Methods, systems, equipment, and media for power supply control of hydrogen-powered unmanned systems.
By acquiring task information and urgency indicators, energy allocation schemes are generated and screened, solving the problem of insufficient adaptability of energy control in the concurrent execution of multiple tasks in hydrogen-powered unmanned systems. This achieves the feasibility of task execution and the rational allocation of resources, improving the dynamic adaptability of the system.
Patent Information
- Application Number
- CN202611181267.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-05
- Publication Date
- 2026-11-17
AI Technical Summary
In the scenario of power supply control for hydrogen-powered unmanned systems that execute multiple tasks concurrently, the existing technology is not adaptable enough to the needs of multi-task execution, resulting in problems such as emergency tasks timeouts due to insufficient energy and low-priority tasks consuming too many resources.
By acquiring current allocable energy and task information, determining urgency indicators, generating multiple candidate energy allocation schemes, and using an energy state prediction model to predict the energy state evolution results of each scheme, the final energy control scheme is selected by combining preset operating constraints to ensure the feasibility of task execution and the rationality of energy allocation.
The energy supply control scheme has been improved to adapt to the needs of multi-task execution, reducing task delays and resource waste, and ensuring energy supply and task completion rate for emergency tasks.
Smart Images

Figure CN123121899A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of unmanned system control technology, and in particular to a method, system, device and medium for power supply control of unmanned systems based on hydrogen power. Background Technology
[0002] With the increasing application of unmanned systems such as hydrogen-powered drones and hydrogen-powered inspection robots in power grid inspection, equipment status data collection, fault verification, and remote operation and maintenance, these systems often need to simultaneously perform multiple tasks, including path movement, inspection data collection, target verification, data processing and transmission, and safety assurance. Different tasks vary significantly in execution time, energy requirements, and importance. Therefore, the power supply system not only needs to meet the platform's continuous operation requirements but also needs to adapt to the dynamic power supply needs under the parallel or alternating execution of multiple tasks.
[0003] In the field of power supply control for hydrogen-powered unmanned systems, existing technologies mainly focus on single objectives such as improving energy utilization efficiency, extending system range, and maintaining the state of charge balance of lithium batteries. For example, these technologies employ rule-based strategies to pre-set power allocation logic, optimization strategies to solve for optimal power allocation in real time, and reinforcement learning to enable the power supply system to adapt to flight conditions. However, when an unmanned system needs to simultaneously execute multiple concurrent tasks with different time limits, energy requirements, and urgency levels, existing methods lack assessment of the feasibility of each task. They can only allocate power based on energy efficiency indicators, failing to effectively match limited system energy with the urgency and feasibility of each task. This easily leads to problems such as urgent tasks exceeding time limits due to insufficient energy and low-priority tasks consuming excessive resources, indicating insufficient adaptability. Therefore, in the scenario of power supply control for hydrogen-powered unmanned systems with multiple concurrent tasks, improving the adaptability of the power supply control scheme to the demands of multi-task execution has become an urgent technical problem to be solved. Summary of the Invention
[0004] This application provides a method, system, device, and medium for power supply control of hydrogen-powered unmanned systems, which can solve the problem that the existing power supply control scheme has low adaptability to the multi-task execution requirements in the scenario of power supply control of hydrogen-powered unmanned systems with multiple tasks running concurrently.
[0005] This application provides a hydrogen-powered unmanned system power supply control method in some embodiments, including: The system acquires the allocable energy of the unmanned system in the current control cycle and the task information of each task to be executed, and determines the urgency index of each task to be executed based on the task information. The feasibility classification results for completing each of the tasks to be executed are determined based on the task information, the allocable energy, and the urgency index. Based on the completed feasibility classification results, multiple candidate energy supply allocation schemes are generated, and based on the pre-built energy supply status prediction model, the energy supply status evolution result of each candidate energy supply allocation scheme is predicted. Based on the energy supply state evolution results, an initial energy supply control scheme is obtained by scoring and selecting the best from each of the candidate energy supply allocation schemes, and the initial energy supply control scheme is modified based on preset operating constraints to determine the final energy supply control scheme. The final power supply control scheme controls the unmanned system to execute each of the tasks to be performed.
[0006] Compared to existing technologies, the above embodiments have the following beneficial effects: In each control cycle, the currently allocable energy and task information of each task to be executed are first summarized, and the urgency index of each task is calculated. Based on this, the feasibility of each task is classified by combining allocable energy and urgency, distinguishing whether the task has the conditions for timely completion. For different categories, multiple candidate energy allocation schemes are generated, and the evolution of the energy state of each scheme in the next few cycles is predicted in advance using an energy state prediction model. Through this prediction, the initial scheme with better performance in terms of energy state evolution can be selected, and then the initial scheme is modified in combination with the constraints of actual operation, finally obtaining a practically executable energy control scheme. This process links task execution requirements, energy allocation feasibility, and future state change trends, making the energy supply scheme more adaptable to the dynamic changes when multiple tasks are executed concurrently, reducing task delays or resource waste caused by mismatch between energy allocation and task requirements.
[0007] Further, the candidate energy allocation schemes include: the planned allocated power of the hydrogen fuel cell, the planned discharge power of the lithium battery, and the task energy quota for each of the tasks to be executed; the energy supply status prediction model includes: a hydrogen energy output prediction module, a lithium battery power prediction module, and a task execution progress prediction module; the prediction of the energy supply status evolution result of each candidate energy allocation scheme based on the pre-constructed energy supply status prediction model includes: Obtain the actual output power of the hydrogen fuel cell, the remaining charge percentage of the lithium battery, and the current remaining execution amount of each task to be executed in the current control cycle; The maximum variable power of a single control cycle of the hydrogen fuel cell is obtained. The maximum variable power, the actual output power, and the planned allocated power are input into the hydrogen energy output prediction module to recursively predict the predicted output power of several future control cycles, thereby obtaining the first energy supply state evolution result of the hydrogen fuel cell. The planned discharge power and the remaining power ratio are input into the lithium battery power prediction module to recursively predict the predicted remaining power ratio for several future control cycles, thereby obtaining the second power supply state evolution result of the lithium battery. For each task to be executed, the task energy quota and the current remaining execution amount of the task to be executed are input into the task execution progress prediction module to recursively predict the predicted remaining execution amount for several future control cycles, thereby obtaining the third energy supply state evolution result of the task to be executed.
[0008] Compared to existing technologies, the above embodiments offer the following advantages: They refine candidate energy allocation schemes to the planned power allocation of hydrogen fuel cells, the planned discharge power of lithium batteries, and the task energy quota for each task. Simultaneously, the energy supply status prediction model is divided into three relatively independent parts: a hydrogen energy output prediction module, a lithium battery power prediction module, and a task execution progress prediction module. By obtaining the actual output power, remaining power ratio, and remaining execution amount of each task in the current control cycle, and combining this with the maximum variable power of the hydrogen fuel cell within a single control cycle, the output power, remaining power ratio, and remaining execution amount for multiple future cycles are recursively predicted. This modular prediction method allows the changing trends of each type of state variable to be tracked and evaluated independently, avoiding the computational burden caused by complex coupled models. Furthermore, the recursive prediction can anticipate potential problems such as power mismatch, low power, or task delays that may arise in future scenarios for different candidate allocation schemes, thus improving the accuracy of decision-making.
[0009] Further, the step of obtaining an initial energy supply control scheme by scoring and selecting the best from each of the candidate energy supply allocation schemes based on the energy supply state evolution results includes: For each of the candidate power allocation schemes, a first score for the candidate power allocation scheme is determined based on the difference between the planned allocated power and the corresponding predicted output power of the candidate power allocation scheme. For each of the candidate energy allocation schemes, a second score for the candidate energy allocation scheme is determined based on the difference between the predicted remaining power ratio corresponding to the candidate energy allocation scheme and the preset minimum allowable remaining power. For each of the candidate power allocation schemes, a third score for the candidate power allocation scheme is determined based on the difference between the predicted remaining execution amount and the preset remaining execution amount threshold for each task to be executed in the candidate power allocation scheme. Based on the first score, the second score, and the third score, the candidate energy allocation schemes are scored and selected for the best one to obtain the initial energy control scheme.
[0010] Compared to existing technologies, the above embodiments have the following beneficial effects: The first score examines the difference between the planned allocated power and the predicted output power; the smaller the difference, the better the hydrogen fuel cell output follows the allocation plan, and the less pressure there is on power regulation. The second score compares the predicted remaining power ratio with the preset minimum allowable remaining power; the larger the difference, the more ample the safety margin of the lithium battery, and the less likely it is to run out of power. The third score compares the predicted remaining execution amount of each task with the preset threshold; the smaller the difference, the greater the certainty of completing the task on schedule. By combining these three scores to select the best candidate scheme, it is possible to avoid compromising other aspects of the control scheme by excessively pursuing a single indicator, resulting in an initial power supply control scheme with better overall performance.
[0011] Further, the completion feasibility classification results include: a first category of tasks that meet the conditions for timely completion and a second category of tasks that do not meet the conditions for timely completion; the task information includes: the task energy requirement of each of the tasks to be executed; the generation of multiple candidate energy allocation schemes based on the completion feasibility classification results includes: Obtain the preset rated output power range of the hydrogen fuel cell, and determine the feasible power traversal range of the hydrogen fuel cell based on the maximum variable power and the preset rated output power range. Multiple different power values are traversed within the power traversal interval, which are used as the planned power allocation of the hydrogen fuel cell in each of the candidate power allocation schemes. Based on the allocable energy and the planned allocated power in the current control cycle, the planned discharge power of the lithium battery in the corresponding candidate power allocation scheme is determined; Allocate the corresponding energy requirement of each of the first type of tasks as the corresponding task energy quota in each of the candidate energy allocation schemes; Based on the remaining allocable energy and the urgency index corresponding to each of the second type of tasks, the task energy quota corresponding to each of the candidate energy allocation schemes for each of the second type of tasks is determined.
[0012] Compared to existing technologies, the above embodiments have the following advantages: First, the rated output power range and maximum variable power per cycle of the hydrogen fuel cell are obtained, thereby determining a feasible power traversal range that can be safely and dynamically adjusted. Within this range, multiple discrete power values are selected as the planned allocation power in the candidate schemes. Then, based on the difference between the allocable energy in the current control cycle and the planned allocation power, the planned discharge power corresponding to the lithium battery is determined. For tasks that have been classified as having the conditions for timely completion, the energy quota is directly allocated in full according to their task energy requirements, because these tasks have relatively abundant time and energy, and prioritizing their successful completion will not consume too many resources. For tasks that do not have the conditions for timely completion, the energy quota is allocated according to the remaining allocable energy and the urgency index of each task. Tasks with higher urgency can obtain more remaining energy. This approach avoids spreading limited energy evenly across all tasks, leading to energy shortages for critical tasks, and also prevents the waste caused by allocating energy to tasks that are clearly unable to be completed on time, thus enabling the candidate schemes to have a relatively reasonable energy allocation logic in the initial stage of generation.
[0013] Further, the task information includes: time limit parameters and estimated completion time corresponding to the tasks to be executed; the step of determining the completion feasibility classification result of each task to be executed based on the task information, the allocable energy, and the urgency index includes: Based on the time limit parameters, the estimated completion time, the task energy requirements, and the available energy, determine the time gap and energy gap for each of the tasks to be executed. When the time gap is less than or equal to a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the first type of task that has the conditions for timely completion. When the time gap is greater than a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient time. When the time gap is less than or equal to a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient energy. When the time gap is greater than a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be a second type of task that has both insufficient time and insufficient energy.
[0014] Compared to existing technologies, the above embodiments have the following advantages: When determining whether a task meets the conditions for timely completion, measurements are taken from two independent dimensions: time and energy. Using the time limit parameters, estimated completion time, task energy requirements, and currently available energy in the task information, the time gap and energy gap for each task are calculated. Then, the time gap and energy gap are compared with preset time thresholds and energy thresholds, respectively, resulting in four different classifications: no shortage of either time or energy, insufficient time only, insufficient energy only, and insufficient in both. This classification method is more refined than simply dividing based on remaining time or energy reserves, and can help the subsequent energy allocation mechanism identify the specific dimensions of the shortage for each task.
[0015] Further, determining the urgency index of each of the tasks to be executed based on the task information includes: Calculate the remaining time for each task to be executed based on the time limit parameters of each task to be executed and the current time of the current control cycle; When the remaining time is less than or equal to the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the first urgency value; When the remaining time is greater than the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the second urgency value; wherein, the urgency of the first urgency value is higher than that of the second urgency value.
[0016] Compared with the prior art, the above embodiments have the following beneficial effects: This application can determine the urgency of the task by comparing the remaining time with the expected completion time. The calculation is simple and has high real-time performance. It does not require complex parameter calibration or training and is easy to execute quickly in actual unmanned systems.
[0017] Furthermore, the step of modifying the initial power supply control scheme based on preset operating constraints to determine the final power supply control scheme includes: When the initial power supply control scheme violates the preset operating constraints, the task energy quota of each task to be executed in the initial power supply control scheme is reduced according to the urgency index, and the target task energy quota of each task to be executed is determined. Based on the energy quota of each target task, the planned allocated power and the planned discharge power in the initial power supply control scheme are adjusted to obtain the target planned allocated power and the target planned discharge power, respectively. The final power supply control scheme is determined based on the target planned power allocation, the target planned discharge power, and the target task energy quota for each of the tasks to be executed.
[0018] Compared to existing technologies, the above embodiments have the following advantages: When revising the initial energy supply control scheme, the energy quota for the tasks to be performed is reduced in an orderly manner according to the urgency index. After determining the target energy quota for each task after reduction, the planned power allocation of the hydrogen fuel cell and the planned discharge power of the lithium battery are adjusted in reverse according to the new energy allocation situation, so that they fall back within the allowable operating range. This ensures that the final energy supply control scheme satisfies the hard constraints while protecting the energy supply for urgent tasks as much as possible. Compared to directly reverting to a conservative scheme or reducing the overall power, it achieves a better trade-off between safety and task completion rate.
[0019] Another embodiment of this application provides a hydrogen-powered unmanned system power supply control system, which is used to implement the steps of the hydrogen-powered unmanned system power supply control method of this application.
[0020] Another embodiment of this application also provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the hydrogen-powered unmanned system power supply control method of this application.
[0021] Another embodiment of this application also provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of the hydrogen-powered unmanned system power supply control method of this application. Attached Figure Description
[0022] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0023] Figure 1 This is a flowchart illustrating a hydrogen-powered unmanned system power supply control method provided in some embodiments of this application. Detailed Implementation
[0024] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.
[0026] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.
[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0028] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.
[0029] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).
[0030] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.
[0031] In the field of power supply control for hydrogen-powered unmanned systems, existing technologies mainly focus on single objectives such as improving energy utilization efficiency, extending system range, and maintaining the state of charge balance of lithium batteries. For example, they employ rule-based strategies to pre-set power allocation logic, optimization strategies to solve for optimal power allocation in real time, and reinforcement learning to enable the power supply system to adapt to flight conditions. However, when unmanned systems need to simultaneously execute multiple concurrent tasks with different time limits, energy requirements, and urgency levels, existing methods lack assessment of the feasibility of each task. They can only allocate power based on energy efficiency indicators, failing to effectively match limited system energy with the urgency and feasibility of each task. This easily leads to problems of insufficient adaptability, such as urgent tasks exceeding time limits due to insufficient energy and low-priority tasks consuming excessive resources.
[0032] Please refer to Figure 1 To address the challenge of improving the adaptability of existing energy supply control schemes to the demands of multi-task concurrent execution in hydrogen-powered unmanned systems, this application provides an embodiment of a hydrogen-powered unmanned system energy supply control method, comprising the following steps S101 to S105: S101: Obtain the allocable energy of the unmanned system in the current control cycle and the task information of each task to be executed, and determine the urgency index of each task to be executed based on the task information.
[0033] Preferably, in some embodiments of this application, the control cycle refers to the system repeatedly executing the "status acquisition—status processing—energy allocation—control output" cycle once at a preset fixed time interval, wherein the fixed time interval is denoted as... The control cycle is preset by task management software or power supply control software and can be determined based on platform type, task change rate, and energy response speed. For hydrogen-powered drones or hydrogen-powered inspection robots under unmanned systems, the control cycle is preferentially set to a short fixed cycle, enabling the system to periodically update the task status and power supply status, and adjust the output of the hydrogen fuel cell system and lithium battery pack in a timely manner.
[0034] Preferably, in some embodiments of this application, the start times of two adjacent control cycles are respectively denoted as... and ,in: ; That is, the system at every moment Perform a data collection and status cleanup for the current period.
[0035] Furthermore, in some embodiments of this application, the task information includes: the task energy requirement of each task to be executed, the time limit parameter corresponding to the task to be executed, and the estimated completion time.
[0036] Preferably, in some embodiments of this application, the task information further includes: task number, task type, task importance level, whether interruption is allowed, whether reduced execution intensity is allowed, and current execution progress.
[0037] Preferably, in some embodiments of this application, to avoid task type overlap, task types are divided into at least five categories according to the main execution purpose of the current task within the current control cycle: Path movement tasks, which refer to tasks whose main purpose is location transfer, including the movement process of flying along the inspection path, walking along the inspection route, going to the designated equipment location, and moving to or from the waypoint or return point; Inspection and data collection tasks, which refer to tasks whose main purpose is to obtain equipment status data, including visible light image acquisition, infrared image acquisition, partial discharge signal acquisition, gas information acquisition, sound information acquisition, and other monitoring information acquisition; Target verification tasks, which refer to tasks of reapproaching, re-photographing, re-testing, or multi-angle confirming the discovered anomalies, alarm points, or suspected defects; Data processing and transmission tasks, which refer to tasks whose main purpose is on-site data compression, edge processing, result uploading, alarm reporting, or link maintenance; Security assurance tasks, which refer to tasks whose main purpose is to maintain the safe operation of the platform, including safe hovering, safe standby, emergency landing, emergency stop, minimum communication maintenance, and minimum control maintenance. The above task types are distinct from each other, and the same task instance is assigned only one task type within the same control cycle; when the purpose of task execution changes, its task type is redefined in the next control cycle.
[0038] Preferably, in some embodiments of this application, it is also necessary to collect energy state data and system load data of the unmanned system. The energy state data is provided by the hydrogen fuel cell management module and the lithium battery management module, and includes at least: the actual output power of the hydrogen fuel cell, the maximum allowable output of the hydrogen fuel cell, the rate of change of hydrogen fuel cell output, the remaining hydrogen quantity, the remaining charge percentage of the lithium battery, and the allowable discharge power of the lithium battery. The system load data is provided by the flight control module, the motion control module, and each mission load drive module, and includes at least: the current drive load power, the current mission load power, the communication load power, the computing load power, and the power required for return or return to base estimated based on the current position and return path.
[0039] Preferably, in some embodiments of this application, after collecting the task information, energy status data, and system load data for the current control cycle, the data of each module is uniformly organized using a method of aligning the most recent time according to the control cycle window: a corresponding time window is established based on the start time of the current control cycle; for the data generated within this window, if a certain type of data is updated at least once in the current control cycle, the last valid update data in that cycle is selected as the representative value; if a certain type of data is not updated in the current control cycle, the most recent valid data from the previous control cycle is called as the continuation data and marked as a continuation of use; if a certain type of data shows multiple anomalies or missing values in the current control cycle, the most recent valid data is retained and an anomaly flag is output simultaneously for subsequent judgment of the reliability of the status quantity. Through the above method, data with different update frequencies and different collection times are uniformly organized into a set of system status data corresponding to the current control cycle, so that before the end of each control cycle, the system can obtain task information, energy status data, and system load data corresponding to that cycle one by one.
[0040] Preferably, in some embodiments of this application, the calculation process for allocable energy is as follows: First, the total power that the system can provide in the current control cycle is determined based on the current energy state set, which is the sum of the power that the hydrogen fuel cell can stably provide and the auxiliary power that the lithium battery can provide; then, the minimum load power required to maintain the basic operation of the system (such as flight control or motion control, communication, and basic computing consumption) is subtracted from this total power to obtain the power available for task allocation in the current control cycle. If this power value is less than or equal to zero, it indicates that there is no surplus power available for ordinary task allocation in the current cycle. Then, the power available for task allocation is multiplied by the duration of the current control cycle to obtain the allocable energy in the current cycle after deducting the basic operating power of the system.
[0041] Preferably, in some embodiments of this application, the estimated completion time of each task to be executed needs to be calculated separately according to the task type. For path movement tasks, it is estimated based on the remaining path length and the current allowed movement speed (provided by the motion control module, or determined based on the current allocable power, environmental conditions, and historical speed). For inspection and data collection tasks, it is calculated based on the number of remaining collection points and the average time required for a single collection point. This average time can be obtained from historical execution records. If historical data is insufficient, the default time in the preset task template is used. For target verification tasks, it is calculated based on the number of remaining verification points and the average time required for a single verification point. If the verification includes multiple actions such as approaching, stopping, re-taking, and re-testing, the time required for each action is added together. For data processing and transmission tasks, it is calculated comprehensively based on the amount of data to be processed, the current processing speed, the current communication rate, and the amount of data to be uploaded. For security assurance tasks, one-time security actions are determined according to the preset time required to complete the action, while security actions that need to be continuously maintained are determined according to the maintenance requirements within the current control cycle. Through the above methods, a more accurate estimated completion time can be obtained for different types of tasks, providing a basis for judging the urgency of subsequent tasks and classifying their feasibility.
[0042] Furthermore, in some embodiments of this application, determining the urgency index corresponding to each of the tasks to be executed based on the task information includes: Calculate the remaining time for each task to be executed based on the time limit parameters of each task to be executed and the current time of the current control cycle; When the remaining time is less than or equal to the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the first urgency value; When the remaining time is greater than the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the second urgency value; wherein, the urgency of the first urgency value is higher than that of the second urgency value.
[0043] Preferably, in some embodiments of this application, the urgency index is determined as follows: First, based on the time limit parameters (i.e., task deadlines) of each task to be executed and the current time of the current control cycle, the remaining time before the deadline for each task is calculated, specifically: Remaining time = Task deadline - Current time. Then, the estimated completion time of each task is obtained (this time can be estimated based on the remaining execution volume of the task and the current allowed execution speed, etc.). The remaining time is compared with the estimated completion time: If the remaining time is less than or equal to the estimated completion time, it means that the task may not be completed before the deadline according to the current progress. In this case, the urgency index of the task is determined as the first urgency value, indicating high urgency; if the remaining time is greater than the estimated completion time, it means that the task still has sufficient time leeway, and it is determined as the second urgency value, indicating low urgency. The urgency of the first urgency value is higher than that of the second urgency value, that is, high-urgency tasks are given priority in power allocation. It should be noted that in practical applications, the urgency level can also be further divided by combining factors such as task importance level and whether interruption is allowed, but this embodiment adopts the above two-level determination method, which is simple to calculate, has good real-time performance, and is easy to implement in engineering.
[0044] This application determines the urgency of a task by comparing the remaining time with the estimated completion time. The calculation is simple and has high real-time performance. It does not require complex parameter calibration or training and can be quickly executed in actual unmanned systems.
[0045] S102: Determine the completion feasibility classification result of each of the tasks to be executed based on the task information, the allocable energy, and the urgency index.
[0046] Furthermore, in some embodiments of this application, the completion feasibility classification results include: a first category of tasks that meet the conditions for timely completion and a second category of tasks that do not meet the conditions for timely completion.
[0047] Furthermore, in some embodiments of this application, determining the completion feasibility classification result of each of the tasks to be executed based on the task information, the allocable energy, and the urgency index includes: Based on the time limit parameters, the estimated completion time, the task energy requirements, and the available energy, determine the time gap and energy gap for each of the tasks to be executed. When the time gap is less than or equal to a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the first type of task that has the conditions for timely completion. When the time gap is greater than a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient time. When the time gap is less than or equal to a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient energy. When the time gap is greater than a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be a second type of task that has both insufficient time and insufficient energy.
[0048] Preferably, in some embodiments of this application, feasibility classification is mainly based on two indicators: time gap and energy gap. First, the remaining time is calculated based on the time limit parameter (i.e., task deadline) of each task to be executed and the current time. Then, this remaining time is compared with the expected completion time of the task to obtain the time gap: Time gap = Expected completion time - Remaining time. A positive value indicates insufficient time. Simultaneously, the minimum energy required to complete the task (estimated from the expected completion time and average power requirement) is compared with the energy available for task allocation within the current control cycle to obtain the energy gap: Energy gap = Task energy requirement - Allocable energy. A positive value indicates insufficient energy. The preset time threshold and preset energy threshold can be flexibly set according to the actual system response capability and task type, and are typically zero or small positive numbers. Then, tasks are categorized according to the following rules: When the time gap is less than or equal to a preset time threshold and the energy gap is less than or equal to a preset energy threshold, it is classified as a first-category task that meets the conditions for timely completion; when the time gap is greater than the preset time threshold and the energy gap is less than or equal to the preset energy threshold, it is classified as a second-category task with insufficient time; when the time gap is less than or equal to the preset time threshold and the energy gap is greater than the preset energy threshold, it is classified as a second-category task with insufficient energy; when the time gap is greater than the preset time threshold and the energy gap is greater than the preset energy threshold, it is classified as a second-category task with both insufficient time and insufficient energy. This classification allows for differentiation of whether the current bottleneck of a task is time, energy, or both, providing a clear basis for generating candidate energy allocation schemes.
[0049] This application measures task completion on time from two independent dimensions: time and energy. Using time limits, estimated completion time, energy requirements, and available energy from the task information, it calculates the time and energy gaps for each task. These gaps are then compared to preset time and energy thresholds, resulting in four different classifications: no time or energy shortages, time shortage only, energy shortage only, and both shortages. This classification method is more refined than simply using remaining time or energy, helping the subsequent energy allocation mechanism identify the specific dimensions of each task's shortage.
[0050] S103: Generate multiple candidate energy supply allocation schemes based on the completed feasibility classification results, and predict the energy supply status evolution result of each candidate energy supply allocation scheme based on the pre-built energy supply status prediction model.
[0051] Furthermore, in some embodiments of this application, the step of generating multiple candidate energy allocation schemes based on the completed feasibility classification results includes: Obtain the preset rated output power range of the hydrogen fuel cell, and determine the feasible power traversal range of the hydrogen fuel cell based on the maximum variable power and the preset rated output power range. Multiple different power values are traversed within the power traversal interval, which are used as the planned power allocation of the hydrogen fuel cell in each of the candidate power allocation schemes. Based on the allocable energy and the planned allocated power in the current control cycle, the planned discharge power of the lithium battery in the corresponding candidate power allocation scheme is determined; Allocate the corresponding energy requirement of each of the first type of tasks as the corresponding task energy quota in each of the candidate energy allocation schemes; Based on the remaining allocable energy and the urgency index corresponding to each of the second type of tasks, the task energy quota corresponding to each of the candidate energy allocation schemes for each of the second type of tasks is determined.
[0052] Preferably, in some embodiments of this application, the process of generating multiple candidate energy allocation schemes is as follows: First, the preset rated output power range of the hydrogen fuel cell is obtained, and combined with the maximum variable power within a single control cycle, a feasible power traversal interval is determined. Multiple discrete power values are taken within this interval as the planned allocation power of the hydrogen fuel cell in each candidate scheme, with the difference between adjacent values not exceeding the maximum variable power. Then, based on the difference between the allocable energy in the current control cycle and the planned allocated power, the planned discharge power corresponding to the lithium battery is determined. For the first category of tasks that meet the conditions for timely completion in the classification results, their energy requirements are directly allocated in full as their task energy quota in each candidate scheme; for the second category of tasks that do not meet the conditions for timely completion, after deducting the energy already allocated to the first category of tasks, the remaining allocable energy is allocated according to the urgency index of each second category of task. Tasks with higher urgency receive more energy quotas, thus forming multiple different combinations of task energy quotas. Through the above method, multiple candidate energy allocation schemes that simultaneously include the planned allocated power of the hydrogen fuel cell, the planned discharge power of the lithium battery, and the energy quotas of each task can be generated.
[0053] This application first obtains the rated output power range and maximum variable power per cycle of the hydrogen fuel cell, thereby determining a feasible power traversal range that can be safely and dynamically adjusted. Within this range, multiple discrete power values are selected as the planned allocation power in the candidate schemes. Then, based on the difference between the allocable energy in the current control cycle and the planned allocation power, the planned discharge power corresponding to the lithium battery is determined. For tasks that have been classified as having the conditions for timely completion, energy quotas are directly allocated in full according to their task energy requirements, because these tasks have relatively abundant time and energy, and prioritizing their successful completion will not consume too many resources. For tasks that do not have the conditions for timely completion, energy quotas are allocated based on the remaining allocable energy and the urgency index of each task. Tasks with higher urgency can receive more remaining energy. This approach avoids spreading limited energy evenly across all tasks, leading to energy shortages for critical tasks, and also prevents the waste of energy by allocating it to tasks that are clearly unable to be completed on time, thus giving the candidate schemes a relatively reasonable energy allocation logic in the initial stage of generation.
[0054] Further, in some embodiments of this application, the candidate energy allocation scheme includes: the planned allocated power of the hydrogen fuel cell, the planned discharge power of the lithium battery, and the task energy quota for each of the tasks to be executed; the energy supply status prediction model includes: a hydrogen energy output prediction module, a lithium battery power prediction module, and a task execution progress prediction module; the step of predicting the energy supply status evolution result of each candidate energy allocation scheme according to the pre-constructed energy supply status prediction model includes: Obtain the actual output power of the hydrogen fuel cell, the remaining charge percentage of the lithium battery, and the current remaining execution amount of each task to be executed in the current control cycle; The maximum variable power of a single control cycle of the hydrogen fuel cell is obtained. The maximum variable power, the actual output power, and the planned allocated power are input into the hydrogen energy output prediction module to recursively predict the predicted output power of several future control cycles, thereby obtaining the first energy supply state evolution result of the hydrogen fuel cell. The planned discharge power and the remaining power ratio are input into the lithium battery power prediction module to recursively predict the predicted remaining power ratio for several future control cycles, thereby obtaining the second power supply state evolution result of the lithium battery. For each task to be executed, the task energy quota and the current remaining execution amount of the task to be executed are input into the task execution progress prediction module to recursively predict the predicted remaining execution amount for several future control cycles, thereby obtaining the third energy supply state evolution result of the task to be executed.
[0055] Preferably, in some embodiments of this application, the remaining execution volume of each task to be executed needs to be calculated separately according to its task type. Specifically: for path movement tasks, it is determined based on the remaining path length between the current position of the unmanned system and the target position; for inspection and data collection tasks, it is determined based on the incomplete inspection and data collection content (such as the number of uncaptured images, unmeasured points, etc.); for target verification tasks, it is determined based on the number of targets to be verified, the number of angles to be recaptured, the number of remeasurements, or the duration of observation to be observed; for data processing and data transmission tasks, it is determined based on the amount of data not yet processed, the amount of data not yet uploaded, or the number of alarm messages to be sent; for safety assurance tasks, it is determined based on the safety actions not yet completed (such as emergency landing, emergency stop, etc.) or the safety status that still needs to be maintained (such as safe hovering, minimum communication maintenance, etc.). In the above way, the execution progress of different types of tasks can be uniformly quantified into remaining execution volume, which is convenient for subsequent energy allocation and progress prediction.
[0056] In some embodiments of this application, the power supply status prediction model is a lightweight, short-term recursive prediction model established for the real-time arbitration process. It reflects the changing trends of hydrogen fuel cell output, lithium battery remaining charge ratio, and task execution progress over several control cycles under the current task and power supply status. The model's parameters are pre-determined offline based on historical operational data statistics and equipment parameters. Specifically, it is based at least on the power change response characteristics of the hydrogen fuel cell at different output levels, the charge reduction pattern of the lithium battery at different discharge powers, the energy consumption pattern corresponding to a unit execution amount during different types of task execution, and the minimum execution energy requirement under different task urgency levels. During system operation, the control unit uses the actual state of the current control cycle as the initial prediction value and progressively calculates the state changes of each subsequent prediction step according to a preset recursive rule, thereby obtaining the prediction result corresponding to each candidate power supply allocation scheme. Furthermore, the model's parameters are also corrected during operation based on the actual execution results to maintain prediction accuracy.
[0057] Preferably, in some embodiments of this application, for each candidate energy allocation scheme, it is necessary to predict the changes in energy supply status over multiple future control cycles. Specifically, the following steps are taken: First, obtain the actual output power of the hydrogen fuel cell, the remaining capacity ratio of the lithium battery, and the current remaining execution amount of each task for the current control cycle. Simultaneously, obtain the maximum variable power of the hydrogen fuel cell within a single control cycle (i.e., the maximum increase in output power within one cycle). Input the actual output power, planned allocation power, and maximum variable power into the hydrogen energy output prediction module, and recursively calculate the output power for each cycle: if the planned allocation power is higher than the current actual output, the output power is gradually increased, but the increase per cycle does not exceed the maximum variable power; if the planned allocation power is lower than or equal to the current output, it is adjusted according to the planned value or a preset decrease rule, thereby obtaining the predicted output power for each future cycle (first energy supply status evolution result). Input the planned discharge power and the current remaining capacity ratio of the lithium battery into the lithium battery capacity prediction module, calculate the amount of electricity reduced per cycle based on the discharge power and cycle duration, and recursively obtain the predicted remaining capacity ratio for each future cycle (second energy supply status evolution result). For each task to be executed, its energy quota and remaining execution amount in the current cycle are input into the task execution progress prediction module and compared with the minimum energy required for the task to complete standard execution progress: if the actual energy obtained is not lower than the required energy, the remaining execution amount is reduced according to the standard progress amount; if the energy obtained is lower than the required energy but greater than zero, the remaining execution amount is reduced according to the proportion of energy obtained to the required energy; if no energy is obtained, the remaining execution amount remains unchanged. By recursively applying this rule, the predicted remaining execution amount of each task in future cycles can be obtained (the result of the third energy supply state evolution).
[0058] This application refines the candidate energy allocation schemes to the planned allocated power of hydrogen fuel cells, the planned discharge power of lithium batteries, and the task energy quota for each task. Simultaneously, it breaks down the energy supply state prediction model into three relatively independent parts: a hydrogen energy output prediction module, a lithium battery power prediction module, and a task execution progress prediction module. By obtaining the actual output power, remaining power ratio, and remaining execution amount of each task in the current control cycle, and combining this with the maximum variable power of the hydrogen fuel cell within a single control cycle, the output power, remaining power ratio, and remaining execution amount for multiple future cycles are recursively predicted. This modular prediction method allows the changing trends of each type of state variable to be tracked and evaluated independently, avoiding the computational burden caused by complex coupled models. Furthermore, the recursive prediction can anticipate potential problems such as power mismatch, insufficient power, or task delays that may arise from different candidate allocation schemes in the future, thus improving the accuracy of decision-making.
[0059] S104: Based on the energy supply state evolution results, an initial energy supply control scheme is obtained by scoring and selecting the best from each of the candidate energy supply allocation schemes, and the initial energy supply control scheme is modified based on preset operating constraints to determine the final energy supply control scheme.
[0060] Furthermore, in some embodiments of this application, the step of obtaining an initial energy supply control scheme by scoring and selecting the best from each of the candidate energy supply allocation schemes based on the energy supply state evolution results includes: For each of the candidate power allocation schemes, a first score for the candidate power allocation scheme is determined based on the difference between the planned allocated power and the corresponding predicted output power of the candidate power allocation scheme. For each of the candidate energy allocation schemes, a second score for the candidate energy allocation scheme is determined based on the difference between the predicted remaining power ratio corresponding to the candidate energy allocation scheme and the preset minimum allowable remaining power. For each of the candidate power allocation schemes, a third score for the candidate power allocation scheme is determined based on the difference between the predicted remaining execution amount and the preset remaining execution amount threshold for each task to be executed in the candidate power allocation scheme. Based on the first score, the second score, and the third score, the candidate energy allocation schemes are scored and selected for the best one to obtain the initial energy control scheme.
[0061] Preferably, in some embodiments of this application, the scoring and selection process is as follows: For each candidate energy allocation scheme, firstly, the planned allocation power of the hydrogen fuel cell is compared with the predicted output power for each future cycle obtained by the hydrogen energy output prediction module. The smaller the difference between the two, the better the fuel cell output can follow the allocation plan, and the less the power regulation pressure. Based on this, a first score is determined. Then, the predicted remaining power ratio for each future cycle obtained by the lithium battery power prediction module is compared with a preset minimum allowable remaining power (i.e., the lower limit that the lithium battery can retain during operation to avoid over-discharge and retain rapid compensation capability). The higher the predicted power is above this lower limit, the greater the power safety margin. Based on this, a second score is determined. Next, for each task to be executed, its remaining execution amount at the end of the prediction is compared with a preset remaining execution amount threshold. The smaller the remaining execution amount or the closer it is to the threshold, the greater the possibility of the task being completed on schedule. The evaluation results of all tasks are combined to obtain a third score. Finally, the first, second, and third scores are fused through preset weighting coefficients to evaluate each candidate scheme. The scheme with the highest score is selected as the initial energy control scheme. In practical applications, gradient boosting tree models can also be used to replace the above manual scoring rules. The model will automatically learn the weight relationship between each factor based on the training samples, but in principle, it is still based on the three dimensions of the deviation between the predicted output power and the plan, the power safety margin, and the degree of task progress guarantee.
[0062] The evaluation process involves three main components: First, the difference between the planned power allocation and the predicted output power. A smaller difference indicates that the hydrogen fuel cell output follows the allocation plan more closely, reducing power regulation pressure. Second, the ratio of predicted remaining charge to preset minimum allowable remaining charge. A larger difference indicates a greater safety margin for the lithium battery, reducing the risk of running out of power. Third, the ratio of predicted remaining execution for each task to preset thresholds. A smaller difference indicates a higher probability of timely task completion. By combining these three evaluations to select the best candidate scheme, the excessive pursuit of a single indicator can be avoided at the expense of other aspects of the control scheme, resulting in an initial power supply control scheme with better overall performance.
[0063] Furthermore, in some embodiments of this application, the step of modifying the initial power supply control scheme based on preset operating constraints to determine the final power supply control scheme includes: When the initial power supply control scheme violates the preset operating constraints, the task energy quota of each task to be executed in the initial power supply control scheme is reduced according to the urgency index, and the target task energy quota of each task to be executed is determined. Based on the energy quota of each target task, the planned allocated power and the planned discharge power in the initial power supply control scheme are adjusted to obtain the target planned allocated power and the target planned discharge power, respectively. The final power supply control scheme is determined based on the target planned power allocation, the target planned discharge power, and the target task energy quota for each of the tasks to be executed.
[0064] Preferably, in some embodiments of this application, when the initial energy supply control scheme violates preset operating constraints, the energy quota of each task to be executed is first reduced according to the urgency index: tasks with higher urgency retain their quotas, while tasks with lower urgency are reduced first, thereby determining the adjusted target energy quota for each task. Based on this, the planned power allocation for the hydrogen fuel cell and the planned discharge power for the lithium battery in the initial scheme are readjusted: if the original planned power allocation exceeds the current allowable output limit of the hydrogen fuel cell or the maximum variable power per cycle, it is reduced to the allowable range; if the remaining capacity of the lithium battery after discharge according to the original plan will be lower than the minimum allowable remaining capacity, the lithium battery discharge power is reduced accordingly, and the energy quota for low-urgency tasks is further reduced. Simultaneously, it is necessary to ensure that the sum of the target energy quotas for all tasks does not exceed the total allocable energy for the current control cycle; if it does, reductions continue in ascending order of urgency. After the above adjustments are completed, the final energy supply control scheme is formed by combining the adjusted target planned power allocation, target planned discharge power, and target energy quota for each task.
[0065] When revising the initial energy supply control scheme, this application systematically reduces the energy quota for tasks to be performed based on the urgency index. After determining the target energy quota for each task after reduction, the planned power allocation of the hydrogen fuel cell and the planned discharge power of the lithium battery are adjusted in reverse according to the new energy allocation, so that they fall back within the allowable operating range. This results in a final energy supply control scheme that satisfies the hard constraints while protecting the energy supply for urgent tasks as much as possible. Compared with directly reverting to a conservative scheme or reducing the overall power, this scheme achieves a better trade-off between safety and task completion rate.
[0066] S105: Control the unmanned system to execute each of the tasks to be executed through the final power supply control scheme.
[0067] Preferably, in some embodiments of this application, after the unmanned system is controlled to execute each of the tasks to be executed through the final power supply control scheme, the method further includes: collecting execution feedback data of the current control cycle, specifically including energy-side feedback (actual output power of the hydrogen fuel cell, actual discharge status of the lithium battery, and the actual remaining charge ratio of the lithium battery at the end of the control cycle) and task-side feedback (actual energy consumption, actual execution progress, and remaining execution amount of each task at the end of the cycle). The actual execution results are compared with the previously predicted power supply state evolution results. If the deviation between the two exceeds a preset threshold, the relevant parameters in the power supply state prediction model are corrected. For example, the output change response parameter is adjusted according to the difference between the actual output and the predicted output of the hydrogen fuel cell, the charge reduction law parameter is adjusted according to the difference between the actual remaining charge and the predicted remaining charge of the lithium battery, and the energy consumption parameter corresponding to the unit execution amount is adjusted according to the difference between the actual execution progress and the predicted progress of each task. Simultaneously, the system also tracks task completion status (e.g., the proportion of critical tasks completed on time, and the proportion of tasks completed late) and energy supply execution status (e.g., the number of cycles in which the remaining lithium battery charge percentage is within the warning range) across multiple consecutive control cycles. Based on this, arbitration parameters in subsequent control cycles are adjusted, such as task urgency weight parameters, minimum execution energy correction parameters for critical tasks, priority allocation parameters for hydrogen fuel cells, and minimum allowable remaining lithium battery charge correction parameters. Through this closed-loop correction, the energy supply control scheme for subsequent control cycles more closely reflects the actual operating conditions.
[0068] One embodiment of this application provides a hydrogen-powered unmanned system power supply control system, which is used to implement the hydrogen-powered unmanned system power supply control method provided in any of the above-mentioned method embodiments of this application.
[0069] It is understood that the above system embodiments correspond to the method embodiments of this application, and can realize the hydrogen-powered unmanned system power supply control method provided by any of the above method embodiments of this application.
[0070] It should be noted that the system embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0071] Based on the above embodiments of the hydrogen-powered unmanned system power supply control method, another embodiment of this application provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the hydrogen-powered unmanned system power supply control method of any embodiment of this application.
[0072] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete this application. The one or more module units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.
[0073] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.
[0074] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.
[0075] Based on the above-described method embodiments, another embodiment of this application provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the hydrogen-powered unmanned system power supply control method described in any of the above-described method embodiments of this application.
[0076] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
Claims
1. A hydrogen-based power supply control method for an unmanned system, characterized by, include: The system acquires the allocable energy of the unmanned system in the current control cycle and the task information of each task to be executed, and determines the urgency index of each task to be executed based on the task information. The feasibility classification results for completing each of the tasks to be executed are determined based on the task information, the allocable energy, and the urgency index. Based on the completed feasibility classification results, multiple candidate energy supply allocation schemes are generated, and based on the pre-built energy supply status prediction model, the energy supply status evolution result of each candidate energy supply allocation scheme is predicted. Based on the energy supply state evolution results, an initial energy supply control scheme is obtained by scoring and selecting the best from each of the candidate energy supply allocation schemes, and the initial energy supply control scheme is modified based on preset operating constraints to determine the final energy supply control scheme. The final power supply control scheme controls the unmanned system to execute each of the tasks to be performed.
2. The hydrogen-based power supply control method for unmanned systems of claim 1, wherein, The candidate energy allocation schemes include: the planned allocated power of hydrogen fuel cells, the planned discharge power of lithium batteries, and the task energy quota for each of the tasks to be executed; the energy supply status prediction model includes: a hydrogen energy output prediction module, a lithium battery power prediction module, and a task execution progress prediction module; the prediction of the energy supply status evolution result for each candidate energy allocation scheme based on the pre-constructed energy supply status prediction model includes: Obtain the actual output power of the hydrogen fuel cell, the remaining charge percentage of the lithium battery, and the current remaining execution amount of each task to be executed in the current control cycle; The maximum variable power of a single control cycle of the hydrogen fuel cell is obtained. The maximum variable power, the actual output power, and the planned allocated power are input into the hydrogen energy output prediction module to recursively predict the predicted output power of several future control cycles, thereby obtaining the first energy supply state evolution result of the hydrogen fuel cell. The planned discharge power and the remaining power ratio are input into the lithium battery power prediction module to recursively predict the predicted remaining power ratio for several future control cycles, thereby obtaining the second power supply state evolution result of the lithium battery. For each task to be executed, the task energy quota and the current remaining execution amount of the task to be executed are input into the task execution progress prediction module to recursively predict the predicted remaining execution amount for several future control cycles, thereby obtaining the third energy supply state evolution result of the task to be executed.
3. The hydrogen-based power supply control method for unmanned systems of claim 2, wherein, The step of obtaining an initial energy supply control scheme by scoring and selecting the best from each of the candidate energy supply allocation schemes based on the energy supply state evolution results includes: For each of the candidate power allocation schemes, a first score for the candidate power allocation scheme is determined based on the difference between the planned allocated power and the corresponding predicted output power of the candidate power allocation scheme. For each of the candidate energy allocation schemes, a second score for the candidate energy allocation scheme is determined based on the difference between the predicted remaining power ratio corresponding to the candidate energy allocation scheme and the preset minimum allowable remaining power. For each of the candidate power allocation schemes, a third score for the candidate power allocation scheme is determined based on the difference between the predicted remaining execution amount and the preset remaining execution amount threshold for each task to be executed in the candidate power allocation scheme. Based on the first score, the second score, and the third score, the candidate energy allocation schemes are scored and selected for the best one to obtain the initial energy control scheme.
4. The hydrogen-based power supply control method for unmanned systems of claim 2, wherein, The completion feasibility classification results include: a first category of tasks that meet the conditions for timely completion and a second category of tasks that do not meet the conditions for timely completion; the task information includes: the energy requirement of each of the tasks to be executed; the generation of multiple candidate energy allocation schemes based on the completion feasibility classification results includes: Obtain the preset rated output power range of the hydrogen fuel cell, and determine the feasible power traversal range of the hydrogen fuel cell based on the maximum variable power and the preset rated output power range. Multiple different power values are traversed within the power traversal interval, which are used as the planned power allocation of the hydrogen fuel cell in each of the candidate energy supply allocation schemes. Based on the allocable energy and the planned allocated power in the current control cycle, the planned discharge power of the lithium battery in the corresponding candidate power allocation scheme is determined; Allocate the corresponding energy requirement of each of the first type of tasks as the corresponding task energy quota in each of the candidate energy allocation schemes; Based on the remaining allocable energy and the urgency index corresponding to each of the second type of tasks, the task energy quota corresponding to each of the candidate energy allocation schemes for each of the second type of tasks is determined.
5. The hydrogen-based power supply control method for unmanned systems of claim 4, wherein, The task information includes: time limit parameters and estimated completion time corresponding to the task to be executed; the determination of the completion feasibility classification result of each task to be executed based on the task information, the allocable energy, and the urgency index includes: Based on the time limit parameters, the estimated completion time, the task energy requirements, and the available energy, determine the time gap and energy gap for each of the tasks to be executed. When the time gap is less than or equal to a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the first type of task that has the conditions for timely completion. When the time gap is greater than a preset time threshold and the energy gap is less than or equal to a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient time. When the time gap is less than or equal to a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be the second type of task with insufficient energy. When the time gap is greater than a preset time threshold and the energy gap is greater than a preset energy threshold, the corresponding task to be executed is determined to be a second type of task that has both insufficient time and insufficient energy.
6. The hydrogen-based power supply control method for unmanned systems of claim 5, wherein, The step of determining the urgency index for each of the tasks to be executed based on the task information includes: Calculate the remaining time for each task to be executed based on the time limit parameters of each task to be executed and the current time of the current control cycle; When the remaining time is less than or equal to the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the first urgency value; When the remaining time is greater than the estimated completion time, the urgency index of the corresponding task to be executed is determined to be the second urgency value; wherein, the urgency of the first urgency value is higher than that of the second urgency value.
7. The hydrogen-based power supply control method for unmanned systems of claim 2, wherein, The step of modifying the initial power supply control scheme based on preset operating constraints to determine the final power supply control scheme includes: When the initial power supply control scheme violates the preset operating constraints, the task energy quota of each task to be executed in the initial power supply control scheme is reduced according to the urgency index, and the target task energy quota of each task to be executed is determined. Based on the energy quota of each target task, the planned allocated power and the planned discharge power in the initial power supply control scheme are adjusted to obtain the target planned allocated power and the target planned discharge power, respectively. The final power supply control scheme is determined based on the target planned power allocation, the target planned discharge power, and the target task energy quota for each of the tasks to be executed.
8. A hydrogen-based power supply control system for an unmanned system, comprising: The hydrogen-powered unmanned system power supply control system is used to implement the hydrogen-powered unmanned system power supply control method as described in any one of claims 1 to 7.
9. A terminal device, comprising: The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the power supply control method for a hydrogen-powered unmanned system as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the power supply control method for a hydrogen-powered unmanned system as described in any one of claims 1 to 7.
