166 results about "Tokamak" patented technology

According to the auxiliary supporting mechanism for assembling the internal parts of the Tokamak and the assembling method, all auxiliary stand columns and a center stand column are installed at the same center distance, and the center stand column is provided with an upper-layer disc-type turn buckle supporting mechanism and a lower-layer disc-type turn buckle supporting mechanism so as to be connected with the outer surface of a vertical section of a vacuum chamber ring petal; the outer side supporting mechanism is connected around the outer surface of the vacuum chamber ring valve set, the double-column type bottom supporting table is located below the vacuum chamber ring valve set, and the vacuum chamber ring valve set is supported from the outer side and the lower portion of the vacuum chamber ring valve set; a stand column connecting rod is connected between every two adjacent auxiliary stand columns, and a limiting mechanism capable of extending into a window in a vacuum chamber ring petal is installed on each stand column connecting rod. The U-shaped operation platform penetrates through the vacuum chamber ring petal set, and a plurality of small internal part assembling auxiliary tools are movably installed on the surface of the U-shaped operation platform. The method is wide in application range, can effectively solve the problems that transportation channels of internal parts are few and the assembly construction space is limited, and reduces the construction difficulty and the construction cost.

The invention relates to the technical field of Tokamak nuclear fusion, and discloses an assembling method of a pre-assembled assembly of a nuclear fusion device and an installing method of the nuclear fusion device. The pre-assembled assembly comprises a vacuum chamber, a window neck tube in the vacuum chamber, a vacuum chamber cold shield and a TF magnet; the assembling method of the nuclear fusion device preassembled assembly comprises the following steps: supporting between a cold shield inner sector and a vacuum chamber through a first tool; a first tool is supported between the cold shield outer sector and the vacuum chamber; the TF magnet is connected with the middle window position of the vacuum chamber through a second tool, the upper arc-shaped section of the outer fan section of the cold shield is connected with the hoisting tool of the TF magnet through a third tool, and the lower window section of the outer fan section of the cold shield is connected with the second tool; and the middle window section and the annular field magnet supporting seat are connected through a fourth tool. According to the assembling method of the nuclear fusion device pre-assembled assembly, interference among installation of the vacuum chamber, the window neck tube in the vacuum chamber, the vacuum chamber cold shield and the TF magnet can be avoided, and the stability and functionality of the pre-assembled assembly structure are guaranteed.

Main characteristic of the invention is the exploitation of the very large flux of high energy neutrons generated in a nuclear fusion tokamak with beam-driven plasma for the purpose of producing radioisotopes. It specifically relates to the deployment of precursor materials into the core of a nuclear fusion tokamak with beam-driven plasma for the purpose of exposing these materials to the neutron flux generated in the plasma by the process of nuclear fusion of Deuterium (D) and Tritium (T) to generate radioactive isotopes, so-called radioisotopes. The precursor materials contained in metallic capsules are incorporated into the blankets that surround the plasma.

The invention relates to the technical field of Tokamak plasma diagnosis and data fusion, in particular to a multi-modal fusion Tokamak high-frequency Thomson scattering reconstruction method. According to the technical scheme, the method comprises the following steps: acquiring multi-modal diagnosis data on a Tokamak device, wherein the multi-modal diagnosis data comprises high-frequency auxiliary diagnosis data and low-frequency Thomson scattering data; and performing feature engineering and physical enhancement processing on the high-frequency auxiliary diagnosis data. According to the method, through multi-modal diagnosis data fusion and interpretable KAN network modeling, the Thomson scattering time resolution is greatly improved on the premise that hardware is not transformed, and high-fidelity reconstruction of sub-millisecond transient processes such as ELM and magnetic island evolution is achieved; the method has physical interpretability, can reveal the implicit relationship between diagnostic quantities, enhances the robustness under complex experimental conditions through strategies such as modal loss, and provides key high-frequency diagnostic support for Tokamak high-performance operation and real-time control.

The invention provides a probe power supply over-current protection circuit and an over-current protection method thereof, relates to the technical field of power electronics, and solves the problems that periodic instantaneous over-current and continuous fault over-current are difficult to accurately distinguish and a control signal and a protection system lack a linkage mechanism in the prior art. In the circuit, a current detection unit generates a single over-current trigger signal when detecting a single over-current event; the central control unit sends a blocking signal after receiving the single overcurrent trigger signal, and sends a reset signal after preset time delay; meanwhile, cumulative counting is carried out on single overcurrent trigger signals, and whether an overcurrent fault occurs or not is judged according to a preset threshold number of times; the output control unit is used for controlling the closing and resetting of a probe power supply under the action of different signals; and the state feedback unit feeds back state information to the main control system when an overcurrent fault occurs. According to the invention, accurate response to the ELM overcurrent of the Tokamak device and reliable blocking of persistent faults are realized.

The invention discloses a Tokamak temperature diagnosis reconstruction method and device based on a data-driven model, and belongs to the technical field of machine learning and nuclear fusion physical crossover, and the method comprises the steps: firstly, carrying out the training of a data-driven model based on the fusion of a recurrent neural network and a self-attention mechanism through the large-scale Tokamak experiment data, and carrying out the reconstruction of the Tokamak temperature diagnosis. The complex correlation between various diagnosis signals and electronic temperature diagnosis is learned; secondly, an efficient data processing mode is provided, tokamak original signals are divided into equal-length sub-data in a window division mode, and the data are packaged and stored in an array form after being subjected to data processing so that the data can be directly used for further processing, reading, training and reasoning of subsequent machine learning; finally, the model can be seamlessly integrated into an existing Tokamak data management system by excluding a control reference signal. According to the invention, relatively accurate electronic temperature diagnosis reference can be provided.

The invention discloses a Tokamak plasma multi-parameter cooperative configuration analysis method, and relates to the technical field of plasma multi-parameter cooperative configuration analysis, and the method comprises the following steps: constructing a parameter database to store plasma parameters; performing normalization processing on the plasma parameters to obtain normalized parameters of the plasma parameters; whether correlation exists between the two plasma parameters or not is analyzed, and an independent incidence relation is obtained; analyzing whether the plasma parameters are simultaneously influenced by more than two plasma parameters, and analyzing a comprehensive association relationship between the plasma parameters; when the configuration of the plasma is regulated and controlled, plasma parameters are cooperatively regulated and controlled based on the independent incidence relation and the comprehensive incidence relation; the plasma multi-parameter cooperative configuration analysis method is used for solving the problem that better cooperative control and analysis cannot be carried out on plasmas due to the fact that correlation analysis of different plasma parameters is not comprehensive enough in an existing plasma multi-parameter cooperative configuration analysis technology.

The invention discloses a Tokamak cold shield assembling system of a liquid nitrogen rapid cooling pipeline, and relates to the technical field of Tokamak cold shield combined machining. The Tokamak cold shield assembling system of the liquid nitrogen rapid cooling pipeline comprises a stepping motor, and a rotating part of the stepping motor is connected with a first limiting plate and a second limiting plate; a rotating part of the gear motor is connected with a third limiting plate and a fourth limiting plate, and the two ends of the third limiting plate and the two ends of the fourth limiting plate are connected with placing plates; an inner containing cavity is formed between the second limiting plate and the third limiting plate, and an outer containing cavity is formed between the first limiting plate and the fourth limiting plate. According to the Tokamak cold shield assembly system of the liquid nitrogen rapid cooling pipeline, the inner cold shield petal plate and the outer cold shield petal plate are placed through the placement rotating piece and rotate to enter and exit from the installation position, the rotating piece is placed through the rotation limiting piece in a matched mode, and the inner cold shield petal plate and the outer cold shield petal plate are limited; and assembling of the inner cold screen petal plate or the outer cold screen petal plate is completed at the two ends of the vacuum chamber petal plate.

The invention discloses a multi-stage isolation high vacuum obtaining system of a neutral beam injection device and a control method, and relates to the field of nuclear fusion Tokamak vacuum pumping design. The low-temperature pump unit comprises a plug-in low-temperature pump in a neutral beam vacuum chamber, and the plug-in low-temperature pump is used for refrigerating and cooling liquid nitrogen and liquid helium to adsorb gas to realize vacuum pumping; the low-temperature pump preceding-stage air extractor unit comprises a molecular pump unit, a roots pump unit and a mechanical pump unit; an inlet of the molecular pump set is connected with the neutral beam vacuum chamber through a gate valve, and an outlet of the molecular pump set is connected with an inlet of the roots pump set through a forestage valve. An inlet of the roots pump set is connected with the neutral beam vacuum chamber through a preceding stage valve and a pre-pumping valve, and an outlet of the roots pump set is connected with the mechanical pump set through the preceding stage valve. By adopting a multi-stage vacuum shielding technology and a step-by-step start-stop strategy, step-by-step transition from low vacuum to high vacuum and from high vacuum to low vacuum can be efficiently, stably and reliably realized, and an air extractor unit and a vacuum environment are protected from being polluted.

The invention relates to the technical field of nuclear fusion, and discloses an ultrasonic molecular beam injection system for a fusion device.The ultrasonic molecular beam injection system comprises a gas source, a pressurizing unit, a storage unit and an injection unit, the pressurizing unit comprises a gas conveying pipeline, and the two ends of the gas conveying pipeline communicate with the gas source and the storage unit correspondingly; a booster pump is arranged on the gas conveying pipeline and used for pressurizing gas in the direction away from the gas source, and the storage unit can receive and store the pressurized gas so that the gas can reach a preset pressure value; the injection unit comprises an injection pipeline and a nozzle, one end of the injection pipeline is communicated with the storage unit, the other end of the injection pipeline is provided with the nozzle, the nozzle is used for stretching into a Tokamak vacuum chamber of the fusion device, and the injection pipeline can convey gas in the storage unit to the nozzle so as to form an ultrasonic molecular beam current through the nozzle. According to the invention, the injection function of the ultrasonic molecular beam in a large fusion device can be realized.

The invention provides a design method and device for a biological shielding wall of a fusion device and electronic equipment. The design method of the biological shielding wall of the fusion device comprises the steps that a biological shielding occupying model is segmented in the height direction to obtain a plurality of biological shielding occupying modules, and the biological shielding occupying modules at least comprise top shielding modules used for arranging embedded parts; the bottom shielding module is used for arranging the bottom of the Tokamak Dewar base and an annular support; based on the performance indexes of the biological shielding wall, performing structural design on each biological shielding occupation module to obtain a biological shielding engineering module; and dividing each biological shielding engineering module in the circumferential direction to obtain a plurality of prefabricated sub-modules, and obtaining the biological shielding wall formed by splicing the plurality of prefabricated sub-modules. According to the method, on-site high-altitude, high-strength and long-time complex operation is reduced, the construction safety risk is reduced, and modular removal of the biological shielding wall and waste treatment are more efficient.

The invention discloses a Tokamak plasma magnetic response model construction method, a Tokamak plasma magnetic response model construction system and a medium. Relates to the technical field of nuclear fusion plasma simulation. According to the scheme, improvement is carried out on the basis of the prior art, modeling is carried out on the magnetic response of the Tokamak plasma on the basis of basic data of the Tokamak plasma, and plasma behaviors are directly learned from experimental data; the simulator taking the plasma magnetic configuration evolution autoregression model as a carrier can process nonlinear terms and high-order terms, and is low in calculation cost, high in speed and capable of being applied in real time; according to the method, the behavior of the plasma under magnetic control can be well simulated, the plasma behavior is directly learned from experimental data without simplifying and assuming plasma properties, and the method can be used for off-line simulation and can also be used for control strategy optimization or fault detection in real time.

The invention discloses a superconducting coil, a Tokamak device applying the superconducting coil and a superconducting coil stacking process, the superconducting coil comprises a plurality of layers of sub-unit coil stacks, each layer of sub-unit coil is provided with two coil joints, and the joints of the adjacent layers are connected in a staggered manner to form a spirally rising conduction structure. The stacking process takes a leveling support platform as a reference, and coil posture adjustment is realized through a replaceable clamp: a first clamp is used for fixing a coil and detecting a contour and a joint position, and a second clamp is used for replacement after the coil is qualified; during hoisting, positioning is carried out through a characteristic line alignment technology, the profile is adjusted through a shape maintaining tool or the angle is finely adjusted through a circumferential adjusting device, and accurate alignment between layers is ensured; the circumferential adjusting device corrects the angle deviation through the telescopic column. According to the scheme, through multi-stage precision control, assembly precision control over stacking of the superconducting coils is achieved, the problems of interlayer dislocation and poor conduction of the large-size coils are solved, and the magnetic field uniformity and operation stability of the Tokamak device are improved.

The invention discloses a cooperative control method for improving Tokamak radio frequency wave current driving locality, and the method mainly comprises the following steps: S1, according to a radial position where locality needs to be enhanced, heating narrow resonance layer electrons at the corresponding radial position through electron cyclotron resonance heating (ECRH); s2, ion cyclotron resonance heating (ICRH) is adjusted for regional heating, so that the radial position of an ICRH driving current main peak value coincides with the radial position of an ECRH current peak value; s3, using GENRAY and CQL3D programs to simulate low hybrid wave (LHW) and high harmonic fast wave (HHFW) combined current driving, calculating a momentum region corresponding to a synergistic effect in a momentum space, performing correlation analysis on the synergistic momentum region, wave emission parameters and current radial distribution, feeding back the parameter adjustment direction, and performing repeated iteration until required local current distribution is achieved; according to the invention, current driving with consideration of driving efficiency and locality control is realized, and the method is of great significance in enhancing the current distribution control capability, improving the constraint performance of plasma and obtaining longer constraint time and higher fusion power.

The invention discloses a Tokamak plasma rupture prediction method based on cWGAN and TCN, which takes AUC as a measurement index to evaluate the effectiveness of a model, and adopts a conditional Wasserstein generative adversarial network (cWGAN)-based minority class sample generation method for solving the problem that the number of minority class samples is limited. Gradient penalty (GP) is introduced in a training stage to improve the quality of generated samples and training stability, and an inverse frequency mini-batch minority class equilibrium training strategy is adopted to enable a generator to fully learn distribution characteristics of broken samples, so that high-quality minority class samples are generated and a training set is expanded. A Transform structure is introduced on the basis of a time convolutional network (TCN), local time sequence features are efficiently extracted by using the TCN, and the Transform enhances the expression ability of a model for a long-time dependency relationship and global features, so that the prediction effect is improved.

The invention provides an assembly construction method of a biological shielding wall of a fusion device and the biological shielding wall. The assembly construction method of the biological shielding wall of the fusion device comprises the following steps: manufacturing a wall body framework of a prefabricated sub-module; the wall body frameworks of the prefabricated submodules are assembled in the circumferential direction to obtain an annular framework of the biological shielding wall module, and the biological shielding wall module at least comprises a top shielding module used for arranging embedded parts and a bottom shielding module used for arranging the bottom of a Tokamak Dewar base and an annular support; and the annular framework is hoisted to a pouring foundation, and the interior of the annular framework is poured after hoisting is completed every time. According to the method, on-site high-altitude, high-strength and long-time complex operation is reduced, and construction safety risks are reduced. In addition, if internal parts need to be replaced or overhauled in the future, corresponding structures on the specific prefabricated submodules can be directionally dismantled, and when the service life of the fusion device is over, modular dismantling of the biological shielding wall and waste treatment are more efficient.

The invention relates to the design, configuration and operation concept of a system, herein called radioisotope recovery system, to deploy and recover different precursor materials into and from the core of a nuclear fusion tokamak with beam-driven plasma for the purpose of exposing these materials to the neutron flux generated in the plasma by the process of nuclear fusion of D and T ions, which will cause the generation of radioactive isotopes (also called radioisotopes). Main characteristics of the invention are (i) a closed loop that circulates precursor materials into and out of a nuclear fusion tokamak with beam-driven plasma, (ii) the compliance of the system with the design and maintenance of a nuclear fusion tokamak with beam-driven plasma.

The present invention discloses a method, device and medium for constructing a compact copper conductor tokamak, which relates to the technical field of index evaluation. By obtaining the evaluation indexes of the tokamak device, a quantitative evaluation system is constructed; based on the device operation conditions, the parameter thresholds of the evaluation parameters in the quantitative evaluation system are determined; based on the parameter thresholds, the major radius and minor radius thresholds of the compact device are determined; according to the major radius threshold and minor radius threshold, the boundary of the plasma is determined; and based on the plasma boundary, the compact device is constructed. By establishing a quantitative index system for the comprehensive engineering technology capabilities of the tokamak, the ranges of core parameters such as the major radius and minor radius of the plasma can be quickly determined, the dimension chain can be determined more accurately, the conflict between physical requirements and engineering implementation caused by large dimension chain errors can be reduced, and the engineering capabilities can meet the standards when the device is miniaturized and compactly designed.

The invention discloses a cantilever beam vibration absorber and a Tokamak device, and belongs to the field of vibration control. The cantilever beam vibration absorber comprises an acoustic black hole cantilever beam which comprises an acoustic black hole structure; the piezoelectric element is located on any surface of the acoustic black hole structure in the thickness direction; the piezoelectric shunt circuit is electrically connected with the piezoelectric element; the damping layer is located on at least one side of the acoustic black hole structure in the thickness direction; the tuned mass block is located on the side, away from the acoustic black hole structure, of the damping layer. Broadband vibration reduction can be achieved, the vibration reduction frequency is flexible, accurate and adjustable, and the vibration reduction effect is effectively improved.

This invention discloses a tokamak plasma turbulence prediction method and system based on layer-theoretic topology and spectral neural operators. The method includes acquiring diagnostic and control data from a tokamak device and converting them into a set of physical field vectors, actuator vectors, and equilibrium parameter vectors; mapping the physical field vector set to a latent grid representation and fusing the actuator and equilibrium parameter vectors to obtain a conditional latent grid representation; using a multi-layer Fourier neural operator for time evolution to obtain the latent grid representation for the next time step; applying multi-region consistency constraints to the latent grid representation using a constraint mapping operator and a glue loss function; and using the constrained predicted latent grid representation for prediction, outputting the hazard level and remaining time prediction values ​​of plasma turbulence. Simultaneously, a projection and repair module applies physical constraints to the restored physical field prediction representation to obtain a corrected physical field prediction representation. This invention improves the accuracy of the results.

The present invention relates to a method for monitoring deposition impurities on the first mirror of a tokamak based on a convolutional neural network. The method uses a spectrometer to real-time collect plasma spectral data of the sputtered and ionized surface material of the first mirror in a tokamak under low-pressure radio frequency conditions in a fusion device. According to the characteristics of plasma spectral data with different wavelengths and intensities corresponding to different sputtered materials on the first mirror surface, a convolutional neural network model is established and trained to be able to identify the spectral data of the composition and content of the sputtered materials on the first mirror surface. The spectral data obtained by plasma sputtering is input into a mature network model for analysis to judge the composition and content of the sputtered materials, so as to monitor the change of deposition impurities on the first mirror surface. The present invention obtains the radio frequency plasma spectral information generated on the first mirror surface through a spectrometer, and can identify the composition and content of the sputtered materials on the first mirror surface through a convolutional neural network model without damage, effectively and without affecting the experiment.

The present invention discloses a device and method for steady-state testing of a tokamak internal cold screen, including a test box and an internal cold screen arranged in the test box, wherein a movable base is arranged at the bottom of the inner cavity of the test box, and the present invention relates to the field of internal cold screen detection technology. The device and method for steady-state testing of a tokamak internal cold screen supports the internal cold screen through a receiving component, and under the setting of the movable base, the internal cold screen is transferred to the test box, and the internal cold screen is precisely positioned by cooperating with the steering drive component and the positioning component, and the end face plugging component is arranged to plug the open end face of the internal cold screen, so that vacuum helium leak detection of the internal cold screen can be performed, and the operation is simple and convenient, and efficient vacuum helium leak detection of the internal cold screen can be realized, and after the test is completed, the internal cold screen can also be stably transferred to the outside of the test box, and the internal cold screen has good adaptability and efficient and stable detection of the internal cold screen can be realized.

The invention relates to the field of cooling systems, in particular to a cooling system for a Tokamak device and a control method.The cooling system comprises a cooling device, a first heat exchange module, a device to be subjected to heat exchange and a water quality treatment module, and the cooling device is suitable for cooling a refrigerant medium flowing through the cooling device; the first heat exchange module is internally provided with a first heat exchange flow channel and a second heat exchange flow channel which can selectively exchange heat with each other, the first heat exchange flow channel and the second heat exchange flow channel are suitable for circulating a refrigerant medium, the first heat exchange flow channel is communicated with the cooling device, the device to be subjected to heat exchange is suitable for being arranged in the Tokamak device, and the device to be subjected to heat exchange is selectively communicated with the second heat exchange flow channel; the water quality treatment module selectively communicates with the second heat exchange flow channel, the water quality treatment module and the second heat exchange flow channel form a treatment loop, the water quality treatment module is suitable for optimizing the water quality of the refrigerant medium flowing through the water quality treatment module, and the overall operation stability of the cooling system can be improved.

The present invention relates to the technical field of nuclear fusion tokamak control systems, and in particular to a plasma control system health assessment method based on a decision table and an analytic hierarchy process. The technical solution comprises the following steps: assessing the plasma control system health status based on a decision table; assessing the plasma control system health status based on an analytic hierarchy process; and combining a decision table-based system health assessment method with an analytic hierarchy process-based health assessment method to comprehensively assess the plasma control system health status. By combining the advantages of decision tables and the analytic hierarchy process, the present invention achieves real-time health assessment and precise fault diagnosis for complex distributed control systems, significantly improving the stability, reliability, and safety of the plasma control system and providing strong support for the efficient operation of the system.

The invention discloses an auxiliary coil positioning device, and belongs to the technical field of nuclear power assembly.The auxiliary coil positioning device is used for assembling a coil and a wire box in a Tokamak device, an assembling hanger of the auxiliary coil positioning device comprises a connecting assembly and a guiding and limiting assembly, and the connecting assembly comprises a hanging part and a mounting part which are connected; the mounting part is detachably hung on the outer edge of the wire groove, the mounting part is attached to the inner side wall of one side of the wire groove, the guiding and limiting assembly comprises a guiding part, a connecting part and a limiting part, the connecting part is mounted on the mounting part, and the guiding part and the limiting part are both connected with the mounting part. According to the coil auxiliary positioning device, the size of the limiting piece can be correspondingly adjusted according to the size error of the contact position of the coil and the guiding piece, so that when the coil has the size error, the assembling precision of the coil and the wire box is improved, the magnetic field precision of a Tokamak device is improved, and stable constraint of the Tokamak device on plasma is guaranteed.

The invention discloses a CuCrZrTa alloy wire with a nanometer oxide dispersion strengthening phase and a preparation method of the CuCrZrTa alloy wire. The CuCrZrTa alloy wire comprises, by weight, 0.6 wt%-1.0 wt% of Cr, 0.08 wt%-0.12 wt% of Zr, 0.03 wt%-0.08 wt% of Ta, 0.2 wt%-0.4 wt% of Y2O3, 0.005 wt%-0.02 wt% of B and the balance Cu, and the phase structure of the wire comprises a nanometer Cr precipitation phase, YTaO dispersion particles and a B segregation grain boundary. Through Ta-Y2O3 interlocking to inhibit coarsening, B grain boundary segregation to resist irradiation, low-temperature sintering is adopted to retain nanocrystals, and two-stage aging is adopted to accurately regulate and control precipitated phase distribution, so that the prepared wire has the characteristics of high strength, high conductivity and excellent irradiation resistance of a copper alloy wire; and the extreme working condition requirements of a Tokamak magnet winding (requiring high conductivity / radiation resistance) and a divertor cooling pipeline (requiring high temperature resistance / creep resistance) are met.

The application discloses a boundary plasma simulation method under a unit magnetic surface coordinate system and belongs to the technical field of magnetic confinement tokamak boundary plasma simulation. Firstly, a physical model is expressed by using a unit magnetic surface coordinate system, physical quantity and geometric effect are separated by using unitized base vectors, and singularity near an X point is avoided; secondly, a toroidal Fourier series representation is combined with difference along a magnetic field line, problem dimension is reduced, and numerical dissipation is reduced; thirdly, a generalized finite difference method and a transformed coordinate system are adopted to process the X point and adjacent points, and the X point can be directly contained in a grid; finally, a grid screening method is used to pre-calculate geometric quantities and difference coefficients under high-density grids, and the calculation precision of geometric effect is improved. The application can accurately simulate the influence of the X point on the stability of the boundary plasma, significantly reduces the consumption of computing resources, and improves the simulation efficiency and accuracy.

The application discloses a kind of nuclear fusion adaptive control system and method based on reinforcement learning, the control system includes physical layer and control layer, physical layer contains tokamak / stellarator device, Thomson scattering and so on diagnostic system and FPAD / PAM projectile injection system, responsible for collecting plasma density profile, temperature profile and other data and executing fuel supply;Control layer is reinforcement learning intelligent agent, contains state perception, strategy network and execution interface module, strategy network uses PPO / SAC algorithm, state perception is included line average density, density change rate and historical sequence data;The control method is trained offline by Torax simulator, is optimized by composite reward function, forms perception-decision-making-execution-feedback closed-loop control, supports real machine deployment and online fine tuning.The application solves the problem of traditional PID control lag, adaptive difference, the control precision is improved by about 40%, density fluctuation is reduced by about 50%, prolongs hardware maintenance cycle, adapts nuclear fusion long pulse operation demand.

The application discloses a kind of plasma posture control agent strategy model construction method, device and medium, by constructing tokamak simulation environment, the environmental configuration parameter of out-of-control scene under the control of PID controller is identified, and the training environment of configuration is obtained;Using reinforcement learning agent model to learn, generate current command;The state of control point at each time is calculated based on tokamak simulation environment;The action command obtained by the control point state at each time, current command and PID controller are input into reward function, and the environment reward is calculated;According to the environment reward, the control point state at each time and current command, reinforcement learning training is carried out, and the command strategy is updated;When reinforcement learning agent model meets convergence condition, output agent strategy model.The application scheme provides a kind of strategy model in nonlinear and complex dynamic environment to deal with the ability of large disturbance of transient state, guarantees the stability and safety of tokamak device.

A modular tokamak is disclosed. The modular tokamak according to the present invention may comprise a plurality of core chambers sequentially connected in the longitudinal direction and may also comprise extension chambers each connecting two adjacent core chambers.