Fuel cell fault-tolerant control method and device based on reinforcement learning

Abstract

The invention discloses a fuel cell fault-tolerant control method and device based on reinforcement learning in the field of fuel cells, and the method comprises the steps: obtaining monitoring parameters of a fuel cell in an operation state, and constructing a state space based on the monitoring parameters; obtaining an executable operation in the running state of the fuel cell, and constructing an action space based on the executable operation; constructing a reward function based on the state space and the action space by taking control cost minimization as a target; constructing a fault-tolerant control strategy model based on a DDPG algorithm, and training and verifying the fault-tolerant control strategy model through a state space, an action space and a reward function; and fuel cell fault-tolerant control is carried out through the trained and verified fault-tolerant control strategy model. According to the method, the DDPG algorithm is introduced, target state parameters are monitored in real time in the operation process, a control strategy is automatically adjusted by using reinforcement learning when a fault occurs, it is ensured that the system can still keep stable power output under the fault condition, and optimal control is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of fuel cells, and in particular to a fuel cell fault-tolerant control method and device based on reinforcement learning. Background Art

[0002] Fault-tolerant control is a technique that defines control actions by considering fault diagnosis, and consists of two parts: fault diagnosis and control reconstruction. Fault-tolerant control of the system requires fault detection and isolation before the fault causes performance degradation, and mitigates the impact of the fault through adaptive control strategies. In industrial applications with many types of faults and high probability of failure, fault-tolerant control of proton exchange membrane fuel cell (PEMFC) systems has broad development potential.

[0003] At present, in the research on PEMFC systems, fault diagnosis and control are usually carried out separately. Only a few studies have integrated fault diagnosis and control actions (i.e., fault-tolerant control). In actual industrial applications, PEMFC systems may face a variety of failure modes, including PEMFC failure itself, component aging, subsystem failure, and sensor failure. Therefore, fault-tolerant control that can complete fault diagnosis online in real time and actively reconstruct the control strategy according to the fault type is crucial to ensure the operational stability of the system. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a fuel cell fault-tolerant control method and device based on reinforcement learning to solve the technical problem of fuel cell fault-tolerant control.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a fuel cell fault-tolerant control method based on reinforcement learning, comprising: Acquiring monitoring parameters of the fuel cell under operating conditions, and constructing a state space based on the monitoring parameters; Obtaining executable operations under the operating state of the fuel cell, and constructing an action space based on the executable operations; Taking minimizing control cost as a goal, constructing a reward function based on the state space and the action space; Constructing a fault-tolerant control strategy model based on the DDPG algorithm, and training and verifying the fault-tolerant control strategy model through the state space, the action space, and the reward function; Fuel cell fault-tolerant control is performed using the trained and verified fault-tolerant control strategy model.

[0006] Optionally, the monitoring parameters include: Minimum fuel cell voltage , the average value of fuel cell cell voltage , fuel cell stack temperature , Hydrogen inlet temperature , Hydrogen inlet pressure , Hydrogen out-of-combustion pressure , Hydrogen inlet flow , Hydrogen medium pressure and the pressure drop on the anode side of the fuel cell ; The state space for: ; In the formula, For the moment.

[0007] Optionally, the executable operations include: Control the coolant thermostat opening , Control the opening of the hydrogen inlet proportional valve , Control the switching frequency of the nitrogen exhaust valve , control the drain valve switching frequency , Control the speed of hydrogen circulation pump And control the speed of the air compressor ; The action space for: ; In the formula, For the moment.

[0008] Optionally, the reward function for: ; In the formula, for Moment-by-moment action overhead, failure cost, and smooth control cost; ; In the formula, is the weight coefficient of action cost, is the working power consumption of the air compressor and hydrogen circulation pump, Adjust the power consumption for the opening of the coolant thermostat and the hydrogen inlet proportional valve. The switching power consumption of nitrogen exhaust valve and water drain valve; ; In the formula, is the weight coefficient of the failure cost, is the fault cost corresponding to the minimum and average values ​​of the fuel cell cell voltage, is the failure cost corresponding to the fuel cell stack temperature and the hydrogen inlet temperature, is the failure cost corresponding to the hydrogen inlet pressure, hydrogen outlet pressure, and hydrogen medium pressure, is the failure cost corresponding to the hydrogen inflow rate, is the failure cost corresponding to the pressure drop on the anode side of the fuel cell; ; In the formula, is the weight coefficient for controlling the smoothness, for Moment of action.

[0009] Optionally, the fault-tolerant control strategy model based on the DDPG algorithm includes an Actor network , Critic Network , Target Actor Network And the target critic network ; The training and verification of the fault-tolerant control strategy model through the state space, the action space and the reward function includes: Initializing the Actor Network Parameters , Critic Network Parameters , Target Actor Network Parameters , Target Critic Network Parameters ; Initialize training rounds , repeat the following steps: Initialize The fuel cell running status of the wheel, get the status at the initial moment ; Based on status For The fault-tolerant control strategy model of the first round is trained to obtain The fault-tolerant control strategy model of the wheel is ; like , for The fault-tolerant control strategy model of the round is subjected to performance verification. If the performance verification passes, the final fault-tolerant control strategy model is obtained.

[0010] Optionally, the state-based Training the fault-tolerant control strategy model includes: initialization , enter the sampling phase and repeat the following steps: Based on status , through the Actor network Select Action , for noise; Execute an action After that, get the reward and status , generate experience samples , and store it in the experience playback buffer ,make ; when , For the preset total number of moments, enter the training phase: From the Experience Replay Buffer Randomly extract small batches of experience samples as training samples; Based on the training samples, the Critic network is updated by minimizing the loss function based on the temporal difference error. Parameters , loss function for: ; ; In the formula, is the discount factor, For the training samples, is the total number of training samples, For the The estimation of training samples value; like , is a constant term, For modulo operation, the Actor network is updated by the policy gradient method Parameters , policy gradient for: ; Based on parameters and Update Target Actor Network Parameters and target critic network Parameters : ; ; In the formula, is the update coefficient.

[0011] In a second aspect, the present invention provides a fuel cell fault-tolerant control device based on reinforcement learning, comprising: A state space construction module is configured to obtain monitoring parameters under the operating state of the fuel cell and construct a state space based on the monitoring parameters; An action space construction module is configured to obtain executable operations under the operating state of the fuel cell and construct an action space based on the executable operations; A reward function construction module, configured to construct a reward function based on the state space and the action space; The strategy model building module is configured to build a fault-tolerant control strategy model based on the DDPG algorithm with the goal of minimizing the control cost; A model training and verification module, configured to train and verify the fault-tolerant control strategy model through the state space, the action space and the reward function; The model deployment application module is configured to perform fuel cell fault-tolerant control through the fault-tolerant control strategy model after training and verification.

[0012] In a third aspect, the present invention provides an electronic device, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0014] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a fuel cell fault-tolerant control method and device based on reinforcement learning, which has 1) adaptive fault-tolerant control based on reinforcement learning: by introducing the DDPG algorithm, the system can adapt to different fault conditions through self-learning in a complex dynamic environment. The method has stronger real-time and flexibility, and can automatically adjust the control strategy to reduce the performance degradation caused by faults. 2) Integration of fault diagnosis and control reconstruction: combining fault diagnosis with control reconstruction, it can detect faults in real time and immediately reconstruct the control strategy, so that the system can still maintain stable operation after the fault occurs, reducing downtime and performance degradation. 3) Multi-dimensional state monitoring and precise control: compared with the traditional method of monitoring a single or a small number of parameters, the method dynamically adjusts key components such as hydrogen supply and air compressor through comprehensive monitoring of multi-dimensional parameters such as current, voltage, temperature, and pressure to ensure that the fuel cell achieves optimal performance under various working conditions. In summary, the present invention enables the system to find the best balance between performance and efficiency in fault-tolerant control, which is more intelligent and efficient than the traditional control scheme that only focuses on the fault cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a fuel cell fault-tolerant control method based on reinforcement learning provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of the structure of a fuel cell and its detection component and control component provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0018] Embodiment 1: like Figure 1 As shown, an embodiment of the present invention provides a fuel cell fault-tolerant control method based on reinforcement learning, comprising the following steps: Step S1: Acquire monitoring parameters of a fuel cell under operating conditions, and construct a state space based on the monitoring parameters.

[0019] like Figure 2 As shown, in this embodiment, the monitoring parameters include: the minimum value of the fuel cell single cell voltage , the average value of fuel cell cell voltage , fuel cell stack temperature , Hydrogen inlet temperature , Hydrogen inlet pressure , Hydrogen out-of-combustion pressure , Hydrogen inlet flow , Hydrogen medium pressure and the pressure drop on the anode side of the fuel cell , .

[0020] Minimum fuel cell voltage It can reflect the balance of the battery pack. The average value of the fuel cell single cell voltage Can reflect the health status of the battery pack. Fuel cell stack temperature If it is too high or too low, it will affect the working efficiency and life of the fuel cell stack. For convenience, this embodiment collects the coolant outlet temperature as the fuel cell stack temperature. If it is too low, it may cause condensation. This reflects that the pressure drop may be abnormal due to pipeline blockage or valve failure. In other optional implementations, other monitoring parameters can also be selected and monitoring methods can be formulated.

[0021] Based on the above monitoring parameters, the state space for: ; In the formula, For the moment.

[0022] Step S2: Obtain executable operations under the operating state of the fuel cell, and construct an action space based on the executable operations.

[0023] like Figure 2 As shown, in this embodiment, the executable operations include: controlling the coolant thermostat opening , Control the opening of the hydrogen inlet proportional valve , Control the switching frequency of the nitrogen exhaust valve , control the drain valve switching frequency , Control the speed of hydrogen circulation pump And control the speed of the air compressor .

[0024] By controlling the coolant thermostat opening Adjust the temperature of the fuel cell stack by controlling the opening of the hydrogen inlet proportional valve Adjust the hydrogen flow into the reactor by controlling the switching frequency of the nitrogen exhaust valve Adjust the nitrogen concentration. Excessive nitrogen accumulation will affect the purity and concentration of hydrogen. By controlling the drain valve switching frequency Adjust the water volume. Excessive water accumulation will affect the purity and concentration of hydrogen. By controlling the speed of the hydrogen circulation pump Adjust the hydrogen supply rate and utilization by controlling the air compressor Adjusting the reaction efficiency of hydrogen and oxygen. In other optional implementations, other executable operations may also be selected.

[0025] Based on the above executable operations, the action space for: ; In the formula, For the moment.

[0026] Step S3: With the goal of minimizing the control cost, a reward function is constructed based on the state space and action space.

[0027] The reward function defines the reward for each state-action pair, which is used to evaluate the effect of each action and guide the algorithm to optimize the control strategy. In fuel cell systems, action overhead and failure cost are key factors in measuring system performance and economy, especially in fault-tolerant control. In the method, the specific reward function design takes into account action overhead, system performance and failure cost. Since the control cost needs to be minimized, the greater the action overhead and failure cost, the smaller the reward.

[0028] Specifically in this embodiment, the reward function for: ; In the formula, for Moment-by-moment action overhead, failure cost, and smooth control cost; ; In the formula, is the weight coefficient of action cost, is the working power consumption of the air compressor and hydrogen circulation pump, Adjust the power consumption for the opening of the coolant thermostat and the hydrogen inlet proportional valve. The switching power consumption of nitrogen exhaust valve and water drain valve; ; In the formula, is the weight coefficient of the failure cost, is the fault cost corresponding to the minimum and average values ​​of the fuel cell cell voltage, is the failure cost corresponding to the fuel cell stack temperature and the hydrogen inlet temperature, is the failure cost corresponding to the hydrogen inlet pressure, hydrogen outlet pressure, and hydrogen medium pressure, is the failure cost corresponding to the hydrogen inflow rate, is the failure cost corresponding to the pressure drop on the anode side of the fuel cell; ; In the formula, is the weight coefficient for controlling the smoothness, for Moment of action.

[0029] The purpose of setting the smooth control cost is to avoid frequent action changes. The more drastic the action changes, the higher the cost.

[0030] Step S4: construct a fault-tolerant control strategy model based on the DDPG algorithm, and train and verify the fault-tolerant control strategy model through state space, action space and reward function.

[0031] The DDPG algorithm combines the advantages of deterministic policy gradients and deep reinforcement learning, and can make effective decisions in complex continuous control problems. The DDPG algorithm can handle high-dimensional, continuous action spaces, which is very useful in complex systems such as fuel cell control. At the same time, the use of target networks and soft update techniques increases the stability of policy training. In addition, the DDPG algorithm uses experience replay to calculate the target Q value, which can effectively utilize previous experience, reduce the correlation between samples, and improve training efficiency. The DDPG algorithm is an algorithm based on the Actor-Critic architecture, which can generate actions through the policy (Actor) network and estimate the value of the current policy through the value (Critic) network: Actor Network: Input Current State , output an action , that is, the control action that the system should take in the current state. By updating the policy parameters, the action selected each time Capable of delivering the highest possible future rewards.

[0032] Critic network: input current state and action, output state-action Q value , which is used to evaluate the quality of the current action and represents the cumulative reward for executing the action in this state.

[0033] The specific processing process is as follows: The fault-tolerant control strategy model based on the DDPG algorithm includes the Actor network , Critic Network , Target Actor Network And the target critic network ; Training and validating the fault-tolerant control strategy model through state space, action space, and reward function includes: (1) Initialize the Actor network Parameters , Critic Network Parameters , Target Actor Network Parameters , Target Critic Network Parameters ; (2) Initialize training rounds , repeat the following steps (3)-(4): (3) Initialize The fuel cell running status of the wheel, get the status at the initial moment ; (4) Based on status For The fault-tolerant control strategy model of the first round is trained to obtain The fault-tolerant control strategy model of the wheel is ; (5) If , for The performance of the fault-tolerant control strategy model of the round is verified. If the performance verification passes, the final fault-tolerant control strategy model is obtained.

[0034] In step (4), based on the state Training the fault-tolerant control strategy model includes: (4.1) Initialization , enter the sampling phase, and repeat the following steps (4.2)-step (4.3): (4.2) Based on status , through the Actor network Select Action , for noise; (4.3) Execution of actions After that, get the reward and status , generate experience samples , and store it in the experience playback buffer ,make ; (4.4) When , For the preset total number of moments, enter the training phase: (4.5) From the experience replay buffer Randomly extract small batches of experience samples as training samples; (4.6) Based on the training samples, update the Critic network by minimizing the loss function based on the temporal difference error Parameters , loss function for: ; ; In the formula, is the discount factor, For the training samples, is the total number of training samples, For the The estimation of training samples value; (4.7) If , is a constant term, For modulo operation, the Actor network is updated by the policy gradient method Parameters , policy gradient for: ; (4.8) Based on parameters and Update Target Actor Network Parameters and target critic network Parameters : ; ; In the formula, is the update coefficient.

[0035] In step (5), The performance of the fault-tolerant control strategy model of the first round is verified. The verification phase is used to evaluate the performance and robustness of the training model. During the verification process, the model no longer updates parameters, but makes decisions through a fixed strategy network and observes its performance in the simulation environment. The evaluation indicators include the system's output power deviation, fault handling capability, fault recovery time, hydrogen supply stability, etc. Specific evaluation indicators can be formulated according to actual needs.

[0036] Step S5: performing fuel cell fault-tolerant control using the trained and verified fault-tolerant control strategy model.

[0037] Embodiment 2:

[0038] Based on the fuel cell fault-tolerant control method proposed in the first embodiment, the present invention provides a fuel cell fault-tolerant control device based on reinforcement learning, including: A state space construction module is configured to obtain monitoring parameters under the operating state of the fuel cell and construct a state space based on the monitoring parameters; An action space construction module is configured to obtain executable operations under the operating state of the fuel cell and construct an action space based on the executable operations; A reward function building module, configured to build a reward function based on a state space and an action space; The strategy model building module is configured to build a fault-tolerant control strategy model based on the DDPG algorithm with the goal of minimizing the control cost; A model training and verification module is configured to train and verify the fault-tolerant control strategy model through state space, action space and reward function; The model deployment application module is configured to perform fuel cell fault-tolerant control through the trained and verified fault-tolerant control strategy model.

[0039] Embodiment three:

[0040] Based on the fuel cell fault-tolerant control method proposed in Embodiment 1, an embodiment of the present invention provides an electronic device, including a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0041] Embodiment 4:

[0042] Based on the fuel cell fault-tolerant control method proposed in the first embodiment, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the above method are implemented.

[0043] Embodiment five:

[0044] Based on the fuel cell fault-tolerant control method proposed in the first embodiment, an embodiment of the present invention provides a computer program product, including a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0045] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0046] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0047] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0048] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0049] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A fuel cell fault-tolerant control method based on reinforcement learning, characterized in that: include: Acquiring monitoring parameters under the operating state of the fuel cell, and constructing a state space based on the monitoring parameters; Obtaining executable operations under the operating state of the fuel cell, and constructing an action space based on the executable operations; Taking minimizing control cost as a goal, constructing a reward function based on the state space and the action space; Constructing a fault-tolerant control strategy model based on the DDPG algorithm, and training and verifying the fault-tolerant control strategy model through the state space, the action space, and the reward function; Fuel cell fault-tolerant control is performed using the fault-tolerant control strategy model after training and verification.

2. The fuel cell fault-tolerant control method based on reinforcement learning according to claim 1, characterized in that: The monitoring parameters include: Minimum fuel cell voltage , the average value of fuel cell cell voltage , fuel cell stack temperature , Hydrogen inlet temperature , Hydrogen inlet pressure , Hydrogen out-of-combustion pressure , Hydrogen inlet flow , Hydrogen medium pressure and the pressure drop on the anode side of the fuel cell ; The state space for: ； In the formula, For the moment.

3. The fuel cell fault-tolerant control method based on reinforcement learning according to claim 2, characterized in that: The executable operations include: Control the coolant thermostat opening , Control the opening of the hydrogen inlet proportional valve , Control the switching frequency of the nitrogen exhaust valve , control the drain valve switching frequency , Control the speed of hydrogen circulation pump And control the speed of the air compressor ; The action space for: ； In the formula, For the moment.

4. The fuel cell fault-tolerant control method based on reinforcement learning according to claim 3 is characterized in that: The reward function for: ； In the formula, for Moment-by-moment action overhead, failure cost, and smooth control cost; ； In the formula, is the weight coefficient of action cost, is the working power consumption of the air compressor and hydrogen circulation pump, Adjust the power consumption for the opening of the coolant thermostat and the hydrogen inlet proportional valve. The switching power consumption of nitrogen exhaust valve and water drain valve; ； In the formula, is the weight coefficient of the failure cost, is the fault cost corresponding to the minimum and average values ​​of the fuel cell cell voltage, is the failure cost corresponding to the fuel cell stack temperature and the hydrogen inlet temperature, is the failure cost corresponding to the hydrogen inlet pressure, hydrogen outlet pressure, and hydrogen medium pressure, is the failure cost corresponding to the hydrogen inflow rate, is the failure cost corresponding to the pressure drop on the anode side of the fuel cell; ； In the formula, is the weight coefficient for controlling the smoothness, for Moment of action.

5. The fuel cell fault-tolerant control method based on reinforcement learning according to claim 4, characterized in that: The fault-tolerant control strategy model based on the DDPG algorithm includes an Actor network , Critic Network , Target Actor Network And the target critic network ; The training and verification of the fault-tolerant control strategy model through the state space, the action space and the reward function includes: Initializing the Actor Network Parameters , Critic Network Parameters , Target Actor Network Parameters , Target Critic Network Parameters ; Initialize training rounds , repeat the following steps: Initialize The fuel cell running status of the wheel, get the status at the initial moment ; Based on status For The fault-tolerant control strategy model of the first round is trained to obtain The fault-tolerant control strategy model of the wheel is ; like , for The fault-tolerant control strategy model of the round is subjected to performance verification. If the performance verification passes, the final fault-tolerant control strategy model is obtained.

6. The fuel cell fault-tolerant control method based on reinforcement learning according to claim 5, characterized in that: The state-based Training the fault-tolerant control strategy model includes: initialization , enter the sampling phase and repeat the following steps: Based on status , through the Actor network Select Action , for noise; Execute an action After that, get the reward and status , generate experience samples , and store it in the experience playback buffer ,make ; when , For the preset total number of moments, enter the training phase: From the Experience Replay Buffer Randomly extract small batches of experience samples as training samples; Based on the training samples, the Critic network is updated by minimizing the loss function based on the temporal difference error. Parameters , loss function for: ； ； In the formula, is the discount factor, For the training samples, is the total number of training samples, For the The estimation of training samples value; like , is a constant term, For modulo operation, the Actor network is updated by the policy gradient method Parameters , policy gradient for: ； Based on parameters and Update Target Actor Network Parameters and target critic network Parameters : ； ； In the formula, is the update coefficient.

7. A fuel cell fault-tolerant control device based on reinforcement learning, characterized in that: include: A state space construction module is configured to obtain monitoring parameters under the operating state of the fuel cell and construct a state space based on the monitoring parameters; An action space construction module is configured to obtain executable operations under the operating state of the fuel cell and construct an action space based on the executable operations; A reward function construction module, configured to construct a reward function based on the state space and the action space; The strategy model building module is configured to build a fault-tolerant control strategy model based on the DDPG algorithm with the goal of minimizing the control cost; A model training and verification module, configured to train and verify the fault-tolerant control strategy model through the state space, the action space and the reward function; The model deployment application module is configured to perform fuel cell fault-tolerant control through the fault-tolerant control strategy model after training and verification.

8. An electronic device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images