Nitrous oxide propulsion self-pressurization storage tank temperature closed-loop control method and system

Abstract

The invention provides a temperature closed-loop control method and system for a nitrous oxide propulsion self-pressurization storage box, and relates to the technical field of spacecraft propulsion system.The method comprises the steps that nitrous oxide concentration, pressure, propellant temperature distribution, spacecraft task stages and propellant surplus data in the storage box are obtained, and the nitrous oxide concentration, pressure, propellant temperature distribution, spacecraft task stages and propellant surplus data are obtained; calculating the power adjustment amount of each partition of the external heating layer based on the concentration and temperature data, and generating a partition control instruction; the partition power output is adjusted by executing the instruction, and a temperature field matched with a preset condition is formed in the storage box; then cooperatively optimizing the temperature field and the pressure data to generate an optimized power distribution proportion; then, the task stage and the propellant remaining amount are processed in a rolling mode, and a temperature control parameter track is generated; and finally, fusing the power distribution result and the parameter track to form a comprehensive control strategy containing an updating instruction. According to the method, the efficiency of the nitrous oxide self-pressurization system in the propellant phase change dynamic process in an on-orbit task is improved.

Description

Technical Field

[0001] This application relates to the field of spacecraft propulsion system control technology, and in particular to a closed-loop control method and system for the temperature of a nitrous oxide propulsion self-pressurization tank. Background Technology

[0002] During long-term on-orbit space missions, nitrous oxide self-pressurization systems are typically employed, and their pressure stability depends on precise control of the propellant's gas-liquid phase change process within the tank. The gas-liquid phase change process of the propellant in different regions of the tank is dynamic and spatially inhomogeneous. This characteristic leads to fluctuations in the evaporation rate, which in turn affects the tank's pressure stability. Therefore, a temperature control method capable of adaptively adjusting to accommodate the dynamic phase change process of the propellant is needed.

[0003] Currently, a typical solution to this requirement is to achieve temperature management by combining fixed-zone heating control with pressure feedback regulation. Specifically, this solution sets up several independent temperature zones in the tank heating layer, performs constant power heating on each zone based on a preset temperature curve, and adjusts the heating power proportionally according to pressure sensor data to maintain the basic stability of the tank pressure.

[0004] However, during long-term operation, this scheme suffers from a mismatch between the dynamic process of propellant vapor-liquid phase change within the tank and the fixed-zone heating mode, leading to a mismatch between the local temperature field and the propellant evaporation / condensation requirements. Furthermore, feedback regulation based on a single pressure parameter often results in control lag when dealing with the dynamic characteristics of a multivariate coupled system, affecting the optimization of propellant utilization efficiency. Summary of the Invention

[0005] This application provides a closed-loop temperature control method and system for nitrous oxide propulsion self-pressurization tanks to solve the problem of low efficiency of nitrous oxide self-pressurization systems in on-orbit missions during propellant phase change dynamics.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a closed-loop temperature control method for a nitrous oxide-propelled self-pressurized tank, comprising:

[0007] Acquire data on nitrous oxide concentration, pressure, propellant temperature distribution, spacecraft mission phase information, and remaining propellant quantity in the self-pressurized tank;

[0008] Based on the nitrous oxide concentration data and the temperature distribution data, the required power adjustment amount for each zone of the external heating layer of the self-pressurized tank is calculated, and a zone power control command corresponding to the power adjustment amount is generated.

[0009] Based on the partition power control command, the power output of each partition is adjusted to form a temperature field distribution in the tank that matches the preset conditions.

[0010] The temperature field distribution and the pressure data are jointly optimized using a reinforcement learning algorithm to generate a proportional optimization result.

[0011] Using a dynamic programming algorithm, the task phase information and the remaining propellant data are subjected to rolling optimization processing to generate a target temperature control parameter sequence;

[0012] The proportional optimization results and the target temperature control parameter sequence are combined to form a comprehensive control strategy. The comprehensive control strategy includes an updated partition power control command, which is used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

[0013] Optionally, the step of using a reinforcement learning algorithm to collaboratively optimize the temperature field distribution and the pressure data to generate a proportional optimization result includes:

[0014] Based on the temperature field distribution and the pressure data, a preliminary scheme containing the power allocation ratio of each partition is generated through a policy network using a reinforcement learning algorithm.

[0015] The pressure stability gains of the preliminary scheme in terms of pressure stability and temperature field uniformity in the self-pressurized tank are evaluated through the value network of the reinforcement learning algorithm.

[0016] The policy optimization module of the reinforcement learning algorithm updates the preliminary scheme through multiple rounds of iteration based on the pressure-stabilized reward value. In each round of iteration, the exploration mechanism of the reinforcement learning algorithm generates a proportional optimization result containing the power allocation ratio of each partition.

[0017] Optionally, the policy optimization module using the reinforcement learning algorithm updates the initial scheme through multiple iterations based on the pressure-stabilized reward value. In each iteration, the algorithm's exploration mechanism generates a proportional optimization result containing the power allocation ratio of each partition, including:

[0018] Based on the sign and magnitude of the pressure stability benefit value, the adjustment direction and adjustment range of the preliminary plan are determined;

[0019] Based on the adjustment direction and the adjustment range, the preliminary scheme is optimized to generate an optimized scheme;

[0020] Based on the optimization scheme, the partition power perturbation components corresponding to each partition that conform to a normal distribution are generated through an exploration mechanism;

[0021] The partition power disturbance component is superimposed onto the corresponding partition power allocation ratio in the optimization scheme to form a power allocation scheme.

[0022] Through the value network, the pressure stability benefit value of the power allocation scheme is compared with the pressure stability benefit value of the optimization scheme. When the comparison result shows that the pressure stability benefit value of the power allocation scheme is greater than the pressure stability benefit value of the optimization scheme, the power allocation scheme is used as the initial scheme for a new round of iteration.

[0023] Repeat the above process until the preset number of iterations is reached, and output the final optimized partition power allocation ratio scheme, which includes the ratio optimization results of the power allocation ratio of each partition.

[0024] Optionally, adjusting the power output of each zone based on the zoned power control command to form a temperature field distribution within the tank that matches preset conditions includes:

[0025] The partition power control command is converted into a power drive signal corresponding to each partition.

[0026] The power output of each zone heating film is controlled according to the power drive signal, and the corresponding heating effect is produced.

[0027] Real-time propellant temperature data for each region within the tank under the heating effect are collected using a temperature detection device.

[0028] The real-time temperature data is compared and analyzed with the expected temperature distribution determined based on the gas-liquid phase change requirements;

[0029] Based on the comparative analysis results, the power drive signal is dynamically corrected, and the heating film of each zone is controlled by the corrected power drive signal to form a temperature field distribution corresponding to the preset conditions.

[0030] Optionally, the step of using a dynamic programming algorithm to perform rolling optimization processing on the mission phase information and the remaining propellant data to generate a target temperature control parameter sequence includes:

[0031] Based on the task stage information, determine the current task stage and the remaining unexecuted task stages;

[0032] Based on the remaining propellant data, predict the trend of propellant consumption rate changes in each mission phase;

[0033] Based on the propellant consumption rate change trend value, the state transition equation of the dynamic programming algorithm is used to calculate multiple candidate values ​​of temperature control parameters for each mission stage.

[0034] With the goal of maximizing propellant utilization efficiency, target values ​​for temperature control parameters corresponding to each mission stage are selected from multiple candidate values ​​of temperature control parameters corresponding to each mission stage.

[0035] The target values ​​of the temperature control parameters corresponding to each task stage are sorted according to the order of the task stages to form an initial temperature control parameter sequence.

[0036] The initial temperature control parameter sequence is periodically corrected through a rolling optimization mechanism to generate the target temperature control parameter sequence.

[0037] Optionally, the fusion of the proportional optimization result and the target temperature control parameter sequence to form a comprehensive control strategy includes:

[0038] The power allocation ratio of each partition in the ratio optimization result is time-sequentially aligned with the target value of the temperature control parameter at the corresponding time in the target temperature control parameter sequence.

[0039] By using a weighted fusion algorithm, the power allocation ratio after time alignment and the target value of the temperature control parameter after time alignment are combined and calculated to obtain the preliminary control strategy after weighted fusion.

[0040] The initial control strategy after weighted fusion is regionally modified to obtain the regionally modified control strategy;

[0041] Verify the coordination between the power allocation ratio of each zone and the target value of the temperature control parameter in the regionally modified control strategy. When the verification is successful, output the modified control strategy as the verified control strategy.

[0042] When a conflict is detected between the power allocation ratio in the verified control strategy and the target value of the temperature control parameter, an arbitration mechanism is initiated. The verified control strategy is readjusted with tank pressure stability as the priority indicator to obtain the arbitration-adjusted control strategy, which is then output as the comprehensive control strategy.

[0043] Optionally, calculating the required power adjustment for each zone of the external heating layer of the self-pressurizing tank based on the nitrous oxide concentration data and the temperature distribution data includes:

[0044] Differential analysis was performed on the nitrous oxide concentration data and the temperature distribution data to extract characteristic parameters that reflect the evaporation and condensation state of the propellant.

[0045] The characteristic parameters are converted into the temperature field distribution in the external heating layer of the self-pressurized tank.

[0046] Based on the temperature field distribution, the heating requirements of each zone of the external heating layer of the self-pressurized tank are determined;

[0047] Based on the heating requirements, calculate the power adjustment amount required for each zone.

[0048] Secondly, this application provides a closed-loop temperature control system for a nitrous oxide-propelled self-pressurized tank, comprising:

[0049] The acquisition module is used to acquire data on nitrous oxide concentration, pressure, temperature distribution of propellant in the self-pressurized tank, mission phase information of the spacecraft, and remaining propellant quantity.

[0050] The calculation module is used to calculate the power adjustment required for each zone of the external heating layer of the self-pressurized tank based on the nitrous oxide concentration data and the temperature distribution data, and to generate a zone power control command corresponding to the power adjustment amount.

[0051] The adjustment module is used to adjust the power output of each zone based on the zone power control command, so as to form a temperature field distribution in the tank that matches the preset conditions.

[0052] The optimization module is used to perform collaborative optimization processing on the temperature field distribution and the pressure data using reinforcement learning algorithms to generate proportional optimization results;

[0053] The generation module is used to perform rolling optimization processing on the task stage information and the remaining propellant data using a dynamic programming algorithm to generate a target temperature control parameter sequence.

[0054] The fusion module is used to fuse the proportional optimization results and the target temperature control parameter sequence to form a comprehensive control strategy. The comprehensive control strategy includes updated partition power control instructions, which are used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

[0055] Thirdly, this application provides an electronic device, comprising:

[0056] Memory, used to store computer programs;

[0057] A processor is configured to execute the computer program to implement the steps of the closed-loop temperature control method for a nitrous oxide propulsion self-pressurized tank as described in the first aspect above.

[0058] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the nitrous oxide propulsion self-pressurizing tank temperature closed-loop control method described in the first aspect above.

[0059] The technical solution provided in this application has the following beneficial effects:

[0060] First, this application establishes a comprehensive perception capability of the control system for environmental conditions through synchronous acquisition of multiple parameters. Then, it achieves precise quantification of heating requirements for each zone based on correlation analysis of concentration and temperature data. Next, it forms a temperature field matching the activity distribution within the tank through zone power adjustment. Then, it uses reinforcement learning algorithms to achieve dynamic balance between temperature field and pressure parameters, thereby improving the system's adaptability. Simultaneously, it uses dynamic programming algorithms to achieve forward-looking planning of temperature control parameters throughout the entire task cycle. Finally, it integrates the optimization results of multiple algorithms to form a comprehensive control strategy adapted to complex operating conditions.

[0061] Furthermore, this application generates a preliminary power allocation scheme through a policy network, evaluates its long-term benefits in terms of pressure stability and temperature field uniformity through a value network, and then performs multiple rounds of iterative updates through a policy optimization module combined with an exploration mechanism, ultimately outputting an optimized power allocation ratio. Therefore, this process achieves autonomous optimization of power allocation under multiple constraints, forming an optimal power allocation strategy that adapts to the real-time state of the system by dynamically balancing the requirements of pressure stability and temperature field uniformity.

[0062] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 A flowchart illustrating a closed-loop temperature control method for a nitrous oxide propulsion self-pressurization tank, provided in an embodiment of this application;

[0065] Figure 2 A schematic diagram illustrating a specific implementation of a closed-loop temperature control method for a nitrous oxide propulsion self-pressurized tank, provided in this application embodiment;

[0066] Figure 3 This is a schematic diagram of the temperature closed-loop control system for a nitrous oxide propulsion self-pressurization tank provided in an embodiment of this application. Detailed Implementation

[0067] The existing control scheme, which uses fixed-zone heating combined with pressure feedback, faces two prominent contradictions during long-term on-orbit operation: First, there is a mismatch between the dynamic process of propellant gas-liquid phase change in the tank and the fixed-zone heating mode, resulting in a mismatch between the local temperature field and the propellant evaporation / condensation requirements; Second, there is a response lag between the single pressure parameter feedback and the dynamic characteristics of the multivariable coupled system, which affects the optimization of propellant utilization efficiency. These contradictions stem from the insufficient adaptability of the static control strategy to the time-varying and spatially varying characteristics of the system.

[0068] This application proposes a closed-loop temperature control method for a nitrous oxide propulsion self-pressurized tank. This method achieves dynamic matching between the temperature field and the propellant phase change requirements through multi-parameter fusion sensing and dual-algorithm collaborative decision-making. It not only solves the local temperature control mismatch problem in the fixed-zone heating mode, but also eliminates the hysteresis effect of single-parameter feedback through multi-variable forward control, thereby improving the pressure stability and propellant utilization efficiency of the propulsion system in long-term on-orbit missions.

[0069] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0070] The core of this application is to provide a closed-loop temperature control method for a nitrous oxide-propelled self-pressurized tank, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0071] Step 101: Obtain nitrous oxide concentration data, pressure data, propellant temperature distribution data, spacecraft mission phase information, and propellant remaining amount data in the self-pressurized tank.

[0072] In step 101, the nitrous oxide concentration data refers to the proportion of nitrous oxide gas at the tank outlet in the mixed gas as measured by a gas sensor, and the pressure data refers to the gas pressure value inside the tank collected by a pressure sensor.

[0073] Propellant temperature distribution data refers to the set of propellant temperature values ​​collected at different spatial locations by an array of temperature sensors arranged on or inside the propellant tank. The spacecraft is the platform supporting the self-pressurized propellant tank, which, as a component of the spacecraft's propulsion system, provides thrust and has a self-pressurization function, which is achieved by heating the propellant inside the tank through an external heating layer. Spacecraft mission phase information refers to the time series of different operating modes, such as the startup phase, steady-state operation phase, and shutdown phase, divided according to the flight plan.

[0074] Nitrous oxide is used as a propellant. Nitrous oxide propellant generates thrust after being heated and vaporized. The remaining propellant data refers to the mass or volume of the remaining propellant in the tank, as measured by a level gauge or mass flow meter.

[0075] In this embodiment, firstly, nitrous oxide concentration data is continuously collected by a gas concentration sensor installed at the tank outlet, and propellant temperature data for each zone is collected by a network of temperature sensors distributed on the inner wall or inside the tank. Then, pressure changes inside the tank are monitored in real time by a pressure sensor, and the current mission phase is identified by the spacecraft control system. Meanwhile, remaining propellant data is periodically reported by the propellant metering device. Finally, these multi-source data are synchronously transmitted to the central processing unit for subsequent analysis.

[0076] Step 102: Based on the nitrous oxide concentration data and the temperature distribution data, calculate the power adjustment required for each zone of the external heating layer of the self-pressurized tank, and generate a zone power control command corresponding to the power adjustment.

[0077] In step 102, the structure of the self-pressurized tank includes a tank cavity for storing propellant and an external heating layer disposed on the tank wall. The power adjustment amount refers to the power value that needs to be increased or decreased for each heating zone calculated based on concentration and temperature data. The zone power control command refers to a control command that includes the specific power output value of each zone.

[0078] In this embodiment, the concentration data and temperature distribution data are first differentially processed to obtain the concentration change rate and temperature gradient data. Then, the evaporation state and phase change requirements of the propellant are calculated based on these characteristic parameters. Based on this requirement distribution, the power compensation value required for each heating zone is determined. Finally, control instructions containing the specific power output values ​​of each zone are generated.

[0079] For example, during the operation of a certain type of satellite propulsion system, the gas sensor detected that the nitrous oxide concentration at the tank outlet was 85%, the pressure sensor showed that the tank pressure was 2.5 MPa, and the propellant temperatures measured at eight temperature measuring points inside the tank were 245℃, 251℃, 238℃, 262℃, 241℃, 255℃, 247℃, and 259℃, respectively. The satellite control system indicated that it was currently in a steady-state operation phase, and the propellant metering device showed that the remaining propellant was 12.3 kg. These data were packaged and sent to the control computer.

[0080] The computer then processes the received concentration and temperature distribution data. During the processing, it can be found that the temperature in the third zone is too low and the concentration change rate is abnormal. It is also calculated that the power needs to be increased by 5 watts, 3 watts, 12 watts, -2 watts, 8 watts, -1 watt, 4 watts, and -3 watts respectively for the first to eighth zones. Based on this, a zone power control instruction containing these power adjustment values ​​is generated.

[0081] Step 103: Based on the partition power control command, adjust the power output of each partition to form a temperature field distribution in the tank that matches the preset conditions.

[0082] In step 103, the preset conditions can refer to the propellant gas-liquid phase change requirements, and the temperature field distribution refers to the temperature spatial distribution pattern formed in the propellant in the tank after zonal heating, which meets the gasification and pressurization requirements.

[0083] In this embodiment, firstly, the power control command for each zone is converted into specific voltage and current signals by the power drive module to control the operation of the heating film in each zone; simultaneously, the temperature sensor monitors the propellant temperature changes in each area of ​​the tank in real time and compares the measured temperature with the expected temperature distribution; finally, the power output is dynamically adjusted according to the comparison results until a temperature field distribution that meets the requirements is formed.

[0084] For example, the power drive module outputs the corresponding power to the eight heating zones according to the received zone power control command. After three control cycles, the propellant temperature in the eight zones of the tank reaches 252℃, 254℃, 250℃, 260℃, 249℃, 254℃, 251℃ and 256℃ respectively, forming a temperature field that matches the propellant vaporization requirements.

[0085] Step 104: Use reinforcement learning algorithm to perform collaborative optimization processing on the temperature field distribution and the pressure data to generate proportional optimization results.

[0086] In step 104, the proportional optimization result refers to the optimal proportional relationship of power allocation in each partition obtained by the optimization algorithm.

[0087] In this embodiment, based on real-time temperature field distribution data and pressure data, a preliminary power allocation scheme is generated using a policy network in a reinforcement learning algorithm. Then, a value network is used to evaluate the long-term benefits of the above scheme in terms of pressure stability and temperature uniformity. After that, a policy optimization module is used to perform multiple rounds of iterative updates based on the evaluation results, and finally outputs the optimal power allocation ratio.

[0088] For example, the reinforcement learning algorithm performed 15 rounds of iterative calculations based on the current temperature field distribution and 2.5 MPa pressure data, and finally output the power allocation ratios of each partition as 0.18, 0.16, 0.22, 0.08, 0.14, 0.06, 0.12, and 0.04. This result effectively balances the requirements of pressure stability and uniform propellant evaporation.

[0089] Step 105: Using a dynamic programming algorithm, perform rolling optimization processing on the mission stage information and the remaining propellant data to generate a target temperature control parameter sequence.

[0090] In step 105, the target temperature control parameter sequence refers to the optimal path for the temperature control parameters to change over time throughout the entire task cycle. This application embodiment does not specifically limit the specific expression of the dynamic programming algorithm, but it can be set according to the actual situation.

[0091] For example, based on the remaining propellant amount of 12.3 kg and mission phase information, the dynamic programming algorithm calculates the temperature control parameter sequence for each 5-hour period during the subsequent 50-hour mission, which are 255℃, 253℃, 251℃, 249℃, 247℃, 245℃, 243℃, 241℃, 239℃, and 237℃, forming the target temperature control parameter sequence.

[0092] Step 106: Integrate the proportional optimization results and the target temperature control parameter sequence to form a comprehensive control strategy. The comprehensive control strategy includes an updated partition power control command, which is used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

[0093] In step 106, the integrated control strategy refers to the final control scheme formed by integrating power allocation optimization and parameter trajectory optimization.

[0094] In this embodiment, the proportional optimization result is time-aligned with the target temperature control parameter sequence, and then a preliminary control strategy is generated based on the aligned result using a weighted fusion algorithm. The preliminary control strategy is then regionally modified according to the temperature field distribution, and finally an executable comprehensive control strategy is output through coordination verification.

[0095] For example, the control computer integrates the proportional optimization results with the target temperature control parameter sequence to generate a comprehensive control strategy that includes updated partition power control instructions. This strategy instructs the first to eighth partitions to output 46 watts, 41 watts, 56 watts, 20 watts, 36 watts, 15 watts, 31 watts, and 10 watts of power respectively in the next control cycle, achieving a precise response to changes in the propellant phase change state.

[0096] This application embodiment establishes a dynamic temperature control mechanism that adapts to the dynamic process of propellant phase change by using multi-source data fusion sensing and dual-algorithm collaborative decision-making. This improves the pressure stability of the propulsion system and the efficiency of working fluid utilization during long-term on-orbit missions, thereby effectively solving the adaptability problem of fixed control strategies under time-varying conditions.

[0097] To address the issue of decreased temperature control accuracy caused by dynamic changes in the propellant phase transition process, in some embodiments, step 104 involves using a reinforcement learning algorithm to perform collaborative optimization processing on the temperature field distribution and the pressure data, generating a proportional optimization result, such as... Figure 2 As shown, it includes:

[0098] Step 201: Using a policy network based on a reinforcement learning algorithm, generate a preliminary scheme containing the power allocation ratio of each partition, based on the temperature field distribution and the pressure data.

[0099] In step 201, the policy network is a computational module in the reinforcement learning algorithm that generates decision schemes based on the environmental state. The preliminary scheme refers to the initial control strategy that includes the power allocation ratio of each heating zone.

[0100] Step 202: Evaluate the pressure stability gain of the preliminary scheme in terms of pressure stability and temperature field uniformity in the self-pressurized tank through the value network of the reinforcement learning algorithm.

[0101] In step 202, the value network is a computational module in the reinforcement learning algorithm that evaluates the long-term benefits of decision-making schemes. The pressure stability benefit value refers to the comprehensive score of the scheme in maintaining pressure stability and temperature field uniformity.

[0102] Step 203: Using the policy optimization module of the reinforcement learning algorithm, the preliminary scheme is iterated and updated multiple times based on the pressure stability reward value. In each iteration update process, the optimization results containing the power allocation ratio of each partition are generated through the exploration mechanism of the reinforcement learning algorithm.

[0103] In steps 201 to 203, the policy network, value network, and policy optimization module are all core computational structures that constitute the reinforcement learning algorithm. The policy network is responsible for generating the initial plan, the value network is responsible for evaluating the benefits of the plan, and the policy optimization module iteratively improves the plan output by the policy network based on the evaluation results of the value network.

[0104] The policy optimization module is the computational unit in the reinforcement learning algorithm that improves decision-making schemes. The exploration mechanism refers to the method of introducing random changes during the optimization process to discover better solutions. The exploration mechanism is a specific functional unit integrated into the policy optimization module. It introduces controllable random changes based on the initial scheme generated by the policy network, forms a feedback loop with the value network, and jointly assists the policy optimization module in completing the exploration and optimization process of the power allocation scheme. The proportional optimization result refers to the optimal power allocation scheme obtained after multiple iterations.

[0105] It should be noted that the embodiments of this application do not specifically limit the specific structure of the reinforcement learning algorithm, and the structure can be adjusted according to the actual situation.

[0106] In this embodiment, the strategy optimization module iterates and updates the initial scheme multiple times based on the pressure stability benefit value. In each iteration, a random perturbation of the power allocation ratio is introduced through an exploration mechanism. After multiple evaluations and improvements, the final ratio optimization result is output.

[0107] Here is a specific example:

[0108] During the operation of the satellite propulsion system, the policy network of the reinforcement learning algorithm generates a preliminary scheme [0.15, 0.14, 0.20, 0.10, 0.16, 0.08, 0.12, 0.05] containing the power allocation ratio of eight zones based on the current temperature field distribution data [252, 254, 250, 260, 249, 254, 251, 256] and 2.5 MPa pressure data;

[0109] Value networks are evaluated through a function. Calculate the stable return under pressure, where This indicates the stable return value under pressure. This represents a pressure stability index; for example, a value of 0.92 can be used. This indicates an index of temperature uniformity, such as a value of 0.85. These are the weighting coefficients, such as 0.6 and 0.4. Substituting them into the formula yields... ;

[0110] The strategy optimization module iterates and updates the initial scheme multiple times based on the benefit value. In each iteration, random perturbations are introduced through an exploration mechanism. After 15 iterations, the final proportional optimization result [0.18, 0.16, 0.22, 0.08, 0.14, 0.06, 0.12, 0.04] is generated. This result satisfies the convergence condition and is output to the control computer as the final optimization scheme.

[0111] In this embodiment, the power allocation scheme achieves the best balance between pressure stability and temperature uniformity through the collaborative optimization processing of reinforcement learning algorithm, which improves the adaptive control capability of the system under the dynamic process of propellant phase change and ensures the stability of tank pressure.

[0112] To further improve the optimization accuracy of the power allocation scheme, in some embodiments, step 203: the strategy optimization module using the reinforcement learning algorithm performs multiple rounds of iterative updates on the preliminary scheme based on the pressure stability reward value. During each round of iterative updates, the exploration mechanism of the reinforcement learning algorithm generates a proportional optimization result containing the power allocation ratio of each partition, including:

[0113] Step 301: Determine the adjustment direction and adjustment range of the preliminary scheme based on the positive and negative signs and the magnitude of the pressure stability benefit value.

[0114] In step 301, the adjustment direction refers to the direction of scheme modification determined based on the changing trend of the pressure stable benefit value, and the adjustment magnitude refers to the degree of scheme modification determined based on the magnitude of the pressure stable benefit value.

[0115] In this embodiment, firstly, the strategy optimization module analyzes the positive and negative characteristics of the pressure stability benefit value to determine whether the preliminary scheme, which includes the power allocation ratio of each partition, needs to strengthen or weaken the current trend; then, it determines the specific degree of adjustment based on the magnitude of the pressure stability benefit value, thereby providing clear guidance for subsequent optimization.

[0116] Step 302: Based on the adjustment direction and the adjustment range, optimize the preliminary scheme to generate an optimized scheme.

[0117] In step 302, the optimized scheme refers to the improved power allocation scheme obtained after preliminary adjustments.

[0118] In this embodiment, the strategy optimization module modifies the preliminary scheme according to the determined adjustment direction and adjustment range, and generates an optimized scheme that improves upon the original scheme framework.

[0119] Step 303: Based on the optimization scheme, generate partition power perturbation components that conform to a normal distribution for each partition through an exploration mechanism.

[0120] In step 303, the normal distribution refers to the symmetrical distribution pattern of the partition power perturbation components, which mathematically exhibits a high frequency of occurrence of the middle value and a gradually decreasing frequency of occurrence of the extreme values. This distribution pattern is automatically generated by the exploration mechanism of the reinforcement learning algorithm based on the expected deviation range between the current power allocation scheme and the optimal solution. The partition power perturbation components refer to the power adjustment amounts of each partition that conform to the probability distribution generated by the exploration mechanism.

[0121] In this embodiment of the application, the exploration mechanism generates a set of random perturbation components that conform to a normal distribution based on the optimization scheme. These components correspond to each heating zone and are used to explore the possibility of better power allocation.

[0122] Step 304: The partition power disturbance component is superimposed onto the corresponding partition power allocation ratio in the optimization scheme to form a power allocation scheme.

[0123] In step 304, the power allocation scheme refers to the new scheme formed by superimposing the disturbance component onto the optimization scheme.

[0124] Step 305: Through the value network, compare the pressure stability benefit value of the power allocation scheme with the pressure stability benefit value of the optimization scheme. When the comparison result shows that the pressure stability benefit value of the power allocation scheme is greater than the pressure stability benefit value of the optimization scheme, the power allocation scheme is taken as the preliminary scheme for a new round of iteration.

[0125] In this embodiment, when the pressure stability benefit value of the exploratory solution is greater than that of the original optimized solution, it indicates that the exploratory direction is effective, and the strategy optimization module will use the exploratory solution as the starting point for a new round of iterations to continue optimization.

[0126] Step 306: Repeat the above process until the preset number of iterations is reached, and output the final optimized partition power allocation ratio scheme, which includes the ratio optimization results of the power allocation ratio of each partition.

[0127] In step 306, the preset number of iterations refers to the upper limit of the number of algorithm loop calculations pre-set based on the dynamic characteristics of the propellant phase change process and the accuracy requirements of tank pressure control. This number is determined comprehensively through ground simulation tests and mission duration requirements.

[0128] In this embodiment, the strategy optimization module repeatedly performs the process of scheme adjustment, perturbation exploration, and benefit comparison until a preset number of iterations is reached, and outputs a fully optimized final power allocation scheme.

[0129] Here is a specific example:

[0130] During the satellite propulsion system optimization process, the strategy optimization module determined the adjustment direction based on the pressure stability benefit value of 0.892 to be increasing the power ratio of the third partition, with an adjustment range of 10% of the current value. The initial scheme [0.15, 0.14, 0.20, 0.10, 0.16, 0.08, 0.12, 0.05] was optimized to obtain the optimized scheme [0.15, 0.14, 0.22, 0.10, 0.16, 0.08, 0.12, 0.05].

[0131] The exploration mechanism generates partitioned power perturbation components [0.01, -0.02, 0.03, -0.01, 0.02, -0.03, 0.01, -0.02] that conform to a normal distribution. The perturbation components are then superimposed on the optimization scheme to form a power allocation scheme [0.16, 0.12, 0.25, 0.09, 0.18, 0.05, 0.13, 0.03].

[0132] Value network computing exploratory scheme stress-stabilized revenue value Among them, pressure stability index It can be 0.93, the temperature uniformity index The weighting coefficient can be 0.87. , Substituting into the formula, we get =0.6×0.93+0.4×0.87=0.906; Since 0.906 is greater than the original scheme's profit value of 0.892, the exploratory scheme is used as the initial scheme for the new round of iterations;

[0133] After repeating the above process 10 times, the final optimized partition power allocation ratio scheme [0.18, 0.16, 0.22, 0.08, 0.14, 0.06, 0.12, 0.04] is output as the ratio optimization result.

[0134] In the embodiments of this application, the power allocation scheme is continuously improved through a combination of multi-round iterative optimization and exploration mechanisms, which effectively enhances the overall performance of the scheme in terms of pressure stability and temperature uniformity.

[0135] To precisely control the temperature distribution of the propellant within the tank, in some embodiments, step 103: adjusting the power output of each zone based on the zoned power control command to create a temperature field distribution within the tank that matches preset conditions, includes:

[0136] Step 401: Convert the partition power control command into a power drive signal corresponding to each partition.

[0137] In step 401, the power drive signal refers to an electrical signal that can control the operation of the heating element, and its signal characteristics correspond to the required power level.

[0138] In this embodiment, the control unit converts the power value contained in the partition power control command into a power drive signal with specific voltage or current characteristics, and the power drive signal for each partition is generated independently.

[0139] Step 402: Control the power output of each zone heating film according to the power drive signal, and produce the corresponding heating effect.

[0140] In step 402, the heating effect refers to the effect of the heat energy generated by the heating element during operation being transferred to the propellant inside the tank through the tank wall, and its intensity is related to the power output.

[0141] In this embodiment of the application, a power drive signal is transmitted to each zone heating film, and the signal is used to control each zone heating film to generate corresponding thermal power output, wherein this thermal power can form a preliminary heating effect in the propellant when it is transmitted into the tank.

[0142] Step 403: Collect real-time propellant temperature data for each area inside the tank under the heating effect using a temperature detection device.

[0143] In step 403, each region refers to the monitoring unit inside the tank that is divided based on temperature detection requirements, and each zone refers to the heating control unit of the external heating layer of the self-pressurized tank that is divided based on independent temperature control requirements. The two correspond to each other in spatial location. Each zone affects the temperature state of the corresponding region through heating control, and each region provides feedback for the power control of each zone through temperature monitoring.

[0144] Real-time temperature data refers to the temperature values ​​of the propellant at various points inside the tank, measured in real time by temperature sensors.

[0145] In this embodiment, a network of temperature sensors distributed on the inner wall or inside the tank collects temperature data of each area in real time. These data reflect the actual temperature distribution under the current heating effect.

[0146] Step 404: Compare and analyze the real-time temperature data with the expected temperature distribution determined based on the gas-liquid phase change requirements.

[0147] In step 404, the expected temperature distribution refers to the ideal temperature range of each region determined according to preset conditions.

[0148] In this embodiment of the application, the collected real-time temperature data is compared and analyzed with the required temperature of the corresponding area in the expected temperature distribution in order to identify areas with large temperature deviations.

[0149] Step 405: Based on the comparative analysis results, the power drive signal is dynamically corrected, and the heating film of each zone is controlled by the corrected power drive signal to form a temperature field distribution corresponding to the preset conditions.

[0150] In this embodiment, firstly, the power drive signal in the area with deviation is adjusted based on the temperature comparison analysis results; then, the heating film is re-controlled by the adjusted power drive signal to gradually form a temperature field distribution that meets the requirements of gas-liquid phase change.

[0151] Here is a specific example:

[0152] During the temperature control process of the satellite propulsion system, the power drive module converts the instructions in the partition power control command that the first to eighth partitions need to increase the power by 5 watts, 3 watts, 12 watts, -2 watts, 8 watts, -1 watt, 4 watts, and -3 watts respectively into corresponding power drive signals, which control the heating film of each partition to output the adjusted power and produce the corresponding heating effect. The temperature detection device collects the real-time propellant temperature data of each area in the tank as 248 degrees Celsius, 252 degrees Celsius, 243 degrees Celsius, 261 degrees Celsius, 247 degrees Celsius, 253 degrees Celsius, 249 degrees Celsius, and 255 degrees Celsius.

[0153] These real-time temperature data are compared and analyzed with the expected temperature distributions determined based on the gas-liquid phase change requirements, namely temperatures of 252 degrees Celsius, 254 degrees Celsius, 250 degrees Celsius, 260 degrees Celsius, 249 degrees Celsius, 254 degrees Celsius, 251 degrees Celsius, and 256 degrees Celsius. The temperature differences for each region are calculated to be -4 degrees Celsius, -2 degrees Celsius, -7 degrees Celsius, 1 degree Celsius, -2 degrees Celsius, -1 degree Celsius, -2 degrees Celsius, and -1 degree Celsius.

[0154] Based on the comparative analysis results, a proportional correction algorithm is used to dynamically correct the power drive signal. The correction amount is calculated using the formula ΔP=K×ΔT, where ΔP represents the power correction amount in watts, K is the proportional coefficient (e.g., 2 watts per degree Celsius), and ΔT represents the temperature difference in degrees Celsius. Substituting these values ​​into the calculation, the power correction amounts for each zone are -8 watts, -4 watts, -14 watts, 2 watts, -4 watts, -2 watts, -4 watts, and -2 watts. After superimposing the correction amounts onto the original power adjustment values, the new zone power control commands are obtained as -3 watts, -1 watt, -2 watts, 0 watts, 4 watts, -3 watts, 0 watts, and -5 watts.

[0155] The heating film is controlled by the modified power drive signal. After three control cycles, the propellant temperature in each area of ​​the tank reaches 252 degrees Celsius, 254 degrees Celsius, 250 degrees Celsius, 260 degrees Celsius, 249 degrees Celsius, 254 degrees Celsius, 251 degrees Celsius, and 256 degrees Celsius, forming a temperature field distribution corresponding to the preset conditions.

[0156] In this embodiment, the temperature distribution inside the tank is precisely matched with the preset conditions through real-time temperature monitoring and dynamic adjustment of the power signal, thus ensuring the stability of the tank pressure control.

[0157] To optimize the temperature control strategy throughout the entire mission cycle, in some embodiments, step 105: using a dynamic programming algorithm to perform rolling optimization processing on the mission stage information and the remaining propellant data to generate a target temperature control parameter sequence includes:

[0158] Step 501: Based on the task stage information, determine the current task stage and the remaining unexecuted task stages.

[0159] In step 501, the remaining unexecuted task phases refer to all unfinished work phases from the current moment until the task ends.

[0160] In this embodiment of the application, the current working stage is identified based on the mission stage information provided by the spacecraft control system, and all subsequent mission stages to be executed are listed in chronological order.

[0161] Step 502: Based on the remaining propellant data, predict the trend value of the propellant consumption rate change in each mission phase.

[0162] In step 502, each task stage includes the current task stage and all remaining unexecuted task stages. The optimization process of the dynamic programming algorithm needs to cover the complete sequence of stages from the current stage to the end of the task. The propellant consumption rate change trend value refers to the amount of propellant consumed per unit time in each future stage based on historical consumption data.

[0163] In this embodiment of the application, the changes in the propellant consumption rate at each mission stage are predicted using trend analysis methods based on the current remaining propellant data and historical consumption records.

[0164] Step 503: Based on the propellant consumption rate change trend value, calculate multiple candidate values ​​of temperature control parameters corresponding to each mission stage using the state transition equation of the dynamic programming algorithm.

[0165] In step 503, the candidate value of the temperature control parameter refers to the possible temperature value that meets the basic control requirements, calculated by the state transition equation. The specific expression of the state transition equation can be set according to the actual situation, and this application embodiment does not make specific limitations.

[0166] Step 504: With the goal of maximizing propellant utilization efficiency, select the target values ​​of temperature control parameters for each mission stage from multiple candidate values ​​of temperature control parameters for each mission stage.

[0167] In step 504, the propellant utilization efficiency refers to the ratio of the mass of propellant actually used to generate thrust per unit time to the total mass of propellant consumed. This parameter is calculated by monitoring the change data of the remaining propellant in the tank and combining it with the working status of the thruster. The target value of the temperature control parameter refers to the optimal temperature value that maximizes the propellant utilization efficiency selected from the candidate values.

[0168] In this embodiment of the application, with the goal of maximizing propellant utilization efficiency, the optimal target value of temperature control parameter for each mission stage is determined by comparing and analyzing the expected effects of each candidate value.

[0169] Step 505: Sort the target values ​​of the temperature control parameters corresponding to each task stage according to the order of the task stages to form an initial temperature control parameter sequence.

[0170] In step 505, the initial temperature control parameter sequence refers to the temperature control sequence formed by arranging the target values ​​of each stage in chronological order.

[0171] Step 506: The initial temperature control parameter sequence is periodically corrected through a rolling optimization mechanism to generate the target temperature control parameter sequence.

[0172] In this embodiment, the latest propellant remaining amount data is periodically obtained through a rolling optimization mechanism to correct the initial temperature control parameter sequence and generate the final target temperature control parameter sequence.

[0173] In this embodiment, the global optimization and rolling correction mechanism of the dynamic programming algorithm realizes the forward planning of temperature control parameters throughout the mission cycle. By precisely managing the heating power, the continuous optimization of propellant utilization efficiency is ensured.

[0174] To generate a coordinated and optimized control strategy, in some embodiments, step 106: fusing the proportional optimization results and the target temperature control parameter sequence to form a comprehensive control strategy includes:

[0175] Step 601: Align the power allocation ratio of each partition in the proportional optimization result with the target value of the temperature control parameter at the corresponding time in the target temperature control parameter sequence.

[0176] In this embodiment of the application, time alignment refers to matching corresponding time nodes.

[0177] Step 602: Using a weighted fusion algorithm, the power allocation ratio after time alignment and the target value of the temperature control parameter after time alignment are combined and calculated to obtain the preliminary control strategy after weighted fusion.

[0178] In this embodiment, firstly, weighting coefficients are assigned to the timing-aligned power allocation ratio and the target value of the temperature control parameter, respectively; then, a preliminary control strategy that comprehensively considers both factors is obtained through weighted calculation.

[0179] Step 603: Perform regional modifications on the initial control strategy after weighted fusion to obtain the regionally modified control strategy.

[0180] In this embodiment of the application, the power allocation ratio of the corresponding region in the preliminary control strategy is adjusted in a targeted manner based on the difference in heating demand in each region of the temperature field distribution, so as to form a control strategy that is more in line with actual needs.

[0181] The specific implementation process is as follows: First, identify the regions with high heating demand in the temperature field distribution. Then, increase the power allocation ratio in the preliminary control strategy for these regions, while decreasing the power allocation ratio for regions with low heating demand, so that the control strategy matches the actual heating demand distribution. The regions with high heating demand are those where the propellant evaporation rate is lower than the preset evaporation rate threshold or the temperature is lower than the preset lower temperature limit. The regions with low heating demand are those where the propellant temperature is higher than the preset upper temperature limit.

[0182] Step 604: Verify the coordination between the power allocation ratio of each zone and the target value of the temperature control parameter in the regionally modified control strategy. When the verification is successful, output the modified control strategy as the verified control strategy.

[0183] Step 605: When a conflict is detected between the power allocation ratio in the verified control strategy and the target value of the temperature control parameter, the arbitration mechanism is activated. The verified control strategy is readjusted with tank pressure stability as the priority indicator to obtain the arbitration-adjusted control strategy. The arbitration-adjusted control strategy is then output as the comprehensive control strategy.

[0184] In step 605, the readjustment is an iterative process that may be performed multiple times. The stopping condition is that the conflict between the power distribution ratio and the target value of the temperature control parameter is eliminated. Specifically, the numerical relationship between the two satisfies a preset coordination threshold with tank pressure stability as the priority indicator.

[0185] In this embodiment of the application, when a parameter conflict is detected in the verified control strategy, the combination relationship between the power distribution ratio and the temperature control parameters is readjusted, with the tank pressure stability as the primary consideration, to generate the final integrated control strategy.

[0186] In the embodiments of this application, a coordinated and consistent comprehensive control strategy is formed through multi-step fusion optimization and conflict arbitration, thereby ensuring the stable and reliable operation of the system under various operating conditions.

[0187] To accurately calculate the required power adjustment for each zone, in some embodiments, step 102: calculating the required power adjustment for each zone of the external heating layer of the self-pressurized tank based on the nitrous oxide concentration data and the temperature distribution data includes:

[0188] Step 701: Perform differential analysis on the nitrous oxide concentration data and the temperature distribution data to extract characteristic parameters that reflect the evaporation and condensation state of the propellant.

[0189] In step 701, the characteristic parameter refers to a quantitative index that can characterize the evaporation and condensation state of the propellant and is used to reflect the dynamic change characteristics of the propellant during the phase change process.

[0190] Step 702: Convert the characteristic parameters into the temperature field distribution in the external heating layer of the self-pressurized tank.

[0191] In step 702, the temperature field distribution refers to a quantitative description that can visually show the intensity of heating power required by each region in the tank to meet the propellant phase change requirements.

[0192] In this embodiment, the extracted feature parameters are normalized and mapped to the corresponding spatial location of the external heating layer of the self-pressurized tank, forming a distribution representation that reflects the heating demand level of each region.

[0193] Step 703: Based on the temperature field distribution, determine the heating requirements of each zone of the external heating layer of the self-pressurized tank.

[0194] In step 703, heating demand refers to the amount of heat that each zone needs to provide, as determined by the temperature field distribution.

[0195] In this embodiment of the application, regions with high heating demand and regions with low heating demand are identified based on the temperature field distribution, so as to determine the heating intensity that needs to be increased or decreased in these regions respectively.

[0196] Step 704: Based on the heating requirements, calculate the power adjustment amount required for each zone.

[0197] In this embodiment, the specific power value that needs to be adjusted for each zone is calculated based on the heating requirements of each zone and the power characteristics of the heating element.

[0198] In this embodiment of the application, by extracting features reflecting the propellant state from concentration and temperature data and converting them into specific power adjustment amounts, a precise response to changes in the propellant phase transition state and targeted power compensation are achieved.

[0199] Figure 3 This application provides a schematic diagram of a closed-loop temperature control system for a nitrous oxide propulsion self-pressurization tank, comprising:

[0200] The acquisition module 31 is used to acquire data on nitrous oxide concentration, pressure, temperature distribution of propellant in the self-pressurized tank, mission phase information of the spacecraft, and remaining propellant quantity.

[0201] The calculation module 32 is used to calculate the power adjustment required for each zone of the external heating layer of the self-pressurized tank based on the nitrous oxide concentration data and the temperature distribution data, and to generate a zone power control command corresponding to the power adjustment amount.

[0202] The adjustment module 33 is used to adjust the power output of each zone based on the zone power control command, so as to form a temperature field distribution in the tank that matches the preset conditions.

[0203] The optimization module 34 is used to perform collaborative optimization processing on the temperature field distribution and the pressure data using a reinforcement learning algorithm to generate a proportional optimization result.

[0204] The generation module 35 is used to perform rolling optimization processing on the task stage information and the remaining propellant data using a dynamic programming algorithm to generate a target temperature control parameter sequence.

[0205] The fusion module 36 is used to fuse the proportional optimization results and the target temperature control parameter sequence to form a comprehensive control strategy. The comprehensive control strategy includes an updated partition power control command, which is used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

[0206] The closed-loop temperature control system for the nitrous oxide propulsion self-pressurized tank in this application is used to implement the aforementioned closed-loop temperature control method for the nitrous oxide propulsion self-pressurized tank. Therefore, the specific implementation of the closed-loop temperature control system for the nitrous oxide propulsion self-pressurized tank can be found in the embodiment section of the nitrous oxide propulsion self-pressurized tank temperature closed-loop control method above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0207] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described nitrous oxide propulsion self-pressurizing tank temperature closed-loop control method.

[0208] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described nitrous oxide propulsion self-pressurizing tank temperature closed-loop control method.

[0209] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0210] The embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the embodiments of the nitrous oxide propulsion self-pressurizing tank temperature closed-loop control method described above.

[0211] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0212] The above provides a detailed description of the closed-loop temperature control method and system for a nitrous oxide propulsion self-pressurizing tank provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A closed-loop temperature control method for a nitrous oxide-propelled self-pressurized tank, characterized in that, include: Acquire data on nitrous oxide concentration, pressure, propellant temperature distribution, spacecraft mission phase information, and remaining propellant quantity in the pressurized tank; Based on the nitrous oxide concentration data and the temperature distribution data, the required power adjustment amount for each zone of the external heating layer of the self-pressurized tank is calculated, and a zone power control command corresponding to the power adjustment amount is generated. Based on the partition power control command, the power output of each partition is adjusted to form a temperature field distribution in the tank that matches the preset conditions. The temperature field distribution and the pressure data are jointly optimized using a reinforcement learning algorithm to generate a proportional optimization result. Using a dynamic programming algorithm, the task phase information and the remaining propellant data are subjected to rolling optimization to generate a target temperature control parameter sequence; The proportional optimization results and the target temperature control parameter sequence are combined to form a comprehensive control strategy. The comprehensive control strategy includes an updated partition power control command, which is used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

2. The method according to claim 1, characterized in that, The step of using reinforcement learning algorithms to collaboratively optimize the temperature field distribution and the pressure data to generate proportional optimization results includes: Based on the temperature field distribution and the pressure data, a preliminary scheme containing the power allocation ratio of each partition is generated through a policy network using a reinforcement learning algorithm. The pressure stability gains of the preliminary scheme in terms of pressure stability and temperature field uniformity in the self-pressurized tank are evaluated through the value network of the reinforcement learning algorithm. The policy optimization module of the reinforcement learning algorithm updates the preliminary scheme through multiple rounds of iteration based on the pressure-stabilized reward value. In each round of iteration, the exploration mechanism of the reinforcement learning algorithm generates a proportional optimization result containing the power allocation ratio of each partition.

3. The method according to claim 2, characterized in that, The policy optimization module using the reinforcement learning algorithm updates the initial scheme through multiple iterations based on the stable pressure reward value. In each iteration, the algorithm's exploration mechanism generates a proportional optimization result containing the power allocation ratio of each partition, including: Based on the sign and magnitude of the pressure stability benefit value, determine the adjustment direction and magnitude of the preliminary plan; Based on the adjustment direction and the adjustment range, the preliminary scheme is optimized to generate an optimized scheme; Based on the optimization scheme, the partition power perturbation components corresponding to each partition that conform to a normal distribution are generated through an exploration mechanism; The partition power disturbance component is superimposed onto the corresponding partition power allocation ratio in the optimization scheme to form a power allocation scheme. Through the value network, the pressure stability benefit value of the power allocation scheme is compared with the pressure stability benefit value of the optimization scheme. When the comparison result shows that the pressure stability benefit value of the power allocation scheme is greater than the pressure stability benefit value of the optimization scheme, the power allocation scheme is used as the initial scheme for a new round of iteration. Repeat the above process until the preset number of iterations is reached, and output the final optimized partition power allocation ratio scheme, which includes the ratio optimization results of the power allocation ratio of each partition.

4. The method according to claim 1, characterized in that, The step of adjusting the power output of each zone based on the zoned power control command to form a temperature field distribution within the tank that matches preset conditions includes: The partition power control command is converted into a power drive signal corresponding to each partition. The power output of each zone heating film is controlled according to the power drive signal, and the corresponding heating effect is produced. Real-time propellant temperature data for each region within the tank under the heating effect are collected using a temperature detection device. The real-time temperature data is compared and analyzed with the expected temperature distribution determined based on the gas-liquid phase change requirements; Based on the comparative analysis results, the power drive signal is dynamically corrected, and the heating film of each zone is controlled by the corrected power drive signal to form a temperature field distribution corresponding to the preset conditions.

5. The method according to claim 1, characterized in that, The step of using a dynamic programming algorithm to perform rolling optimization on the mission phase information and the remaining propellant data to generate a target temperature control parameter sequence includes: Based on the task stage information, determine the current task stage and the remaining unexecuted task stages; Based on the remaining propellant data, predict the trend of propellant consumption rate changes in each mission phase; Based on the propellant consumption rate change trend value, the state transition equation of the dynamic programming algorithm is used to calculate multiple candidate values ​​of temperature control parameters corresponding to each mission stage. With the goal of maximizing propellant utilization efficiency, target values ​​for temperature control parameters corresponding to each mission stage are selected from multiple candidate values ​​of temperature control parameters corresponding to each mission stage. The target values ​​of the temperature control parameters corresponding to each task stage are sorted according to the order of the task stages to form an initial temperature control parameter sequence. The initial temperature control parameter sequence is periodically corrected through a rolling optimization mechanism to generate the target temperature control parameter sequence.

6. The method according to claim 1, characterized in that, The integrated control strategy, formed by combining the proportional optimization results and the target temperature control parameter sequence, includes: The power allocation ratio of each partition in the ratio optimization result is time-sequentially aligned with the target value of the temperature control parameter at the corresponding time in the target temperature control parameter sequence. By using a weighted fusion algorithm, the power allocation ratio after time alignment and the target value of the temperature control parameter after time alignment are combined and calculated to obtain the preliminary control strategy after weighted fusion. The initial control strategy after weighted fusion is regionally modified to obtain the regionally modified control strategy; Verify the coordination between the power allocation ratio of each zone and the target value of the temperature control parameter in the regionally modified control strategy. When the verification is successful, output the modified control strategy as the verified control strategy. When a conflict is detected between the power allocation ratio in the verified control strategy and the target value of the temperature control parameter, an arbitration mechanism is initiated. The verified control strategy is readjusted with tank pressure stability as the priority indicator to obtain the arbitration-adjusted control strategy, which is then output as the comprehensive control strategy.

7. The method according to claim 1, characterized in that, The calculation of the required power adjustment for each zone of the external heating layer of the self-pressurized tank based on the nitrous oxide concentration data and the temperature distribution data includes: Differential analysis was performed on the nitrous oxide concentration data and the temperature distribution data to extract characteristic parameters that reflect the evaporation and condensation state of the propellant. The characteristic parameters are converted into the temperature field distribution in the external heating layer of the self-pressurized tank. Based on the temperature field distribution, the heating requirements of each zone of the external heating layer of the self-pressurized tank are determined; Based on the heating requirements, calculate the power adjustment amount required for each zone.

8. A closed-loop temperature control system for a nitrous oxide-propelled self-pressurized tank, characterized in that, include: The acquisition module is used to acquire data on nitrous oxide concentration, pressure, temperature distribution of propellant in the self-pressurized tank, mission phase information of the spacecraft, and remaining propellant quantity. The calculation module is used to calculate the power adjustment required for each zone of the external heating layer of the self-pressurized tank based on the nitrous oxide concentration data and the temperature distribution data, and to generate a zone power control command corresponding to the power adjustment amount. The adjustment module is used to adjust the power output of each zone based on the zone power control command, so as to form a temperature field distribution in the tank that matches the preset conditions. The optimization module is used to perform collaborative optimization processing on the temperature field distribution and the pressure data using reinforcement learning algorithms to generate proportional optimization results; The generation module is used to perform rolling optimization processing on the task stage information and the remaining propellant data using a dynamic programming algorithm to generate a target temperature control parameter sequence. The fusion module is used to fuse the proportional optimization results and the target temperature control parameter sequence to form a comprehensive control strategy. The comprehensive control strategy includes updated partition power control instructions, which are used to adjust the power output of each partition in real time to achieve closed-loop temperature control for propellant state changes.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the closed-loop temperature control method for a nitrous oxide propulsion self-pressurized tank as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the closed-loop temperature control method for a nitrous oxide propulsion self-pressurizing tank as described in any one of claims 1 to 7.

Images