Intelligent control method and system for removing tritium retention facing plasma material

Abstract

The invention is suitable for the technical field of industrial safety control, and provides an intelligent control method and system for removing plasma-oriented material tritium retention, and the method comprises the following steps: monitoring and collecting state data in a tritium removal process in real time; identifying the stage of the current tritium removal process; receiving current state information and preset tritium removal process parameters, and predicting a desorption trend curve in a future time period according to a prediction model based on diffusion and desorption physical rules of tritium in the plasma-oriented material; comparing the predicted desorption trend curve with the tritium desorption rate monitored in real time to obtain a deviation; if the deviation meets a preset judgment condition, generating a new control instruction; and according to the control instruction, dynamic adjustment of tritium removal process parameters and cooperative switching of multiple methods are realized. According to the method, the tritium removal state can be sensed in real time, the tritium removal process parameters are dynamically adjusted according to the tritium removal state, and precision, self-adaption and optimization of the tritium removal process are achieved.

Description

Intelligent control method and system for removing tritium retention in plasma-oriented materials Technical Field

[0001] This invention belongs to the field of industrial safety control technology, and in particular relates to an intelligent control method and system for removing tritium retention in plasma-oriented materials. Background Technology

[0002] Magnetic confinement fusion energy is considered one of the ideal ways to solve future energy problems. During the operation of devices such as tokamak, tritium fuel plasma interacts strongly with plasma-facing materials (PFMs). Some tritium becomes trapped in the PFMs. As operating time accumulates, the total amount of trapped tritium increases, posing two major challenges: first, safety risks, as tritium is a radioactive isotope, and its potential release poses a threat to personnel and the environment; second, economic and sustainability risks, as tritium resources are extremely scarce and expensive, and excessive trapping reduces fuel cycle efficiency, limiting the long-term continuous operation of fusion reactors. Therefore, regular, efficient, and thorough detritium removal from PFMs is an indispensable key element for the safe maintenance and economical operation of fusion reactors.

[0003] Currently, existing tritium removal technologies all work by providing sufficient energy (such as thermal, chemical, or kinetic energy) to the trapped tritium atoms, enabling them to overcome the trap binding energy and desorb. The main methods include:

[0004] Thermal baking method: This method promotes tritium diffusion and desorption by heating the material, but the high temperature may damage the internal components of the fusion reactor.

[0005] Isotope exchange method: introduce substitute gases such as deuterium to remove surface tritium through isotope exchange reaction;

[0006] Plasma cleaning method: This method utilizes the interaction between plasma and the material surface to remove tritium through physical sputtering or chemical reaction.

[0007] Laser ablation: This method removes tritium by local heating through laser irradiation, but the equipment is complex and the cost is high.

[0008] However, fusion reactor internals (especially ITER, CFETR, etc.) impose clear cryogenic operating limitations on PFMs (typically ≤350°C). Within this "cryogenic" window, vacuum baking methods relying solely on thermal diffusion are extremely inefficient and have very limited ability to remove tritium from bulk phases and deep defects. Therefore, developing tritium removal technologies that can still operate efficiently under cryogenic confinement has become an urgent need in this field.

[0009] Furthermore, in the prior art, a paper published in the *Journal of Nuclear Materials* discloses a "stepwise isotope exchange and baking" method. The specific operational procedure of this method is as follows: the tritium-containing tungsten component to be treated is placed in an ultra-high vacuum chamber; the heating system is activated to stabilize the component temperature at a preset value (e.g., 300°C) and maintain it for a period of time (e.g., 2 hours) for vacuum thermal desorption; while maintaining a constant temperature, deuterium gas (D2) at a certain pressure is introduced into the vacuum chamber, and the isotope exchange reaction is allowed to proceed for another preset time (e.g., 4 hours); the deuterium gas supply is stopped, the high vacuum is re-evacuated, and baking may be performed for a further period to remove the surface-adsorbed deuterium; after the entire process is completed, samples are taken to analyze the detritium removal rate.

[0010] While this method recognizes the limitations of thermal desorption alone and introduces a chemical displacement mechanism, representing a significant advancement, its core drawback lies in its entirely sequential control process based on a pre-set program. All parameters (temperature, pressure, and duration of each step) are set before the process begins, lacking monitoring of key state variables (such as real-time desorption rate) and any mechanism for dynamic feedback adjustment based on monitoring results. Its effectiveness is highly dependent on the operator's understanding of specific material batches and historical experience, making it a trial-and-error open-loop control method.

[0011] In summary, existing tritium removal technologies are mostly open-loop, experience-driven, single-process technologies. Their fundamental flaw lies in their inability to perceive the actual state of the tritium removal process, resulting in the following core problems:

[0012] 1. Low energy utilization efficiency and the existence of a "processing blank zone": Existing technologies employ a fixed energy input mode (such as constant temperature). Under low-temperature constraints, this uniform energy field cannot intelligently match the non-uniform distribution characteristics of tritium retention. For shallow traps, the fixed energy may be excessive, posing a risk of over-processing; while for deep-level traps requiring higher activation energies, the fixed energy is insufficient, resulting in a large number of deeply trapped tritium atoms not being effectively removed, forming a "processing blank zone" that limits the upper limit of tritium removal depth.

[0013] 2. Lack of transparency and data-driven optimization: Existing technologies treat the tritium removal process as a "black box." Operators can only set the inputs (process parameters) and measure the output (final tritium removal rate) after the process is complete. Key information such as the dynamic progress of tritium removal, the dominant mechanism (thermal desorption or isotope exchange), and the primary depth from which the removed tritium originates is completely missing. This makes process optimization heavily reliant on researchers' experience and numerous repetitive experiments, rather than data-driven scientific decision-making, resulting in long development cycles and difficulty in obtaining a globally optimal solution.

[0014] 3. Inability to achieve intelligent synergy among multiple methods: When more methods (such as pulsed plasma) need to be introduced, existing technologies lack a "brain" to decide when to introduce them, at what intensity, and how to coordinate the interactions between different methods. Simple sequential execution may lead to energy waste or mutual interference.

[0015] 4. Lack of anti-interference and adaptive correction capabilities: When there are batch differences in the processed objects (such as different material microstructures) or slight fluctuations occur in the process (such as slight changes in vacuum), existing technologies with fixed parameters cannot detect these disturbances, let alone perform real-time compensation and correction. This directly leads to large fluctuations in process stability and processing results, making it difficult to meet the high reliability requirements of fusion devices for maintenance processes.

[0016] 5. Optimization relies on experience, not scientific prediction: The optimization of process parameters depends entirely on the experience of the experimenters and a lot of trial and error, rather than scientific decision-making based on physical models and data, resulting in a long development cycle and difficulty in obtaining the global optimal solution.

[0017] Therefore, there is an urgent need to build a closed-loop control system that can sense the tritium removal status in real time and dynamically adjust the tritium removal process parameters accordingly. Summary of the Invention

[0018] The purpose of this invention is to provide an intelligent control method for removing tritium retention in plasma-oriented materials, thereby solving the aforementioned technical problems.

[0019] A smart control method for removing tritium retention in plasma-oriented materials includes the following steps:

[0020] The status data during the tritium removal process is monitored and collected in real time to form a real-time data stream; the status data includes tritium desorption rate, temperature, pressure, and gas flow rate;

[0021] Based on the real-time data stream, the current stage of the detritium removal process is identified;

[0022] The system receives current status information and preset tritium removal process parameters, and predicts the desorption trend curve in the future time period based on a prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials. The status information includes the stage of the tritium removal process and status data.

[0023] The predicted desorption trend curve is compared with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets the preset judgment condition, a new control command is generated based on the current state information and the preset optimization strategy.

[0024] According to the control commands, the dynamic adjustment of the tritium removal process parameters and the coordinated switching of multiple methods can be realized.

[0025] Furthermore, the tritium desorption rate is monitored by a quadrupole mass spectrometer; the temperature is the surface temperature of the plasma-facing material; the pressure is the pressure of the vacuum chamber used for tritium removal; and the gas flow rate is the inlet gas flow rate.

[0026] Furthermore, the detritium removal process is in any one of the following stages: surface desorption period, bulk diffusion control period, and deep trap desorption period.

[0027] Furthermore, the prediction model is a kinetic model based on the activation energy distribution and diffusion equation. Its initial parameters are preset according to the type of plasma-oriented material and are calibrated online based on the tritium removal process parameters.

[0028] Furthermore, the determination condition is as follows:

[0029] If the real-time monitored tritium desorption rate is 10%-20% lower than the predicted value and continues for more than the preset duration; or if the real-time monitored tritium desorption rate enters a plateau period for more than the set time; then it is determined that the current tritium removal process effect has not met expectations, and new control commands are generated based on the current status information and the preset optimization algorithm.

[0030] Furthermore, the optimization strategy is as follows: logical judgment is performed based on a preset rule base; and / or optimization of multiple parameters such as temperature, pressure, gas type and flow rate, and plasma discharge power is performed based on an optimization algorithm within the low-temperature constraint range.

[0031] Furthermore, the dynamic adjustment of the tritium removal process parameters includes at least one of the following adjustment methods:

[0032] The heating power is adjusted by a temperature controller to stabilize the surface temperature of the plasma-facing material within a preset range.

[0033] The gas flow controller automatically adjusts the opening of the proportional valve through its internal closed-loop control circuit to achieve the preset target gas flow rate.

[0034] The plasma power supply's start / stop, power, and pulse parameters are controlled via its controller.

[0035] Furthermore, the coordinated switching of the multiple methods includes automatically switching or combining at least two of the following processes during the detritium removal process: vacuum thermal desorption, isotope exchange, and plasma discharge cleaning-assisted desorption.

[0036] Another object of the present invention is to provide an intelligent control system for removing tritium retention in plasma-oriented materials, for implementing the above-mentioned intelligent control method, comprising:

[0037] The data sensing layer is used to monitor and collect status data during the tritium removal process in real time, forming a real-time data stream; the status data includes tritium desorption rate, temperature, pressure, and gas flow rate;

[0038] The status recognition module is used to identify the current stage of the detritium removal process based on the real-time data stream;

[0039] The prediction module is used to receive current status information and preset tritium removal process parameters, and predict the desorption trend curve in the future time period based on a prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials; the status information includes the stage of the tritium removal process and status data.

[0040] The optimization decision module is used to compare the predicted desorption trend curve with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets the preset judgment conditions, a new control command is generated based on the current state information and the preset optimization strategy.

[0041] The execution control layer is used to dynamically adjust the tritium removal process parameters and coordinate the switching of multiple methods according to the control instructions.

[0042] The intelligent control method for removing tritium retention in plasma-oriented materials provided by this invention can sense the tritium removal status in real time and dynamically adjust the tritium removal process parameters accordingly to achieve closed-loop control. This ensures the safety of the plasma-oriented material structure (low temperature, low damage) while achieving precision, adaptability and optimization of the tritium removal process. Attached Figure Description

[0043] Figure 1 is a schematic diagram of the structure of the intelligent control method for removing tritium retention in plasma-oriented materials provided in an embodiment of the present invention.

[0044] Figure 2 is a schematic flowchart of an intelligent control system for removing tritium retention in plasma-oriented materials provided in an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0046] As shown in Figure 1, in one embodiment of the present invention, a smart control method for removing tritium retention in plasma-oriented materials is provided, comprising the following steps:

[0047] S1. Real-time monitoring and acquisition of status data during the tritium removal process to form a real-time data stream; the status data includes tritium desorption rate, temperature, pressure, and gas flow rate;

[0048] S2. Based on the real-time data stream, identify the current stage of the tritium removal process;

[0049] S3. Receive the current status information and preset tritium removal process parameters, and predict the desorption trend curve in the future time period based on the prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials; the status information includes the stage of the tritium removal process and status data.

[0050] S4. Compare the predicted desorption trend curve with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets the preset judgment condition, generate a new control command based on the current state information and the preset optimization strategy; otherwise, end the operation, record the data, form a database AI autonomous learning to optimize the above prediction model and optimization strategy.

[0051] S5. Based on the control instructions, dynamically adjust the tritium removal process parameters and coordinate the switching of multiple methods.

[0052] Plasma-facing materials (PFMs) refer to materials that directly face the high-temperature plasma in nuclear fusion devices, such as tungsten, molybdenum, and their alloys. Tritium retention refers to the phenomenon where tritium, the fusion fuel, is adsorbed and trapped on the surface and internal defects of PFMs under irradiation and plasma scouring. The tritium desorption rate refers to the mass of tritium-containing substances desorbed from the surface of PFMs per unit time, as monitored in real time by a quadrupole mass spectrometer, and is a key indicator reflecting the tritium removal process.

[0053] In practical applications, the basic workflow of this method is as follows: continuously collect data → analyze the data and determine whether the current state matches the expectation → if not, calculate a new set of optimized parameters → drive the device to run according to the new parameters. This process is repeated until the process termination condition is met.

[0054] In this embodiment of the invention, by integrating real-time sensing technology, physical mechanism models, and optimized control algorithms, real-time monitoring, status assessment, dynamic decision-making, and precise execution of the tritium removal process are achieved. This systematically overcomes the shortcomings of existing technologies, ultimately realizing the precision, adaptability, and optimization of the PFMs tritium removal process. Furthermore, this embodiment of the invention can also accumulate full-process data from each tritium removal process to construct an iteratively optimized process knowledge base, enabling the system to continuously improve its adaptability to different material properties and operating conditions, ultimately achieving a highly efficient, reliable, reproducible, and self-improving tritium removal effect.

[0055] Correspondingly, as shown in Figure 2, in another embodiment of the present invention, an intelligent control system for removing tritium retention in plasma-oriented materials is also provided. It adopts a closed-loop control architecture of perception-decision-execution to realize autonomous optimization and adaptive management of the tritium removal process. Specifically, it includes a data perception layer, a decision optimization layer and an execution control layer.

[0056] The data sensing layer is used to monitor and collect status data during the tritium removal process in real time, forming a real-time data stream; the status data includes tritium desorption rate, temperature, pressure, and gas flow rate;

[0057] The decision optimization layer includes:

[0058] The status recognition module is used to identify the current stage of the detritium removal process based on the real-time data stream;

[0059] The prediction module is used to receive current status information and preset tritium removal process parameters, and predict the desorption trend curve in the future time period based on a prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials; the status information includes the stage of the tritium removal process and status data.

[0060] The optimization decision module is used to compare the predicted desorption trend curve with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets the preset judgment conditions, a new control command is generated based on the current state information and the preset optimization strategy.

[0061] The execution control layer includes a process control layer and a physical execution layer, which are used to dynamically adjust the tritium removal process parameters and coordinate the switching of multiple methods according to the control instructions.

[0062] In practical applications, the data sensing layer consists of various sensors deployed in the tritium removal process vacuum chamber and its pipelines, responsible for real-time acquisition of all raw data reflecting the process status. The decision optimization layer, also known as the central processing and decision-making layer, is the brain of the system, composed of an industrial computer and dedicated software, responsible for data analysis, status identification, trend prediction, and optimization decisions. The execution control layer consists of various execution units (such as heaters, pneumatic valves, and plasma power supplies) and their associated controllers, responsible for receiving instructions and driving the physical process. Correspondingly, the physical execution layer includes vacuum and baking units (including heaters and other execution units), pretreatment units (Nd:YAG lasers (nanosecond pulses)) for laser ablation and removal of atomic-level material surfaces, gas injection and mixing units (including pneumatic valves and other execution units), and plasma discharge units (including plasma power supplies and other execution units). It should be noted that in applications with lower requirements, the decision optimization layer and process control layer can be merged to form a lightweight embedded intelligent controller.

[0063] In a preferred embodiment of the present invention, the data sensing layer includes core sensors and auxiliary sensors. The core sensor is a quadrupole mass spectrometer installed on the tritium removal process vacuum chamber, used to monitor and output the signal intensity of particles with different mass numbers in real time, especially signals related to tritium desorption (such as HT, DT, T2, etc.), whose signal intensity directly reflects the real-time tritium desorption rate. Temperature refers to the surface temperature of the PFMs; pressure refers to the pressure of the vacuum chamber used for tritium removal; and gas flow rate refers to the inlet gas flow rate. Auxiliary sensors include thermocouples (monitoring the surface temperature of the PFMs), pressure gauges (monitoring the vacuum chamber pressure), and flow controllers (monitoring and providing feedback on the inlet gas flow rate). It should be noted that all sensor data can be transmitted to the decision optimization layer in real time via a data acquisition card.

[0064] In a preferred embodiment of the present invention, the state recognition module is used to extract and process the real-time data stream from the data perception layer, calculate the tritium desorption rate at the current moment, and combine it with historical data to identify the stage of the current tritium removal process; the stage of the tritium removal process includes any one of the following: surface desorption period, bulk diffusion control period, and deep trap desorption period.

[0065] In a preferred embodiment of the present invention, the prediction model is a kinetic model based on activation energy distribution and diffusion equations. Its initial parameters (such as activation energy and diffusion coefficient) are preset according to the type of PFMs and calibrated online based on tritium removal process parameters. By receiving current state information (such as temperature and current desorption rate) and preset tritium removal process parameters, the prediction model can predict the desorption trend curve over a future period (such as the next 10 minutes). It should be noted that the above prediction model, a core physics-driven model, can also be replaced by a data-driven model based on machine learning; for example, a time-series prediction model (such as a Long Short-Term Memory network LSTM) can be trained, using historical tritium removal data (process parameters as input, desorption curve as output) to establish predictive relationships.

[0066] In a preferred embodiment of the present invention, the above-mentioned determination condition is as follows: if the real-time monitored tritium desorption rate is lower than 10%-20% of the predicted value and continues for more than a preset duration; or if the real-time monitored tritium desorption rate enters a plateau period exceeding a set time; then it is determined that the current tritium removal process effect has not met expectations, and a new control command is generated based on the current status information and a preset optimization algorithm. Furthermore, the above-mentioned optimization strategy is as follows: logical judgment is performed based on a preset rule base; and / or multi-parameter optimization is performed on temperature, pressure, gas type and flow rate, plasma power and duty cycle within the low-temperature constraint range based on the optimization algorithm.

[0067] In practical applications, the workflow of the optimization decision-making module is as follows:

[0068] Initial decision: Before the operation begins, the system uses a predictive model to perform simulation calculations based on the target PFMs material type and the set removal rate target, and recommends a set of initial process parameters (e.g., the starting temperature is set to 250℃, and hydrogen (deuterium or hydrogen-deuterium mixture) is introduced).

[0069] Real-time decision-making: During operation, the optimization decision-making module continuously compares the predicted desorption curve with the actual desorption curve measured by the mass spectrometer. Once a significant deviation occurs (such as the measured value being more than 15% lower than the predicted value), it is determined that the current process effect has not met expectations.

[0070] Dynamic optimization: At this point, the optimization decision module will activate the optimization strategy (such as a preset rule base or a simple optimization algorithm in the existing technology) and generate new control instructions. For example: "The current isotope exchange efficiency is decreasing. It is recommended to raise the temperature to 280℃ and start low-power plasma-assisted cleaning. Select plasma parameters: gas type (D2, He, Ar), ion energy, RF power, gas flow rate, etc."

[0071] In a preferred embodiment of the present invention, the execution control layer includes a temperature controller (connected to the heater), a gas flow controller (connected to the gas cylinder and the valve), a plasma power supply controller, etc., for receiving control commands from the decision optimization layer, converting them into specific, executable signals (such as analog voltage / current signals), precisely controlling the actions of each actuator, and realizing dynamic adjustment of tritium removal process parameters and coordinated switching of multiple methods.

[0072] The dynamic adjustment of the tritium removal process parameters includes at least one of the following adjustment methods:

[0073] The temperature setpoint control command of the decision optimization layer is sent to the temperature controller, which adjusts the heating power (the power supplied to the resistance heater or infrared heating lamp array) to stabilize the surface temperature of PFMs within the preset range or set value.

[0074] The flow setpoint control command and parameters of the decision optimization layer are sent to the gas flow controller. The gas flow controller automatically adjusts the opening of the proportional valve through the closed-loop control circuit inside the gas flow controller so that the gas flow reaches the preset target value.

[0075] The decision optimization layer sends the switching control commands and parameters to the controller of the radio frequency (RF) or pulsed DC plasma power supply. Through the controller of the plasma power supply, the start-up, shutdown, power and pulse parameters of the plasma power supply are precisely controlled.

[0076] In addition, the coordinated switching of multiple methods includes automatically switching or combining at least two of the following processes during the detritium removal process: vacuum thermal desorption, hydrogen isotope exchange, and plasma discharge cleaning-assisted desorption.

[0077] In a preferred embodiment of the invention, the system further includes a dedicated database for an adaptive learning mechanism, recording "full-process data" for each complete tritium removal process, including but not limited to: time-series data from all sensors, all instructions issued by the decision optimization layer, and subsequent offline analysis results (such as the final tritium removal rate). Periodically (e.g., after processing every 10 samples), this accumulated data can be used to refit and calibrate key physical parameters (such as the diffusion coefficient) in the prediction model using parameter estimation algorithms (such as the least squares method). This allows the model to better reflect the characteristics of the actual system, and with the accumulation of operational experience, the overall control performance of the system continuously improves.

[0078] The following embodiments are practical application examples of the methods and systems provided by the present invention, and are only illustrative examples and are not limited thereto.

[0079] Example 1: Adaptive energy enhancement for deep tritium retention, which addresses the problem that existing fixed-temperature baking methods are ineffective at removing tritium from deep-level traps.

[0080] In this embodiment of the invention, the system's workflow is as follows:

[0081] Start: The system runs according to the initial parameters (baking at 200℃).

[0082] Monitoring: Quadrupole mass spectrometry showed that the desorption rate decreased rapidly after the initial peak and then entered a plateau period.

[0083] Decision: The decision optimization layer determines that the current period is "diffusion control period" and the current energy is insufficient, so it generates the instruction: "Increase the temperature from 200℃ to 240℃; increase the deuterium flow rate by 20%".

[0084] Execution and Results: The execution control layer precisely adjusted the parameters. Subsequently, the mass spectrometer detected a new peak in the desorption rate, proving that deep-layer tritium was effectively activated and removed. This breakthrough overcomes the "processing blank zone" under fixed parameters.

[0085] Example 2: The introduction of multi-method intelligent collaboration addresses the problem that the efficiency of a single method has reached its limit, and a new method needs to be introduced, but there are concerns about parameter mismatch.

[0086] In this embodiment of the invention, the system's workflow is as follows:

[0087] Start: The system is undergoing isotope exchange at 200°C.

[0088] Monitoring: The desorption rate has once again entered a low-level plateau period.

[0089] Decision: The decision optimization layer determines the timing for introducing plasma assistance based on the rules and generates a coordinated instruction: "Maintain 200°C; initiate plasma discharge cleaning (D2, ion energy 30eV, RF power 200W, gas pressure 10Pa); reduce deuterium flow rate to the maintenance value."

[0090] Implementation and Results: The system smoothly introduced plasma cleaning, creating a synergistic effect. Mass spectrometry showed a further increase in the desorption rate, and material damage was avoided by controlling plasma parameters. This achieved a smart synergy where "1+1>2".

[0091] Example 3: Adaptive adjustment for material batch differences, which solves the problem that different batches of PFMs (different irradiation conditions or different materials) have different microstructures (such as grain size and defect density), resulting in different tritium retention depth distributions; existing fixed processes cannot differentiate between them, resulting in insufficient or excessive treatment.

[0092] In this embodiment of the invention, the system's workflow is as follows:

[0093] Start: The system initiates the standard initial process for a new batch of PFMs samples (e.g., setting it to bake at 240℃).

[0094] Monitoring: The peak value of the initial tritium desorption rate detected by the quadrupole mass spectrometer was significantly higher than the average record of similar materials in the system database.

[0095] Decision: The state recognition module determines that the tritium retention concentration on the material surface is abnormally high. Based on the rule base, the optimization decision module immediately generates the instruction to avoid overheating or blistering of local micro-areas of the material due to excessively rapid surface desorption: "Reduce the heating rate from 5℃ / min to 2℃ / min; hold at 200℃ for 15 minutes."

[0096] Execution and Results: The system employs a gentle desorption scheme, allowing for the gradual release of surface tritium and avoiding potential material damage risks. This demonstrates the system's adaptability to the individual differences of the processed objects.

[0097] Example 4: Dynamic correction to deal with sudden abnormalities in the process. The problem it solves is: during the process, a small leak may occur in the vacuum system, which will cause the background pressure to rise and interfere with the isotope exchange efficiency; the traditional open-loop system cannot detect this interference and will continue to process it ineffectively.

[0098] In this embodiment of the invention, the system's workflow is as follows:

[0099] Start: The system is undergoing isotope exchange at 240°C, and deuterium gas is being introduced.

[0100] Monitoring: The pressure gauge detected an abnormal, continuous, slight increase in the vacuum chamber pressure. Simultaneously, the quadrupole mass spectrometer showed that the tritium desorption rate began to deviate from the predicted curve and decreased.

[0101] Decision-making: The decision optimization layer integrates two signals—abnormal pressure and decreased desorption rate—to diagnose a "vacuum degradation" fault. The optimization decision module does not simply alarm and shut down, but generates corrective instructions: "Fine-tune the deuterium gas inlet flow rate to compensate for pressure changes and maintain effective exchange pressure; at the same time, extend the processing time by the estimated 20% to make up for efficiency loss."

[0102] Execution and Results: The system automatically compensates for external interference, ensuring the final processing effect and demonstrating strong anti-interference and fault tolerance capabilities.

[0103] Example 5: Depth optimization based on desorption kinetics, which solves the problem that tritium in traps at different depths requires different activation energies even for the same batch of materials; a fixed temperature cannot achieve precise matching of energy input.

[0104] In this embodiment of the invention, the system's workflow is as follows:

[0105] Start: The system has completed the removal of surface tritium and has entered the bulk diffusion control stage.

[0106] Monitoring: The decision optimization layer performs kinetic analysis on the real-time desorption rate curve (such as calculating the apparent activation energy) and finds that the curve characteristics indicate that the current main desorbed tritium originates from traps of medium depth.

[0107] Decision: Predictive model simulations show that if the temperature is gradually increased from the current 240°C to 300°C, and hydrogen is intermittently introduced, tritium at this energy level can be desorbed more efficiently, while avoiding the potential risks of higher temperatures to the material; the optimization decision module generates control commands accordingly.

[0108] Execution and Results: The system achieves "precision strikes" against tritium retention at specific depths, rather than "indiscriminate bombardment," greatly improving energy utilization efficiency and removal depth while ensuring safety.

[0109] Example 6: The manifestation of the system's self-learning function, which solves the problem of how the system becomes more intelligent with increasing number of runs.

[0110] In this embodiment of the invention, the system's workflow is as follows:

[0111] Record: After processing 10 PFMs samples of the same type, the database has recorded 10 sets of complete "process parameters-desorption curve-final removal rate" data.

[0112] Learning: The system initiated a self-learning program and used this data to recalibrate the value of the tritium diffusion coefficient of the material in the prediction model through a parameter estimation algorithm. It was found that the calibrated value was about 15% higher than the initial theoretical value.

[0113] Application: When processing the 11th sample, the system used the calibrated new model for prediction and optimization. Its recommended initial temperature was 10°C lower than before because the new prediction model predicted faster diffusion, eliminating the need for excessively high temperatures.

[0114] Results: The treatment results demonstrate that the same removal effect was achieved with lower energy consumption, realizing energy efficiency optimization. This reflects the intelligent characteristics of the system, which learns from experience and continuously improves.

[0115] In summary, the intelligent control method and system for removing tritium retention in plasma-oriented materials provided by the embodiments of the present invention have the following beneficial technical effects compared with the prior art:

[0116] 1. Achieved closed-loop adaptive control: This embodiment of the invention monitors the tritium desorption rate in real time as a direct parameter and uses it as a feedback signal to dynamically adjust the process parameters, enabling the system to automatically adapt to materials and external disturbances in different initial states. This solves the core problems of unstable processing effect and poor anti-interference ability caused by fixed parameters in the prior art.

[0117] 2. Enhanced scientific rigor in decision-making: The predictive model built into this invention provides a scientific basis for short-term prediction and optimization. Under low-temperature constraints, the system can proactively seek better process paths based on real-time data, significantly reducing reliance on operators' personal experience and potentially achieving deeper and more thorough tritium removal within a shorter cycle.

[0118] 3. Achieves intelligent collaboration among multiple methods: The decision optimization layer of this invention can flexibly and dynamically adjust different units such as baking temperature, hydrogen isotope exchange, and plasma cleaning based on real-time feedback of process status. It can decide when to introduce which method and at what intensity, and coordinate parameter matching between methods, thereby truly achieving a synergistic effect of "1+1>2" while avoiding the risks caused by parameter mismatch.

[0119] It should be noted that each of the above modules can be implemented as a computer program, which can run on a computer device. The computer device's memory can store the computer program that makes up each module, enabling the processor to execute each step of the above method.

[0120] It should be understood that although the steps in the flowcharts of the embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0121] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A smart control method for removing tritium retention in plasma-oriented materials, characterized in that, The process includes the following steps: real-time monitoring and acquisition of state data during the tritium removal process to form a real-time data stream; the state data includes tritium desorption rate, temperature, pressure, and gas flow rate; based on the real-time data stream, identifying the current stage of the tritium removal process; receiving current state information and preset tritium removal process parameters, and predicting the desorption trend curve for a future time period based on a prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials; the state information includes the stage of the tritium removal process and state data; comparing the predicted desorption trend curve with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets preset judgment conditions, generating a new control command based on the current state information and a preset optimization strategy; and realizing dynamic adjustment of tritium removal process parameters and coordinated switching of multiple methods according to the control command.

2. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The tritium desorption rate is monitored by a quadrupole mass spectrometer; the temperature is the surface temperature of the plasma-facing material; the pressure is the pressure of the vacuum chamber used for tritium removal; and the gas flow rate is the inlet gas flow rate.

3. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The detritium removal process is in any one of the following stages: surface desorption period, bulk diffusion control period, and deep trap desorption period.

4. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The prediction model is a kinetic model based on the activation energy distribution and diffusion equation. Its initial parameters are preset according to the type of plasma-oriented material and calibrated online based on the tritium removal process parameters.

5. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The determination criteria are as follows: if the real-time monitored tritium desorption rate is lower than the predicted value by 10%-20% and continues for more than the preset duration; or if the real-time monitored tritium desorption rate enters a plateau period for more than the set time; then it is determined that the current tritium removal process effect has not met expectations, and new control instructions are generated based on the current status information and the preset optimization algorithm.

6. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1 or 5, characterized in that, The optimization strategy is as follows: logical judgment is performed based on a preset rule base; and / or optimization of multiple parameters such as temperature, pressure, gas type and flow rate, and plasma discharge power is performed based on an optimization algorithm within the low-temperature constraint range.

7. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The dynamic adjustment of the tritium removal process parameters includes at least one of the following adjustment methods: adjusting the heating power through a temperature controller to stabilize the surface temperature of the plasma-facing material within a preset range; automatically adjusting the opening of the proportional valve through the closed-loop control circuit inside the gas flow controller to achieve a preset target gas flow rate; and controlling the start / stop, power, and pulse parameters of the plasma power supply through the plasma power supply controller.

8. The intelligent control method for removing tritium retention in plasma-oriented materials according to claim 1, characterized in that, The coordinated switching of the multiple methods includes automatically switching or combining at least two of the following processes during the detritium removal process: vacuum thermal desorption, isotope exchange, and plasma discharge cleaning-assisted desorption.

9. An intelligent control system for removing tritium retention in plasma-oriented materials, used to implement the intelligent control method according to any one of claims 1-8, characterized in that, include: The data sensing layer is used to monitor and collect status data during the tritium removal process in real time, forming a real-time data stream; The status data includes tritium desorption rate, temperature, pressure, and gas flow rate; The status identification module is used to identify the current stage of the detritium removal process based on the real-time data stream; the prediction module is used to receive the current status information and preset detritium removal process parameters, and predict the desorption trend curve in the future time period according to the prediction model based on the physical laws of tritium diffusion and desorption in plasma-oriented materials; the status information includes the stage of the detritium removal process and status data. The optimization decision module is used to compare the predicted desorption trend curve with the real-time monitored tritium desorption rate to obtain the deviation; if the deviation meets the preset judgment conditions, a new control command is generated based on the current state information and the preset optimization strategy; the execution control layer is used to realize the dynamic adjustment of the tritium removal process parameters and the coordinated switching of multiple methods according to the control command.

Images