Power control method and system of hybrid power system

Abstract

The invention discloses a power control method of a hybrid power system, the hybrid power system comprises a hydrogen fuel cell and a power cell, and the method comprises the following steps: establishing a hydrogen fuel cell model, a power cell model and a load model; monitoring the SOC of a power battery, the opening degree of an accelerator pedal and the real-time rotating speed of a motor in real time; based on an intelligent control algorithm, with the hydrogen consumption, the hydrogen fuel cell output power variation and the power cell SOC as optimization targets, designing a target function and formulating constraint conditions; performing smooth filtering processing on the optimal control sequence solved based on the intelligent control algorithm; if the current working mode is in a hybrid mode, real-time acquisition and optimization are repeated in each control period, and otherwise, power control is performed according to a pure electric mode; and applying the solved control sequence to the hybrid power system. While power dynamic response is guaranteed, the hydrogen consumption is effectively reduced, and the problem of frequent load changing of a hydrogen fuel system is solved.

Description

Technical Field

[0001] The present invention relates to intelligent energy control, and in particular to a power control method and system for a hybrid power system. Background Art

[0002] At present, there are two relatively mature forms of power sources for new energy engineering machinery driven by electric motors: hydrogen fuel cells and power batteries. Hydrogen fuel cells have the advantages of high energy density and zero emissions, but their dynamic response is slow, and frequent load changes will cause the performance of the hydrogen fuel system to decay. Power batteries have the characteristics of fast response, but low energy density. For some high-power engineering machinery, a hybrid power system consisting of hydrogen fuel cells and power batteries has become a practical solution. In this solution, how to optimize the power distribution between the two to achieve efficient and stable energy management is a difficulty in current technology.

[0003] The existing technology has power allocation strategies such as "thermostat strategy" and "power following strategy". Due to their low requirements on controller performance, they can be deployed in mature controllers. At the same time, they also reduce hydrogen consumption and the load change frequency of hydrogen fuel cells to a certain extent. However, these methods are difficult to take into account hydrogen consumption, hydrogen fuel cell life and system dynamic response performance at the same time. Especially when the load power demand is large or changes rapidly, these strategies cannot effectively coordinate the working status of hydrogen fuel cells and power batteries. Although artificial intelligence algorithms such as machine learning and neural networks can take into account hydrogen consumption, fuel cell life and system dynamic response performance, they have high requirements on controller performance and are generally difficult to deploy in existing mature controllers, or they can be deployed but with low real-time performance. Summary of the invention

[0004] Purpose of the invention: To solve the above problems, this application proposes a power control method and system for a hybrid power system, designs an objective function, and solves the optimal control sequence for each control cycle by means of rolling optimization. While ensuring the dynamic response of power, this application effectively reduces hydrogen consumption and solves the frequent load change problem of the hydrogen fuel system.

[0005] Technical solution: A power control method for a hybrid power system, the hybrid power system includes a hydrogen fuel cell and a power battery, comprising the following steps:

[0006] (1) Establish hydrogen fuel cell model, power battery model and load model;

[0007] (2) Real-time monitoring of power battery SOC, accelerator pedal opening, and motor speed;

[0008] (3) Based on the intelligent control algorithm, the hydrogen consumption, the change in hydrogen fuel cell output power, and the power battery SOC are taken as optimization targets, and the objective function is designed and constraint conditions are formulated;

[0009] (4) Smoothing and filtering the optimal control sequence solved based on the intelligent control algorithm;

[0010] (5) If the current working mode is in the hybrid mode, repeat steps (2) to (4) in each control cycle; otherwise, perform power control in the pure electric mode;

[0011] (6) Apply the solved control sequence to the hybrid power system.

[0012] Furthermore, the hydrogen fuel cell model describes the relationship between the output power of the hydrogen fuel cell and the hydrogen consumption:

[0013]

[0014] Among them, P fc : Output power of hydrogen fuel cell; η fc : Efficiency of hydrogen fuel cells; Hydrogen consumption rate refers to the amount of hydrogen consumed per unit time; The low heating value of hydrogen;

[0015] The power battery model describes the relationship between the power battery SOC change and the charge and discharge power:

[0016]

[0017] Among them, SOC Bat (t): SOC of the power battery at the current moment; P Bat : The charge and discharge power of the power battery, positive value means charging, negative value means discharging; Δt: time step; Q Bat : The total capacity of the power battery;

[0018] The load model describes the power relationship between the bus load power demand and the hydrogen fuel cell and the power battery:

[0019] P Load =P fc +P Bat

[0020] P Load : Required power of bus load.

[0021] Furthermore, the intelligent control algorithm adopts the TinyMPC algorithm. The present invention uses TinyMPC as a power allocation control algorithm, which can take into account hydrogen consumption, hydrogen fuel cell life and system dynamic response performance, and is easy to deploy in a resource-constrained microcontroller, solving the problem of real-time operation.

[0022] Furthermore, the step (3) comprises:

[0023] (3.1) Real-time monitoring of bus load power demand P Load ;

[0024] (3.2) Real-time monitoring of the bus load change rate ΔP Load :

[0025]

[0026] (3.3) Real-time monitoring of hydrogen consumption and the output power P of the hydrogen fuel cell fc ;

[0027] (3.4) Taking hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as optimization targets, the design objective function is as follows:

[0028]

[0029] in, is the hydrogen consumption at the kth moment; N is the prediction time domain length; ΔP fc (k) = P fc (k)-P fc (k-1), is the change in the output power of the hydrogen fuel cell, where P fc (k) is the output power of the hydrogen fuel cell at the kth moment; SOC Bat (k) is the SOC value of the power battery at the kth moment; SOC ref is the SOC reference value of the power battery; Penalty(k)=max(0,P load (k)-P threshold ) 2 , is the high load penalty term, used to give priority to meeting the power demand of high power loads, P threshold is the required power threshold; α, β, γ, and δ are all weight coefficients used to balance hydrogen consumption, power fluctuation, and SOC management of the power battery;

[0030] (3.5) The constraints are formulated as follows:

[0031] Hydrogen fuel cell output power range:

[0032]

[0033] Power battery charging and discharging power range:

[0034]

[0035] Power battery SOC Bat scope:

[0036]

[0037] Load power balance:

[0038] P Load (k) = P fc (k)+P Bat (k).

[0039] Furthermore, the bus load consists of fixed load and variable load.

[0040] The fixed load consists of equipment other than the motor. The total required power is calculated based on the technical parameters of each equipment. Setting the power margin τ 1 , suppress the instantaneous sharp increase of power and ensure its smooth transition to maintain system stability and prevent overload risk. The value range is [1,1.5], is the required power of the i-th device, K is the number of devices;

[0041] The variable load comes from the motor. By monitoring the accelerator pedal opening and the real-time motor speed, combined with the motor's traction external characteristic curve, the required power of the motor can be obtained by looking up the table Among them, the power margin τ of the motor is set 2 , the value range is (0,0.98], to prevent the motor from working outside the external characteristic curve, is the required power of the i-th motor, J is the number of motors;

[0042]

[0043] Therefore, the load power balance in the constraint is:

[0044] P Load (k) = P fc (k)+P Bat (k)

[0045]

[0046] Furthermore, in step (4), a smoothing filter is performed, specifically, a first-order low-pass filter is used to smooth the optimal control sequence:

[0047]

[0048] in, is the filtered power at the kth moment, μ is the filter coefficient, and its value range is [0,1].

[0049] A power control system of a hybrid power system, the hybrid power system includes a hydrogen fuel cell and a power battery, the hybrid power system has two working modes: a pure electric mode and a hybrid mode. In the pure electric mode, only the power battery provides the required power for the bus load, and the hydrogen fuel cell control system is turned off; in the hybrid mode, the hydrogen fuel cell gives priority to charging the power battery, and when the power battery SOC reaches or exceeds a threshold, the hydrogen fuel cell and the power battery jointly provide the required power for the bus load;

[0050] The power control system establishes a hydrogen fuel cell model, a power battery model and a load model; monitors the power battery SOC, the accelerator pedal opening and the real-time speed of the motor in real time; based on the intelligent control algorithm, takes the hydrogen consumption, the change in the output power of the hydrogen fuel cell and the power battery SOC as the optimization targets, designs the objective function and formulates the constraint conditions; performs smoothing and filtering on the optimal control sequence solved based on the intelligent control algorithm; if the current working mode is in the hybrid mode, repeatedly obtains real-time data and optimizes it in each control cycle, otherwise, performs power control in the pure electric mode;

[0051] The control sequence solved by the power control system is applied to the hybrid power system.

[0052] Furthermore, the hydrogen fuel cell model describes the relationship between the output power of the hydrogen fuel cell and the hydrogen consumption:

[0053]

[0054] Among them, P fc : Output power of hydrogen fuel cell; η fc : Efficiency of hydrogen fuel cells; Hydrogen consumption rate refers to the amount of hydrogen consumed per unit time; The low heating value of hydrogen;

[0055] The power battery model describes the relationship between the power battery SOC change and the charge and discharge power:

[0056]

[0057] Among them, SOC Bat (t): SOC of the power battery at the current moment; P Bat : The charge and discharge power of the power battery, positive value means charging, negative value means discharging; Δt: time step; Q Bat : The total capacity of the power battery;

[0058] The load model describes the power relationship between the bus load power demand and the hydrogen fuel cell and the power battery:

[0059] P Load =P fc +P Bat

[0060] P Load : Required power of bus load.

[0061] Furthermore, based on the intelligent control algorithm, the hydrogen consumption, the change in the output power of the hydrogen fuel cell, and the SOC of the power battery are taken as optimization targets, the objective function is designed and the constraint conditions are formulated, including:

[0062] (a) Real-time monitoring of bus load power demand P Load ;

[0063] (b) Real-time monitoring of the bus load change rate ΔP Load :

[0064]

[0065] (c) Real-time monitoring of hydrogen consumption and the output power P of the hydrogen fuel cell fc ;

[0066] (d) Taking hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as optimization targets, the design objective function is as follows:

[0067]

[0068] in, is the hydrogen consumption at the kth moment; N is the prediction time domain length; ΔP fc (k) = P fc (k)-P fc (k-1), is the change in the output power of the hydrogen fuel cell, where P fc (k) is the output power of the hydrogen fuel cell at the kth moment; SOC Bat (k) is the SOC value of the power battery at the kth moment; SOC ref is the SOC reference value of the power battery; Penalty(k)=max(0,P load (k)-P threshold ) 2 , is the high load penalty term, used to give priority to meeting the power demand of high power loads, P threshold is the required power threshold; α, β, γ, and δ are all weight coefficients used to balance hydrogen consumption, power fluctuation, and SOC management of the power battery;

[0069] (e) The constraints are as follows:

[0070] Hydrogen fuel cell output power range:

[0071]

[0072] Power battery charging and discharging power range:

[0073]

[0074] Power battery SOC Bat scope:

[0075]

[0076] Load power balance:

[0077] P Load (k) = P fc (k)+P Bat (k).

[0078] Furthermore, the bus load consists of fixed load and variable load.

[0079] The fixed load consists of equipment other than the motor. The total required power is calculated based on the technical parameters of each equipment. Setting the power margin τ 1 , suppress the instantaneous sharp increase of power and ensure its smooth transition to maintain system stability and prevent overload risks. The value range is [1,1.5]. is the required power of the i-th device, K is the number of devices;

[0080] The variable load comes from the motor. By monitoring the accelerator pedal opening and the real-time motor speed, combined with the motor's traction external characteristic curve, the required power of the motor can be obtained by looking up the table Among them, the power margin τ of the motor is set 2 , the value range is (0,0.98], to prevent the motor from working outside the external characteristic curve, is the required power of the i-th motor, J is the number of motors;

[0081]

[0082] Therefore, the load power balance in the constraint is:

[0083] P Load (k) = P fc (k)+P Bat (k)

[0084]

[0085] Beneficial effects: (1) The present invention realizes flexible switching of multiple working modes and efficient energy management by optimizing and controlling the output power of hydrogen fuel cells and power batteries. (2) Real-time monitoring of key parameters, design of objective functions and optimization control are carried out. By adjusting the output power of hydrogen fuel cells and power batteries in real time, it is ensured that the system can operate efficiently under various working conditions. (3) Smoothing and filtering processing makes the power output more stable, reduces the mechanical and electrical stress of the system, and improves the service life and operating stability of the system. (4) The system sets a power margin to prevent the power from suddenly changing to a large value, ensuring that the system can still maintain stable operation when the load suddenly changes. (5) The system monitors and optimizes the SOC of the power battery in real time to ensure that the SOC of the power battery is always maintained within a reasonable range, avoiding overcharging or discharging. Reasonable SOC management extends the service life of the power battery and improves the reliability and economy of the system. (6) In resource-constrained microcontrollers with limited computing power, the low computational complexity of TinyMPC enables it to run in real time with limited computing resources, meeting the system's requirements for real-time control. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 It is the topological diagram of the hybrid power system structure;

[0087] Figure 2 It is the overall power control flow chart of the hybrid system;

[0088] Figure 3 This is the flow chart for power control implementation based on TinyMPC algorithm in hybrid mode. DETAILED DESCRIPTION

[0089] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0090] The present invention provides a technical solution: a power control method of a hybrid power system, comprising the following steps: S1: The present invention requires the following hardware topology: the hydrogen fuel cell can charge the power battery; the hydrogen fuel cell can directly output power to the bus load; the power battery can directly output power to the bus load; the power battery can receive energy recovery from the bus load, such as Figure 1 shown.

[0091] S2: Figure 2 It is the overall control flow chart. Two working modes (PowerMode) are defined: "EcoMode" and "HybridMode". The default working mode is "EcoMode" when the device is turned on.

[0092] S3: Add a "mechanical" or "electronic" selection switch in the driver's cab for the functions mentioned in S2. The driver can independently select and switch the "working mode" in the cab. In the "pure electric mode", only the power battery provides the required power for the bus load, and the hydrogen fuel cell control system is turned off; in the "hybrid mode", the hydrogen fuel cell gives priority to charging the power battery. When the power battery SOC reaches or exceeds the threshold, the hydrogen fuel cell and the power battery can jointly provide the required power for the bus load.

[0093] S4: The steps for optimizing the output power of the hydrogen fuel cell and the power battery in the "hybrid mode" are as follows:

[0094] S5: Model the system, including the hydrogen fuel cell model, power battery model and load model.

[0095] S6: Model describing the relationship between the output power and hydrogen consumption of hydrogen fuel cells:

[0096]

[0097] in:

[0098] P fc : Output power of hydrogen fuel cell;

[0099] η fc : Efficiency of hydrogen fuel cells;

[0100] Hydrogen consumption rate;

[0101] The low heating value of hydrogen;

[0102] S7: Model describing the relationship between the change in SOC of power batteries and the charge and discharge power:

[0103]

[0104] in:

[0105] SOC Bat (t): current battery SOC;

[0106] P Bat : The charging and discharging power of the power battery (positive value means charging, negative value means discharging);

[0107] Δt: time step;

[0108] Q Bat : The total capacity of the power battery;

[0109] S8: Model describing the power relationship between bus load power demand and hydrogen fuel cells and power batteries:

[0110] P Load =P fc +P Bat

[0111] P Load : Required power of the load;

[0112] S9: Solve the optimization problem based on the system model: take hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as the optimization targets, design the objective function, process (such as Figure 3 As shown) as follows:

[0113] S10: Real-time monitoring of the state of charge (SOC) of the power battery: This can be obtained in real time through the battery management system (BMS).

[0114] S11: Get the accelerator pedal opening (Aps_Thrrottle): The VCU controller can directly collect it.

[0115] S12: Get the real-time speed of the motor Mot_Rpm: it can be obtained in real time through the motor controller (MCU).

[0116] S13: Real-time monitoring of bus load power demand P Load :The bus load generally consists of two parts: fixed load and variable load. The fixed load consists of equipment other than the motor. The total required power can be calculated based on the technical parameters of each equipment. Setting the power margin τ 1 , suppress the instantaneous sharp increase of power and ensure its smooth transition to maintain system stability and prevent overload risks. The value range is [1,1.5]. is the required power of the i-th device, K is the number of devices; the variable load mainly comes from the motor. By monitoring the accelerator pedal opening (Aps_Thrrottle) and the current motor speed (Mot_Rpm), combined with the traction external characteristic curve of the motor, the required power of the motor can be obtained by looking up the table Among them, the power margin τ of the motor is set 2 , the value range is (0,0.98], to prevent the motor from working outside the external characteristic curve, is the required power of the i-th motor, and J is the number of motors.

[0117] S14:

[0118] S15: Real-time monitoring of the bus load change rate ΔP Load :

[0119]

[0120] S16: Real-time monitoring of hydrogen consumption (W H2 ) and hydrogen output power (P fc ): can be obtained in real time through its hydrogen fuel control system (FCS).

[0121] S17: Based on the TinyMPC algorithm, the hydrogen consumption, the change in hydrogen fuel cell output power, and the power battery SOC are taken as optimization targets. The design objective function is as follows:

[0122]

[0123] in:

[0124] is the hydrogen consumption at the kth moment;

[0125] N: prediction time domain length;

[0126] ΔP fc (k) = P fc (k)-P fc (k-1), is the power change of the hydrogen fuel cell, where P fc (k) is the output power of the hydrogen fuel cell at the kth moment;

[0127] SOC Bat (k) is the SOC value of the power battery at the moment;

[0128] SOC ref is the SOC reference value of the power battery;

[0129] Penalty(k)=max(0,P load (k)-P threshold ) 2 , is the high load penalty term, used to give priority to meeting the power demand of high power loads, P threshold is the required power threshold;

[0130] α, β, γ, δ: weight coefficients used to balance hydrogen consumption, power fluctuation and balance, and SOC management of the power battery.

[0131] S18: The constraints are as follows:

[0132] The following constraints are met during the TinyMPC optimization process:

[0133] Hydrogen fuel cell output power range:

[0134]

[0135] Power battery charging and discharging power range:

[0136]

[0137] Power battery SOC Bat scope:

[0138]

[0139] Load power balance:

[0140] P Load (k) = P fc (k)+P Bat (k)

[0141] S19: Further, the optimal control sequence solved based on TinyMPC is smoothed and filtered. Specifically, a first-order low-pass filter is used to smooth the optimal control sequence:

[0142]

[0143] in, is the filtered power at the kth moment, μ is the filter coefficient, and its value range is [0,1].

[0144] S20: If the current working mode is in the "hybrid mode", repeat the process S10 to S19 in each control cycle. Otherwise, perform power control as described in S3. The program code is executed periodically in the vehicle VCU controller, and the time for the code to be executed in a loop is the "control cycle".

[0145] S21: Finally, the solved control sequence is applied to the system.

[0146] This application achieves flexible switching of multiple working modes and efficient energy management by optimizing and controlling the output power of hydrogen fuel cells and power batteries. The following are the main advantages of this application and its reasoning process:

[0147] Topology design:

[0148] The bidirectional energy flow design of hydrogen fuel cells and power batteries: hydrogen fuel cells can charge power batteries and directly output power to bus loads; power batteries can directly output power to bus loads and receive energy recovery from bus loads at the same time. This topology achieves efficient utilization and flexible allocation of energy, providing the basis for multiple working modes for the system.

[0149] Flexible working mode switching:

[0150] Pure electric mode (EcoMode): In this mode, only the power battery provides power to the bus load, and the hydrogen fuel cell control system is turned off. This mode is suitable for low-power demand scenarios, which can reduce hydrogen consumption and reduce operating costs. Hybrid mode (HydMode): In this mode, the hydrogen fuel cell charges the power battery first. When the SOC of the power battery reaches or exceeds the threshold, the hydrogen fuel cell and the power battery jointly provide power to the bus load. This mode is suitable for high-power demand scenarios, which can extend the cruising range and improve the overall efficiency of the system. The dual working mode design enables the system to switch flexibly according to actual needs, meeting both low-power and high-power demands.

[0151] By adding mechanical or electronic selection switches, the driver can flexibly switch working modes according to actual needs, which can not only meet the needs of different driving scenarios but also optimize energy utilization efficiency.

[0152] Optimized energy management:

[0153] The coordinated work of hydrogen fuel cells and power batteries: In hybrid mode, hydrogen fuel cells give priority to charging power batteries to ensure that the SOC of the power batteries is maintained within a reasonable range. When the SOC of the power battery reaches the threshold, the hydrogen fuel cell and the power battery jointly provide power for the load, avoiding excessive use of a single energy source. This coordinated working mode not only improves the energy utilization efficiency of the system, but also extends the service life of the power battery, reduces hydrogen consumption, and reduces operating costs.

[0154] Real-time monitoring and optimized control:

[0155] The system monitors key parameters such as the SOC of the power battery, the speed of the motor, the accelerator pedal opening, the power demand of the bus load, and the hydrogen consumption in real time through the battery management system (BMS), the motor controller (MCU) and the hydrogen fuel control system (FCS).

[0156] Based on the TinyMPC algorithm, the system takes hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as optimization targets, designs the objective function and performs optimization control. By adjusting the output power of hydrogen fuel cells and power batteries in real time, the system can ensure efficient operation under various working conditions.

[0157] Real-time monitoring and optimized control enable the system to respond quickly to load changes, maintain power balance, avoid power fluctuations, and improve system stability and reliability.

[0158] Smoothing filter processing:

[0159] The optimal control sequence solved based on TinyMPC is smoothed and filtered, and a first-order low-pass filter is used to smooth the optimal control sequence, reducing the fluctuation of power output. Smoothing filtering makes the power output more stable, reduces the mechanical and electrical stress of the system, and improves the service life and operation stability of the system.

[0160] Load power balancing and margin management:

[0161] The system monitors the power demand of the bus load in real time to ensure that the output power of the hydrogen fuel cell and power battery can meet the load demand and avoid insufficient or excessive power. The system sets a power margin to prevent the power from suddenly changing to a large value, ensuring that the system can still maintain stable operation when the load suddenly changes. Load power balance and margin management enable the system to cope with various complex load changes, improving the adaptability and reliability of the system.

[0162] Power battery SOC management:

[0163] The system ensures that the SOC of the power battery is always maintained within a reasonable range by real-time monitoring and optimization control, thus avoiding overcharging or discharging. Reasonable SOC management prolongs the service life of the power battery and improves the reliability and economy of the system.

[0164] Deployment of TinyMPC algorithm in controller:

[0165] The TinyMPC (Tiny Model Predictive Control) algorithm is a lightweight model predictive control algorithm that is particularly suitable for deployment in resource-constrained embedded systems or microcontrollers. The TinyMPC algorithm reduces the computational complexity by simplifying the solution process of the optimization problem. It usually uses linearized models and simplified constraints to reduce the scale of matrix operations. It also reduces memory usage by reducing the number of variables and constraints in the optimization problem. It usually uses sparse matrix storage and efficient numerical calculation methods to further save memory. In addition, the TinyMPC algorithm has a short calculation time and can quickly solve the optimization problem in each control cycle to meet the needs of real-time control. At the same time, the framework of the TinyMPC algorithm is flexible, and the optimization objectives and constraints can be adjusted according to specific application scenarios to adapt to different control requirements. Furthermore, the TinyMPC algorithm can consider the dynamic characteristics and future state changes of the system through the framework of model predictive control (MPC), and has strong robustness.

[0166] In resource-constrained microcontrollers, the computing power is limited. The low computational complexity of TinyMPC enables it to run in real time under limited computing resources, meeting the system's demand for real-time control. The low memory usage of TinyMPC enables it to run efficiently in resource-constrained environments, avoiding the risk of memory overflow. The fast solution capability of TinyMPC ensures that the system can respond to load changes in a timely manner and maintain power balance. In the present invention, the TinyMPC algorithm is used to optimize hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC. Its scalability allows further optimization targets (such as system efficiency, load priority, etc.) to be added in the future without changing the hardware platform. In complex power systems, load demand and operating condition changes are uncertain. Finally, the robustness of TinyMPC enables the system to maintain stable operation under various operating conditions, avoiding power fluctuations and energy waste. In microcontrollers, complex algorithm implementations may be limited by hardware resources. The simple implementation characteristics of TinyMPC enable it to be quickly integrated into existing control systems, reducing development difficulty and cost. It solves the technical status quo that excellent control algorithms cannot be deployed in existing mature VCU controllers. The TinyMPC algorithm can even be deployed in microcontrollers with a minimum of 8 bits, 128B (RAM), and 2KB (FLASH).

Claims

1. A power control method for a hybrid power system, the hybrid power system comprising a hydrogen fuel cell and a power battery, characterized in that: The following steps are involved: (1) Establish hydrogen fuel cell model, power battery model and load model; (2) Real-time monitoring of power battery SOC, accelerator pedal opening, and motor speed; (3) Based on the intelligent control algorithm, the hydrogen consumption, the change in hydrogen fuel cell output power, and the power battery SOC are taken as optimization targets, and the objective function is designed and constraint conditions are formulated; (4) Smoothing and filtering the optimal control sequence solved based on the intelligent control algorithm; (5) If the current working mode is in the hybrid mode, repeat steps (2) to (4) in each control cycle; otherwise, perform power control in the pure electric mode; (6) Apply the solved control sequence to the hybrid power system.

2. The power control method of a hybrid power system according to claim 1, characterized in that: The hydrogen fuel cell model describes the relationship between the output power of the hydrogen fuel cell and the hydrogen consumption: Among them, P fc : Output power of hydrogen fuel cell; η fc : Efficiency of hydrogen fuel cells; Hydrogen consumption rate refers to the amount of hydrogen consumed per unit time; The low heating value of hydrogen; The power battery model describes the relationship between the power battery SOC change and the charge and discharge power: Among them, SOC Bat (t): SOC of the power battery at the current moment; P Bat : The charge and discharge power of the power battery, positive value means charging, negative value means discharging; Δt: time step; Q Bat : The total capacity of the power battery; The load model describes the power relationship between the bus load power demand and the hydrogen fuel cell and the power battery: P Load =P fc +P Bat P Load : Required power of bus load.

3. The power control method of a hybrid power system according to claim 1, characterized in that: The intelligent control algorithm adopts the TinyMPC algorithm.

4. The power control method of a hybrid power system according to claim 1, characterized in that: The step (3) comprises: (3.1) Real-time monitoring of bus load power demand P Load ; (3.2) Real-time monitoring of the bus load change rate ΔP Load : (3.3) Real-time monitoring of hydrogen consumption and the output power P of the hydrogen fuel cell fc ; (3.4) Taking hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as optimization targets, the design objective function is as follows: in, is the hydrogen consumption at the kth moment; N is the prediction time domain length; ΔP fc (k) = P fc (k)-P fc (k-1), is the change in the output power of the hydrogen fuel cell, where P fc (k) is the output power of the hydrogen fuel cell at the kth moment; SOC Bat (k) is the SOC value of the power battery at the kth moment; SOC ref is the SOC reference value of the power battery; Penalty(k)=max(0,P load (k)-P threshold ) 2 , is the high load penalty term, P threshold is the required power threshold; α, β, γ, δ are all weight coefficients; (3.5) The constraints are formulated as follows: Hydrogen fuel cell output power range: Power battery charging and discharging power range: Power battery SOC Bat scope: Load power balance: P Load (k)=P fc (k)+P Bat (k)。 5. The power control method of a hybrid power system according to claim 4, characterized in that: The bus load consists of fixed load and variable load. The fixed load consists of equipment other than the motor. The total required power is calculated based on the technical parameters of each equipment. Set the power margin τ1, the value range is [1,1.5], is the required power of the i-th device, K is the number of devices; The variable load comes from the motor. By monitoring the accelerator pedal opening and the real-time motor speed, combined with the motor's traction external characteristic curve, the required power of the motor can be obtained by looking up the table. Among them, the power margin τ2 of the motor is set, and the value range is (0,0.98], is the required power of the i-th motor, J is the number of motors; Therefore, the load power balance in the constraint is: P Load (k)=P fc (k)+P Bat (k) 6. The power control method of a hybrid power system according to claim 1, characterized in that: In step (4), a smoothing filter is performed, specifically a first-order low-pass filter is used to smooth the optimal control sequence: in, is the filtered power at the kth moment, μ is the filter coefficient, and its value range is [0,1].

7. A power control system for a hybrid power system, the hybrid power system comprising a hydrogen fuel cell and a power battery, characterized in that: The hybrid power system has two working modes: pure electric mode and hybrid mode. In the pure electric mode, only the power battery provides the required power for the bus load, and the hydrogen fuel cell control system is turned off; in the hybrid mode, the hydrogen fuel cell gives priority to charging the power battery. When the power battery SOC reaches or exceeds the threshold, the hydrogen fuel cell and the power battery jointly provide the required power for the bus load. The power control system establishes a hydrogen fuel cell model, a power battery model and a load model; monitors the power battery SOC, the accelerator pedal opening and the real-time speed of the motor in real time; based on the intelligent control algorithm, takes the hydrogen consumption, the change in the output power of the hydrogen fuel cell and the power battery SOC as the optimization targets, designs the objective function and formulates the constraint conditions; performs smoothing and filtering on the optimal control sequence solved based on the intelligent control algorithm; if the current working mode is in the hybrid mode, repeatedly obtains real-time data and optimizes it in each control cycle, otherwise, performs power control in the pure electric mode; The control sequence solved by the power control system is applied to the hybrid power system.

8. The power control system of a hybrid power system according to claim 7, characterized in that: The hydrogen fuel cell model describes the relationship between the output power of the hydrogen fuel cell and the hydrogen consumption: Among them, P fc : Output power of hydrogen fuel cell; η fc : Efficiency of hydrogen fuel cells; Hydrogen consumption rate refers to the amount of hydrogen consumed per unit time; The low heating value of hydrogen; The power battery model describes the relationship between the power battery SOC change and the charge and discharge power: Among them, SOC Bat (t): SOC of the power battery at the current moment; P Bat : The charge and discharge power of the power battery, positive value means charging, negative value means discharging; Δt: time step; Q Bat : The total capacity of the power battery; The load model describes the power relationship between the bus load power demand and the hydrogen fuel cell and the power battery: P Load =P fc +P Bat P Load : Required power of bus load.

9. The power control system of a hybrid power system according to claim 7, characterized in that: Based on the intelligent control algorithm, the hydrogen consumption, the change in the output power of the hydrogen fuel cell, and the SOC of the power battery are taken as optimization targets, and the objective function is designed and the constraint conditions are formulated, including: (a) Real-time monitoring of bus load power demand P Load ; (b) Real-time monitoring of the bus load change rate ΔP Load : (c) Real-time monitoring of hydrogen consumption and the output power P of the hydrogen fuel cell fc ; (d) Taking hydrogen consumption, hydrogen fuel cell output power change, and power battery SOC as optimization targets, the design objective function is as follows: in, is the hydrogen consumption at the kth moment; N is the prediction time domain length; ΔP fc (k) = P fc (k)-P fc (k-1), is the change in the output power of the hydrogen fuel cell, where P fc (k) is the output power of the hydrogen fuel cell at the kth moment; SOC Bat (k) is the SOC value of the power battery at the kth moment; SOC ref is the SOC reference value of the power battery; Penalty(k)=max(0,P load (k)-P threshold ) 2 , is the high load penalty term, P threshold is the required power threshold; α, β, γ, δ are all weight coefficients; (e) The constraints are as follows: Hydrogen fuel cell output power range: Power battery charging and discharging power range: Power battery SOC Bat scope: Load power balance: P Load (k)=P fc (k)+P Bat (k)。 10. The power control system of a hybrid power system according to claim 9, characterized in that: The bus load consists of fixed load and variable load. The fixed load consists of equipment other than the motor. The total required power is calculated based on the technical parameters of each equipment. Set the power margin τ1, the value range is [1,1.5], is the required power of the i-th device, K is the number of devices; The variable load comes from the motor. By monitoring the accelerator pedal opening and the real-time motor speed, combined with the motor's traction external characteristic curve, the required power of the motor can be obtained by looking up the table. Among them, the power margin τ2 of the motor is set, and the value range is (0,0.98], is the required power of the i-th motor, J is the number of motors; Therefore, the load power balance in the constraint is: P Load (k)=P fc (k)+P Bat (k)

Images