisition module, a candidate sequence generation module, a constraint pre-examination module, a pre-examination information generation module, a propagation cost determination module, a target sequence determination module, and a real command generation module. The state acquisition module communicates with vehicle sensors, the power battery controller, the drive motor controller, the brake controller, and the range extender controller. It acquires the vehicle's operating state, the driving environment state, and the powertrain state, and generates a comprehensive state vector. The candidate sequence generation module generates multiple candidate joint control sequences that have not yet been issued for execution based on the comprehensive state vector. These candidate joint control sequences include joint control vectors for the current control cycle and subsequent predicted control cycles. The constraint pre-examination module inputs the candidate joint control sequences into the constraint mirror ports corresponding to at least two powertrain subsystems and pre-examines each joint control vector based on the real-time constraints of the corresponding powertrain subsystems without outputting execution signals. The pre-examination information generation module generates pre-examination constraint information based on the pre-examination results. This pre-examination constraint information includes the powertrain subsystem identifier, constraint over-limit strength, constraint over-limit period, and constraint propagation object identifier. The propagation cost determination module is used to construct the constraint propagation topology corresponding to each candidate joint control sequence based on the dynamic subsystem identifier and the constraint propagation object identifier, and to determine the constraint propagation cost of each constraint propagation topology. The target sequence determination module is used to determine the target candidate joint control sequence based on each constraint propagation cost. The real instruction generation module is used to convert the joint control vector of the current control cycle in the target candidate joint control sequence into a real joint control instruction and send it to the real execution port of the corresponding dynamic subsystem.

[0054] The powertrain joint control unit can be implemented using a processor, memory, and communication interface within the vehicle controller. The memory stores programs for performing state acquisition, candidate joint control sequence generation, constraint mirror port pre-screening, pre-screening constraint information generation, constraint propagation topology construction, constraint propagation cost calculation, target candidate joint control sequence determination, actual joint control command generation, and actual execution feedback correction. When the processor executes the programs in the memory, it reads the vehicle operating state, driving environment state, and powertrain state according to the current control cycle and subsequent predictive control cycles, generates a comprehensive state vector, and completes the generation, pre-screening, correction, cost calculation, and selection of candidate joint control sequences. The communication interface is used for data interaction with the power battery controller, drive motor controller, brake controller, and range extender controller. The power battery controller, drive motor controller, brake controller, and range extender controller each have constraint mirror ports and actual execution ports. The constraint mirror ports are used for pre-screening candidate joint control sequences, and the actual execution ports are used to receive and execute actual joint control commands.

[0055] After adopting the powertrain joint control method of this embodiment, the vehicle controller can determine in advance whether candidate joint control sequences will cause constraint overruns in the power battery system, drive motor system, braking system, or range extender generator system before the actual joint control command is issued, through constraint mirror ports; the vehicle controller can determine whether constraint overruns in one powertrain subsystem will propagate to other powertrain subsystems through pre-screening constraint information and constraint propagation topology; the vehicle controller can compare the risk magnitude of different candidate joint control sequences through constraint propagation costs; and the vehicle controller can correct the normalized constraint overrun intensity in subsequent control cycles through actual execution feedback. Coupling coefficient of constraint propagation Therefore, under operating conditions such as vehicle acceleration, uphill and downhill braking, high-temperature operation, high SOC of the power battery, increased temperature of the range extender generator system, or simultaneous approach of the upper limit of the allowable control range of multiple power subsystems, the vehicle controller can reduce the risk of multiple power subsystems exceeding the constraint limit simultaneously within the same control cycle while meeting the vehicle's target power or braking requirements, and improve the smoothness and safety of the actual joint control commands.

[0056] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Those skilled in the art can readily implement the present invention based on the accompanying drawings and the above description. However, any modifications, alterations, or variations made by those skilled in the art without departing from the scope of the present invention, utilizing the disclosed technical content, are equivalent embodiments of the present invention. Furthermore, any modifications, alterations, or variations made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the present invention.

Claims

1. A joint control method for the power system of a new energy vehicle, characterized in that, include: Collect vehicle operating status, driving environment status, and power system status to generate a comprehensive state vector; Multiple candidate joint control sequences that have not yet been issued for execution are generated based on the comprehensive state vector. The candidate joint control sequences include multiple joint control vectors in the current and subsequent predictive control cycles, and each joint control vector includes control quantities of at least two power subsystems. The candidate joint control sequence is input into the constraint mirror port corresponding to at least two power subsystems. The constraint mirror port is used to call the constraint judgment logic of the corresponding power subsystem and does not output an execution signal. Based on the real-time constraints of the corresponding power subsystem, each joint control vector is pre-examined. Preliminary review constraint information is generated based on the preliminary review results. The preliminary review constraint information includes the power subsystem identifier corresponding to the constraint exceeding the limit, the constraint exceeding the limit intensity, the constraint exceeding the limit period, and the constraint propagation object identifier. The constraint propagation topology corresponding to each candidate joint control sequence is constructed based on the power subsystem identifier and the constraint propagation object identifier, and the constraint propagation cost of each constraint propagation topology is determined based on the constraint over-limit strength and the constraint over-limit period. The target candidate joint control sequence is determined based on the propagation cost of each constraint. The joint control vector of the current control cycle in the target candidate joint control sequence is converted into a real joint control command and sent to the real execution port of the corresponding power subsystem.

2. The joint control method for the power system of new energy vehicles according to claim 1, characterized in that, The power subsystem includes at least two of the following: a power battery system, a drive motor system, a braking system, and a range extender generator system. The vehicle operating state includes at least one of the following: vehicle speed, accelerator pedal opening, brake pedal opening, and wheel slip ratio; The driving environment status includes at least one of the following: road slope, estimated road adhesion coefficient, and future road slope prediction information; The power system status includes at least one of the following: power battery status, drive motor status, braking status, and range extender generator system status. The power battery status includes at least one of the following: power battery SOC, power battery temperature, power battery allowable charging power, and power battery allowable discharging power. The drive motor status includes at least one of the drive motor speed, drive motor temperature, drive motor allowable output torque, and drive motor allowable regenerative torque. The braking state includes at least one of braking demand intensity, regenerative braking availability, mechanical braking compensation capability, and braking thermal load state. The range-extended power generation system status includes at least one of the following: current power generation, range-extended power generation system temperature, and allowable power generation.

3. The joint control method for the power system of new energy vehicles according to claim 2, characterized in that, Based on the comprehensive state vector, multiple candidate joint control sequences that have not yet been issued for execution are generated, including: Based on the vehicle operating state and drive motor state in the comprehensive state vector, a first candidate joint control sequence is generated. The first candidate joint control sequence is used to improve the following speed of the drive motor control quantity to the vehicle target power demand determined by the vehicle speed, accelerator pedal opening and overall vehicle driving demand. Based on at least two of the following states in the comprehensive state vector: power battery state, drive motor state, braking state, driving environment state, and range extender generator system state, a second candidate joint control sequence is generated. The second candidate joint control sequence is used to stagger the critical control value peaks in at least two power subsystems and distribute them to different control cycles to reduce the risk of multiple power subsystems simultaneously exceeding the constraint limit within the same control cycle. The critical control quantity is the control quantity that is close to the upper limit of the corresponding allowable control range or the upper limit of the rate of change.

4. The joint control method for the power system of new energy vehicles according to claim 3, characterized in that, The plurality of candidate joint control sequences that have not been issued for execution also include at least one of a third candidate joint control sequence and a fourth candidate joint control sequence; Based on the braking state, driving environment state and power battery state in the comprehensive state vector, the third candidate joint control sequence is generated. The third candidate joint control sequence is used to generate a joint control vector including regenerative braking control quantity and power battery charging control quantity in the braking state or downhill state. Based on at least one of the power battery state, drive motor state, range extender generator system state, and braking state in the comprehensive state vector, the fourth candidate joint control sequence is generated. The fourth candidate joint control sequence is used to reduce the power change amplitude, torque change amplitude, or braking force change amplitude of the corresponding power subsystem in the subsequent predictive control cycle when the temperature of the corresponding power subsystem reaches or exceeds a preset temperature warning threshold.

5. The joint control method for the power system of new energy vehicles according to claim 1, characterized in that, Pre-screening via the constrained mirror port includes: The joint control vector of each control cycle in the candidate joint control sequence is compared with the allowable control range of the corresponding power subsystem in the corresponding control cycle. When the control quantity corresponding to a certain power subsystem in the joint control vector exceeds the allowable control range of the power subsystem in the corresponding control cycle, the power subsystem identifier corresponding to the constraint over-limit is determined, and the constraint over-limit intensity is determined according to the over-limit magnitude. The prediction cycle number of the first occurrence of constraint over-limit is determined as the constraint over-limit cycle. Based on the preset power subsystem constraint linkage relationship, determine the constraint propagation object identifier corresponding to the power subsystem identifier.

6. The joint control method for the power system of a new energy vehicle according to claim 5, characterized in that, When generating the pre-examination constraint information, the constraint mirror port also outputs the alternative control range of the corresponding power subsystem in the current state. The alternative control range includes at least one of the following: the allowable charge and discharge power range of the power battery, the allowable torque range of the drive motor, the allowable power range of regenerative braking, and the mechanical braking compensation range. Based on the alternative control range, the corresponding control quantities in the candidate joint control sequences that have exceeded the constraint limits are limited and smoothed, or, under the condition of meeting the vehicle's target power requirements or target braking requirements, the compensation portion of the corresponding control quantities is allocated to the power subsystem corresponding to the constraint propagation object identifier, and the corrected candidate joint control sequences are re-input into the corresponding constraint mirror port for pre-screening.

7. The joint control method for the power system of new energy vehicles according to claim 1, characterized in that, Determining the constraint propagation cost for each of the aforementioned constraint propagation topologies includes determining the constraint propagation cost corresponding to each candidate joint control sequence according to the following formula: ； in, Indicates the first The constraint propagation cost corresponding to each candidate joint control sequence; Indicates the first The set of directed propagation relations in the constraint propagation topology corresponding to each candidate joint control sequence; The constraint is represented by the first The propagation of the first dynamic subsystem to the second One power subsystem; Indicates the first The first dynamic subsystem is related to the first Constraint propagation coupling coefficients of each dynamic subsystem; Indicates by the first The first power subsystem in the first The maximum value of the normalized constraint overlimit strength for each predictive control cycle under each candidate joint control sequence; This indicates that when the current control cycle is the 0th cycle, the 1st cycle... The first power subsystem in the first The prediction period number in which the first constraint exceedance occurs under each candidate joint control sequence; in, and All are non-negative numbers. It is a non-negative integer; when the first... When there is no directed propagation relationship in the constraint propagation topology corresponding to the candidate joint control sequence, the constraint propagation cost is determined by the ontological excess cost of each dynamic subsystem; when the first... When no constraint over-limit occurs in any of the dynamic subsystems, The corresponding item is not included in the constraint propagation cost.

8. The joint control method for the power system of a new energy vehicle according to claim 7, characterized in that, Determining the target candidate joint control sequence based on the constraint propagation cost includes: selecting candidate joint control sequences whose constraint propagation cost is lower than a preset cost threshold as optional candidate joint control sequences; When there are multiple candidate joint control sequences, the degree of matching of power demand, the constraint margin of the power subsystem and the smoothness evaluation result of each candidate joint control sequence are determined according to the comprehensive state vector and the joint control vector in each candidate joint control sequence, and the target candidate joint control sequence is determined accordingly. When the constraint propagation cost corresponding to each candidate joint control sequence is not lower than the preset cost threshold, a security reduction control sequence is generated according to the preset security reduction strategy. The security reduction control sequence is input into the constraint mirror port for pre-examination. When it meets the preset security constraint conditions, the security reduction control sequence is used as the target candidate joint control sequence. The safety derating control sequence includes at least one of limiting the rate of change of drive motor torque, limiting the power surge of range extender generator system, limiting the regenerative braking surge, and increasing the mechanical braking smoothing compensation.

9. The joint control method for the power system of a new energy vehicle according to claim 7, characterized in that, After sending the actual joint control command to the actual execution port of the corresponding power subsystem, the process also includes: Collect the actual execution feedback corresponding to the actual joint control command. The actual execution feedback is used to characterize whether the corresponding power subsystem has experienced an actual constraint exceedance. When the actual execution feedback indicates that a certain power subsystem has exceeded the actual constraint limit, and the corresponding pre-examination constraint information does not contain the identifier of the power subsystem or does not contain the identifier that the power subsystem is the constraint propagation object, the constraint limit exceedance strength correction coefficient corresponding to the power subsystem is increased. When the corresponding pre-qualification constraint information contains an identifier for a certain power subsystem, and the actual execution feedback indicates that the power subsystem has not experienced an actual constraint overrun, the constraint overrun intensity correction coefficient corresponding to that power subsystem is reduced; based on the constraint overrun intensity correction coefficient, the normalized constraint overrun intensity of the corresponding candidate joint control sequence in the next control cycle is adjusted. Or, the constraint propagation coupling coefficient between corresponding dynamic subsystems. Make corrections.

10. A power system joint control device for a new energy vehicle, characterized in that, include: The status acquisition module is used to collect vehicle operating status, driving environment status and power system status, and generate a comprehensive status vector; The candidate sequence generation module is used to generate multiple candidate joint control sequences that have not yet been issued for execution based on the comprehensive state vector. The candidate joint control sequences include joint control vectors in the current control period and subsequent predicted control periods. The constraint pre-examination module is used to input the candidate joint control sequence into the constraint mirror ports of at least two power subsystems, so as to pre-examine each joint control vector based on the real-time constraints of the corresponding power subsystems without outputting execution signals; The pre-approval information generation module is used to generate pre-approval constraint information based on the pre-approval results. The pre-approval constraint information includes the power subsystem identifier, constraint over-limit strength, constraint over-limit period, and constraint propagation object identifier. The propagation cost determination module is used to construct the constraint propagation topology corresponding to each candidate joint control sequence based on the power subsystem identifier and the constraint propagation object identifier, and to determine the constraint propagation cost of each constraint propagation topology; The target sequence determination module is used to determine the target candidate joint control sequence based on the propagation cost of each constraint. The real instruction generation module is used to convert the joint control vector of the current control cycle in the target candidate joint control sequence into a real joint control instruction and send it to the real execution port of the corresponding power subsystem.

Images