Quasi-pulse frc deuterium-deuterium fusion reaction device and nuclear fuel self-sustaining cycle steady power generation method thereof

Abstract

A quasi-pulsed FRC deuterium-deuterium fusion reactor and its self-sustaining nuclear fuel cycle steady-state power generation method are disclosed. The reactor comprises a linear vacuum chamber, an FRC plasma formation zone, a collision fusion zone, a conical transition section, a plasma formation coil, a fast compression coil, a magnetic mirror coil assembly, a deuterium-tritium feeding system, a zoned vacuum pumping system, and a pulsed power control system. The magnetic mirror coil assembly is located near the collision fusion zone in the conical transition section. Within a single pulse cycle, it sequentially executes four functional modes: compression assistance, magnetic mirror reflection, magnetic bottle confinement, and recovery attenuation. During the recovery attenuation mode, it is reused as a direct energy conversion induction coil. The operation mode employs a trace amount of tritium to catalyze deuterium-deuterium fusion. Tritium and helium-3 are confined to participate in the secondary reaction through magnetic mirror confinement. Between pulses, the difference in chemical adsorption of tritium and the difference in cryogenic condensation of helium-3 are utilized to achieve selective recovery of the two components, forming a quasi-pulsed self-sustaining operation. The power is then converted into steady-state electrical power and connected to the grid via a rectification-storage-inverter system, realizing the internal recycling of tritium and helium-3.

Description

Technical Field

[0001] This invention belongs to the field of magnetic inertial confinement nuclear fusion energy technology, specifically involving a deuterium-deuterium (DD) fusion reactor device based on field-reversed configuration (FRC), and an operation method that utilizes the time-sequential multiplexing of magnetic mirror coil groups to achieve magnetic confinement recovery, selective separation, nuclear fuel self-sustaining cycle, direct energy conversion (DEC), and steady-state grid connection of pulsed power for the fusion product tritium (T) and helium-3 (³He). Background Technology

[0002] Controlled nuclear fusion is considered humanity's ultimate energy solution. However, current mainstream magnetic confinement fusion technologies (such as tokamas and stellarators) and most commercial FRC projects primarily use deuterium-tritium (DT) or deuterium-helium-3 (D-³He) fuels: tritium is radioactive and extremely scarce; helium-3 has very low natural reserves on Earth, making it difficult to support commercial energy needs. The resource bottlenecks and radioactive safety issues of these fuel routes limit their universality as the ultimate energy source.

[0003] Deuterium (D) can be extracted from seawater in near unlimited quantities through heavy water separation. Deuterium-deuterium (DD) fusion does not depend on the input of external radioactive fuel, is not limited by the scarcity of helium-3, has an inexhaustible fuel source, and is environmentally friendly and controllable. Only the deuterium-deuterium route has the resource foundation to truly serve as the ultimate energy source for mankind.

[0004] In the field of linear FRC fusion devices, Helion Energy's initial technology uses deuterium-tritium (DT) fuel, with plans to later switch to deuterium-helium-3 (D-³He) fuel. Their device generates fusion energy through collisional compression and employs a direct induction energy recovery method: as the fusion plasma expands, it drives a magnetic field, inducing a pulse current in a coil according to Faraday's law, recovering energy to a capacitor bank for the next pulse. This energy recovery scheme involves the capacitive storage and release of pulsed power within the device to maintain the pulsed power supply system; it does not involve the magnetic confinement and recycling of tritium and helium-3 from the fusion products, nor does it involve a power conversion system to convert pulsed power into steady-state AC power for grid baseload supply.

[0005] On the other hand, the FRC two-stage cascaded magnetic compression technology proposed by Academician Pan Yuan's team aims at a pure deuterium-deuterium fusion neutron source. It directly compresses deuterium-deuterium plasma to fusion conditions through a cascaded combination of fast and slow compression. This technology is highly complex, requiring the retention of a second-stage active slow compression coil and a supporting high-power power supply system. Directly compressing deuterium to achieve fusion is technically extremely challenging, and it does not involve the internal recycling of fusion products tritium and helium-3.

[0006] Existing FRC fusion technologies generally suffer from the following technical bottlenecks: (1) The ignition temperature requirement for the deuterium-deuterium reaction is extremely high (about 300 million degrees), which is difficult to achieve economically by relying solely on magnetic compression; (2) Tritium is an important byproduct and potential igniter of deuterium-deuterium fusion, but existing technologies have not achieved closed-loop recovery within the device; Helium-3 is another byproduct with high fuel value, but existing technologies have not included it in the recovery system. (3) In FRC pulse operation mode, the charged particles of fusion products (tritium nuclei, helium-3 nuclei, protons) are rapidly diffused and lost after collision, and do not participate in the secondary reaction to amplify energy output; (4) The existing FRC unit operates in a single-pulse "open-loop" mode, with fuel products being discharged and then refilled, resulting in low average power density and a lack of energy relay between pulses. (5) Existing FRC devices lack a conversion scheme from pulsed fusion power to steady-state power frequency AC grid connection, resulting in an incomplete industrial closed loop. It is particularly important to note that FRC fusion power is released in microsecond-level pulses with repetitive frequencies from Hertz to 100 Hertz, which is completely different from the continuous and smooth power of photovoltaic and wind power. Conventional rectifier-energy storage-inverter grid connection systems are designed for continuous power and cannot be directly applied to the steep rise and fall characteristics of pulsed power. There are multiple technical obstacles, such as synchronizing the pulse frequency with the grid's 50Hz / 60Hz, smoothing the microsecond-level peak power with the continuous base load, and matching the recovery decay sequence with the inverter switching sequence. (6) The DEC induction coil and magnetic confinement coil of the existing FRC device are arranged independently, which is redundant and has low axial space utilization. (7) Existing FRC devices lack a pulse-to-pulse dual-component selective recovery mechanism, which cannot simultaneously address the removal of residual impurities and the retention of tritium and helium-3. Summary of the Invention

[0007] 1. Technical problems to be solved The present invention aims to provide a quasi-pulse FRC deuterium-deuterium fusion reactor and its operation method to solve the following problems existing in the prior art: (1) The existing FRC device has too high an ignition threshold for direct compression of deuterium, making it difficult to implement in engineering. (2) Existing FRC devices have not achieved effective recovery and recycling of the deuterium-deuterium fusion products tritium and helium-3, resulting in the waste of scarce strategic resources and external dependence; (3) The existing FRC device operates in a single-pulse "open-loop" mode, resulting in large losses of charged particles from fusion products escaping, and a lack of energy relay between the preceding and following pulses. (4) Existing FRC devices lack a conversion scheme from pulsed fusion power to steady-state AC grid connection, resulting in an incomplete industrial closed loop. In particular, conventional grid-connected systems cannot adapt to the pulsed power characteristics of FRC, presenting technical obstacles to pulse-to-steady-state conversion; (5) The DEC induction coil and magnetic confinement coil of the existing FRC device are arranged independently, resulting in redundant device structure; (6) Existing FRC devices lack a pulse-to-pulse bicomponent selective recovery mechanism, which cannot simultaneously address the removal of residual impurities and the retention of tritium and helium-3. 2. Technical Solution

[0008] This invention provides a quasi-pulse FRC deuterium-deuterium fusion reactor, characterized in that it comprises: A linear vacuum chamber has FRC plasma formation regions symmetrically arranged at both ends and a straight cylindrical collision fusion region in the middle. The formation region and the collision fusion region are connected by a conical transition section. The diameter of the formation region is larger than the diameter of the collision fusion region. An FRC plasma forming coil is arranged inside the forming region to pre-ionize the injected deuterium gas and establish a reverse magnetic field to form FRC plasma. A fast compression coil, arranged outside the formation region, is used to generate a pulsed magnetic field to accelerate the FRC plasma axially toward the collision fusion region; The magnetic mirror coil assembly is symmetrically arranged at the small-diameter end of the conical transition section near the collision fusion zone. The coil frame of the magnetic mirror coil assembly is conical, and its inner diameter gradually changes from the outer diameter of the forming zone to the outer diameter of the collision fusion zone along the axial direction and closely fits the outer wall of the conical transition section. The magnetic mirror coil assembly sequentially executes the following four functional modes within a single pulse cycle: • Compression-assisted mode: In the fast compression phase, it works in conjunction with the fast compression coil to form a magnetic field gradient pointing towards the collision fusion region, assisting in the acceleration of FRC plasma; • Magnetic mirror reflection mode: A peak magnetic field intensity is formed during the collision and fusion stage, reflecting high-energy charged fusion products; • Magnetic bottle confinement mode: Maintains a weak magnetic field confinement region between two magnetic field peaks, prolonging the residence time of tritium nuclei, helium-3 nuclei, and protons in the fusion region; • Recovery attenuation mode: The magnetic field attenuates controllably according to a preset time sequence, guiding the confined enriched plasma to the formation region. At the same time, the magnetic mirror coil group is reused as a direct energy conversion induction coil, and the plasma expansion drives the change of magnetic flux to induce a pulse current in the magnetic mirror coil group. An auxiliary stabilizing coil is arranged close to the periphery of the collision fusion zone to provide a continuous or quasi-steady-state weak background magnetic field, forming a magnetic cage to dampen the violent movement of the plasma after the collision and to confine the charged fusion products to reduce wall losses. A pulse power supply control system is used to provide timing pulse current to the magnetic mirror coil group to realize the switching of the four functional modes, and includes a rectifier module for converting the pulse current generated by the direct energy conversion induction coil into direct current. The deuterium main fuel feeder and the tritium igniter injector are symmetrically arranged at the ends of the two forming areas. The deuterium main fuel feeder is used to replenish the main fuel deuterium between pulses, and the tritium igniter injector is used to quantitatively inject a trace amount of tritium igniter. A temporary tritium adsorption trap is arranged at the end of the formation region for adsorbing and recovering tritium between pulses and releasing it by heating before the next pulse; A partitioned vacuum pumping system is arranged on the side wall of the conical transition section. The partitioned vacuum pumping system includes a cryogenic condensation and recovery branch. The cryogenic condensation and recovery branch is equipped with a cryogenic condensation trap for liquefying and separating the helium-3 that diffuses into the conical transition section between pulses in the cryogenic condensation trap for recovery. The pumping port of the partitioned vacuum pumping system is far away from the formation zone to avoid pumping out the tritium recovered to the formation zone.

[0009] The present invention also provides a method for steady-state power generation of nuclear fuel self-sustaining cycle based on the above-mentioned fusion reactor, comprising the following steps: First pulse operation: Deuterium, the main fuel, and a small amount of tritium, the igniter, are injected into the two-end formation regions. The FRC plasma formation coil is activated to generate pre-ionized plasma and establish a reverse magnetic field. Then, the fast compression coil is activated to drive the two FRC plasmas to collide and compress in the collision fusion region. The deuterium-deuterium fusion reaction produces tritium, helium-3, protons, and neutrons. Combustion confinement stage: Switch the magnetic mirror coil group to magnetic mirror reflection mode and magnetic bottle confinement mode to confine the charged particles in the fusion products to the collision fusion region. Utilize the difference in cyclotron frequency between tritium nuclei and helium-3 nuclei to make them stay together in the weak field region of the magnetic bottle and undergo secondary reactions with the background deuterium plasma to amplify the energy output. Inter-pulse recovery phase: Switch the magnetic mirror coil group to recovery attenuation mode. The magnetic field attenuation forms a weak gradient pointing towards the formation region, guiding the plasma enriched with tritium, helium-3 and unreacted deuterium to the formation region. Two-component selective recovery stage: Start the zoned vacuum pumping system to remove residual protons and impurity gases that have diffused into the conical transition section; at the same time, use the tritium temporary adsorption trap to selectively adsorb tritium at room temperature to 100°C, and use the low temperature condenser trap to liquefy and separate helium-3 at below 4K, retaining tritium and helium-3 in the formation zone. Direct energy conversion stage: During the recovery decay mode, the magnetic mirror coil group is reused as a direct energy conversion induction coil, and the induced current generated by the change in magnetic flux driven by plasma expansion is converted into direct current by the rectifier module; Quasi-pulse self-sustaining operation: The above pulse process is repeated. In subsequent pulses, deuterium is replenished through the deuterium main fuel feeder. The tritium igniter injector dynamically adjusts the external tritium injection amount according to the recovery amount. The tritium catalytic cycle is maintained by the recovered products, forming a quasi-pulse continuous operation. The tritium and helium-3 recovered in the previous pulse are used as igniters and energy amplification media in the subsequent pulses to participate in the secondary reaction. Steady-state power generation stage: The pulse power and direct energy released by fusion are collected and converted into induced electrical energy. The rectifier converts the induced electrical energy into DC power, the energy storage and smoothing array absorbs the pulse fluctuations, and then the grid-connected inverter converts the DC power into AC power for output to the grid. The capacity and power density of the energy storage and smoothing array are designed to match the pulse repetition frequency of the fusion reaction device (1-100Hz) and the peak pulse power, so as to convert the pulse power with pulse widths of Hertz to 100Hz and microseconds into a smooth DC bus. 3. Beneficial effects

[0010] Compared with the prior art, the present invention has the following beneficial effects: (1) Lowering the ignition threshold and simplifying the engineering structure: This invention adopts a single-stage fast compression combined with the time-sequential multiplexing of the magnetic mirror coil group, eliminating the need for a cascaded magnetic compression structure. By using a trace amount of tritium igniter to catalyze deuterium-deuterium fusion, the ignition temperature requirement is reduced from the extreme condition of direct ignition of pure deuterium-deuterium to a more easily achievable level, significantly simplifying the power system and device structure and improving engineering feasibility.

[0011] (2) Achieving dual-component magnetic confinement recovery and nuclear fuel self-sufficiency of tritium and helium-3: Utilizing the time-sequential multiplexing of the same set of magnetic mirror coils, charged fusion products are confined during the combustion phase, and tritium and helium-3 are recovered together into the formation zone during pulse intervals. Tritium is retained through a chemisorption trap, while helium-3 is liquefied and separated through a cryogenic condensation trap, resulting in a pulse-by-pulse reduction in the external tritium injection amount and internal circulation of helium-3. This scheme reduces annual tritium consumption to the gram level while fully utilizing the high-value clean fuel properties of helium-3.

[0012] (3) Improved average power density: Unlike the existing FRC single-pulse "open-loop" operation (recharging after combustion products are discharged), the quasi-pulse operation of this invention enables charged fusion products to participate in secondary reactions (D+T→) through inter-pulse product recovery and multi-pulse energy superposition. 4 He+n,D+³He→ 4 He+p) amplifies energy, significantly improving the average power density of the device.

[0013] (4) Realizing an industrial closed loop from pulsed fusion to steady-state power generation, and solving the technical obstacles of pulsed power grid connection: Existing FRC technologies (such as Helion) use direct induction recovery to recover pulse energy to capacitor banks for release in the next pulse inside the device. This is an internal energy cycle at the experimental device level, physically isolated from the external power grid. This invention uses magnetic mirror coil groups multiplexed into direct energy conversion induction coils to convert the magnetic energy changes of plasma expansion into electrical energy. This electrical energy is then converted into steady-state power frequency AC through a rectification-energy storage-inverter link tightly coupled with the FRC pulse characteristics, enabling the FRC pulsed fusion reactor to be connected to the power grid as a base load power source. The capacity and power density of the energy storage smoothing array are specially matched according to the pulse repetition frequency (1-100Hz) and microsecond-level pulse width, solving the technical obstacle that conventional grid-connected systems cannot adapt to the steep rise and fall characteristics of pulse power, forming a complete industrial closed loop from plasma physics to power systems.

[0014] (5) Enhance plasma stability and safety: The auxiliary stabilizing coils around the collision fusion zone provide a continuous weak magnetic cage to dampen the violent movement of the plasma after the collision, reduce the bombardment of high-energy charged particles on the vacuum chamber wall, and reduce the risk of wall material activation, tritium retention and structural damage.

[0015] (6) The structure is the simplest and the axial space is compact: the magnetic mirror coil group is reused as a direct energy conversion induction coil, eliminating the need for a separate DEC coil and reducing one set of coils and its power supply system; the magnetic mirror coil group is arranged close to the outer wall of the conical transition section, making full use of the conical space, improving the axial space utilization of the vacuum chamber and reducing the manufacturing cost of the device.

[0016] (7) Balancing impurity removal with tritium and helium-3 retention: The partitioned vacuum pumping system is arranged on the side wall of the conical transition section, away from the formation zone, and selectively removes residual protons and impurity gases between pulses. At the same time, the contradiction between tritium and helium-3 self-sustaining operation and residual gas removal is resolved through the dual-component recovery mechanism of tritium adsorption trap and helium-3 condensation trap. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall structure of the quasi-pulse FRC deuterium-deuterium fusion reactor of the present invention; Figure 2 The timing diagram shows the four functional modes of the magnetic mirror coil group within a single pulse cycle. Figure 3 This is a schematic diagram of the magnetic field topology and the constrained trajectories of charged particles in the collision fusion region; Figure 4 A pulse sequence list and a tritium injection rate decrease curve for self-sustaining operation of tritium-helium-3 dual-component recovery; Figure 5 This is a system block diagram showing how pulsed power is converted into steady-state electrical power via a rectifier-energy storage-inverter system tightly coupled with FRC pulse characteristics. Figure 6 The time-series relationship between inter-pulse zone pumping and two-component recovery is shown in the diagram. Figure 7 Detailed diagram of the arrangement of the magnetic mirror coil assembly in the conical transition section and magnetic field distribution diagram; Figure 8 This is a flow chart of material balance for the recovery of tritium-helium-3 dual components. Detailed Implementation

[0018] Overall technical concept

[0019] This invention employs a technical architecture of "single-stage fast compression + magnetic mirror time-series multiplexing + dual-component selective recovery," which differs from existing two-stage cascaded magnetic compression routes and single-pulse open-loop operation routes. The device achieves magnetic confinement recovery and inter-pulse recycling of fusion products tritium and helium-3 through four-mode switching of the magnetic mirror coil group within a microsecond to millisecond time series; it retains tritium and helium-3 by selectively removing residual impurities between pulses through a zoned vacuum pumping system, forming a quasi-pulse self-sustaining operation; and it converts pulsed fusion power into steady-state power frequency AC by multiplexing the magnetic mirror coil group into direct energy conversion induction coils, achieving baseload grid connection.

[0020] The specific implementation methods are described below in three progressive levels: proof-of-concept device, engineering verification device, and steady-state power generation system. Example 1: Proof-of-Concept Device (Desktop)

[0021] like Figure 1 The quasi-pulse FRC deuterium-deuterium fusion reactor of this embodiment includes: the main body materials of the vacuum chamber 1, the conical transition section 4 and the forming region 2 are selected from one or more combinations of non-magnetic austenitic stainless steel, quartz glass, borosilicate glass or ceramic glass; when glass or ceramic materials are used, the magnetic mirror coil group 6 is attached to the outer wall of the vacuum chamber through an external support sleeve or a segmented clamp structure, and an insulating and heat-conducting layer is provided between the support sleeve and the vacuum chamber wall.

[0022] A linear vacuum chamber 1, with a total length of 1.5-2.5 meters, has FRC plasma formation regions 2 (inner diameter 15-25 cm, length 30-50 cm) symmetrically arranged at both ends, and a straight cylindrical collision fusion region 3 (inner diameter 5-10 cm, length 10-20 cm) in the middle. The formation region 2 and the collision fusion region 3 are connected by a conical transition section 4, with a cone angle of 15°-30°. The diameter of the formation region 2 is larger than the diameter of the collision fusion region 3.

[0023] The FRC plasma forming coil 5a is located on the inner side of the outer wall of the forming region 2, close to the outer wall of the vacuum chamber. It is a θ-pinch type or coaxial gun type coil with an inductance of 0.1-1μH. It is powered by a low-voltage, high-current pulse power supply (voltage 1-5kV, current 10-100kA) with a pulse width of 10-100μs. During operation, deuterium gas is first introduced into the forming region 2 to start the forming coil 5a to generate pre-ionized plasma. A reverse magnetic field is then established by a rapid reverse current to form the FRC topology (plasma current loop + internal reverse field + external confinement field).

[0024] Fast compression coil 5, located on the outer side of the outer wall of formation region 2, is coaxial with 5a. Its inner diameter is larger than that of 5a. It is a θ-pinch type coil with an inductance of 1-5μH. It is powered by a high-voltage pulse capacitor bank (energy storage 10-100kJ, voltage 10-50kV), with a pulse rise time of 1-5μs and a peak current of 100-500kA. It is used to accelerate the FRC plasma axially to 10 6 -10 7 On the order of cm / s, it pushes towards the collision fusion region 3.

[0025] Magnetic mirror coil groups 6 are symmetrically arranged at the small-diameter end (throat) of the conical transition section 4 near the collision fusion zone 3. The coil frame of the magnetic mirror coil group 6 is conical, with its inner diameter gradually changing axially from the outer diameter of the forming zone 2 (e.g., 20cm) to the outer diameter of the collision fusion zone 3 (e.g., 12cm), closely fitting the outer wall of the conical transition section 4 with a gap of less than 5mm. Each magnetic mirror coil group 6 is a low-inductance hollow coil (inductance 0.5-2μH, number of turns 20-100), and is supplied with sequential pulse current by the pulse power supply control system 7.

[0026] • The magnetic mirror coil group 6 sequentially executes the following four functional modes within a single pulse cycle (total duration approximately 100μs-10ms): • Compression-assisted mode (T0-T1, approximately 1-5 μs): The current in the magnetic mirror coil group 6 rises synchronously with the fast compression coil 5, forming a magnetic field gradient pointing towards the collision fusion region 3 (the magnetic field strength increases from 0.1-0.5 T at the entrance of the formation region to 1-2 T at the entrance of the fusion region), which assists in the acceleration of the FRC plasma. • Magnetic mirror reflection mode (T1-T2, approximately 5-20 μs): The current in magnetic mirror coil group 6 reaches its peak (5-20 T), forming magnetic field strength peaks on both sides of the collision fusion region 3 (magnetic mirror points M1 and M2, located at the junction of the conical transition section and the fusion region). High-energy charged fusion products (tritium nucleus kinetic energy approximately 1.01 MeV, helium-3 nucleus kinetic energy approximately 0.82 MeV, proton kinetic energy approximately 3.02 MeV) are reflected at the magnetic mirror points. Due to the different charge states of tritium nuclei (³H⁺, Z=1) and helium-3 nuclei (³He²⁺, Z=2), their cyclotron frequency ratio is 1:2 and their cyclotron radius ratio is 2:1 under the same magnetic field. Both remain together in the weak field region of the magnetic bottle and undergo secondary reactions with the background deuterium. • Magnetic bottle confinement mode (T2-T3, approximately 20-100 μs): The current in magnetic mirror coil group 6 maintains a plateau or decays slowly, forming a weak magnetic field confinement region (magnetic bottle, with a central field strength of 1 / 3-1 / 2 of the peak field) between the two magnetic field peaks. This prolongs the residence time of charged particles in the fusion region, allowing them to undergo secondary fusion reactions with the background deuterium plasma (D+T→). 4 He+n,D+³He→ 4 He+p), amplifying energy output; • Recovery decay mode (T3-T4, approximately 0.5-5ms): The current of the magnetic mirror coil group 6 decays controllably according to a preset exponential or linear curve. The magnetic field "valve" opens, guiding the confined enriched plasma (containing tritium, helium-3 and unreacted deuterium) to the formation region 2.

[0027] During the decay mode, the magnetic mirror coil group 6 is simultaneously reused as a direct energy conversion induction coil: plasma expansion drives changes in residual magnetic flux, inducing a pulse current in the magnetic mirror coil group 6 according to Faraday's law. Assuming the magnetic mirror coil group 6 has N=50 turns, an effective area per turn S=0.01m² (average cone radius approximately 0.08m), and the magnetic field decays from a peak value B_max=10T to B_min=0.1T, with a timescale τ=1ms, the induced electromotive force is estimated as: ε ≈ N·S·ΔB / τ = 50 × 0.01 × 9.9 / 0.001 ≈ 4950V. This induced current is converted to DC by the rectifier module (equipped with a reverse current isolation diode) in the pulse power control system 7 and fed into the DC bus of the steady-state power generation system.

[0028] The auxiliary stabilizing coil 8, arranged close to the periphery of the collision fusion zone 3, is powered by an independent DC power supply or a quasi-steady-state power supply. It generates a continuous weak background magnetic field (0.1-0.5T, which is 1 / 10-1 / 20 of the peak field of the magnetic mirror), forming a magnetic cage to dampen the macroscopically violent motion of the plasma after the collision, reduce MHD instability, and confine charged fusion products to reduce vacuum chamber wall losses. The coil uses high-temperature resistant insulating material (polyimide / ceramic matrix composite material) and is equipped with a neutron shielding layer (boron-containing polyethylene) and active cooling channels.

[0029] The deuterium main fuel feeder 9a and the tritium ignition agent injector 9b are symmetrically arranged at the ends of the two forming zones 2. The deuterium feeder 9a is controlled by a piezoelectric valve or a solenoid valve, injecting approximately 1g of deuterium gas per pulse; the tritium injector 9b uses a micro-metering pump, injecting approximately 10mg of tritium in the first pulse, and dynamically reducing the amount injected in subsequent pulses according to the recovery amount.

[0030] A zoned vacuum pumping system 16 is arranged on the side wall of the conical transition section 4, employing a turbomolecular pump or a cryogenic pump. The axial distance between the pumping port and the forming zone 2 is greater than half the length of the conical transition section 4. The zoned vacuum pumping system 16 includes a cryogenic condensation and recovery branch 18. The cryogenic condensation trap 18 is arranged downstream of the tritium temporary adsorption trap 17, with an operating temperature below 4K (using liquid helium or a GM refrigerator). Utilizing the difference in boiling point between helium-3 (3.2K) and deuterium (23K) and tritium (25K), the helium-3 diffused into the conical transition section is liquefied, separated, and recovered.

[0031] A temporary tritium adsorption trap 17, located at the end of formation zone 2, uses a palladium film or uranium bed material to adsorb and recover tritium at low temperatures (room temperature to 100°C) between pulses. Before the next pulse, it is heated to 200-400°C and released to mix with the newly injected deuterium. Liquid helium-3 recovered from the cryogenic condensation trap 18 is vaporized and quantitatively reinjected into formation zone 2 as secondary fuel to participate in the D+³He reaction in the next pulse.

[0032] The core verification objective of this embodiment is to verify the confinement and recovery efficiency of the magnetic mirror coil group 6 for charged fusion products within a single pulse, and the power generation efficiency of reusing it as a DEC induction coil. If the dual-component recovery efficiency reaches more than 10% and the DEC induction efficiency reaches more than 5%, it proves that the concept of "time-sequence multiplexing + energy multiplexing + dual-component recovery" is feasible. Example 2: Engineering Verification Level Device

[0033] Based on the proof of concept, the device was scaled up to a total length of 3-5 meters, the inner diameter of the collision fusion zone 3 was 10-15 centimeters, and the energy storage of the fast compression coil 5 was increased to 0.5-2 MJ.

[0034] The peak magnetic field of the magnetic mirror coil group 6 is increased to 20-30T, and the duration of the magnetic bottle confinement mode is extended to 100-500μs to match the combustion timescale of higher density plasmas. The conical frame of the magnetic mirror coil group 6 is manufactured in segments using high-strength stainless steel or titanium alloy, with each segment having a cone angle of 5°-10°, and the segments are vacuum-sealed and welded together.

[0035] The pulse power supply control system 7 adds a feedback control module: based on the signal from the neutron detector (arranged around the fusion region) of the previous pulse, it dynamically calculates the tritium injection amount for the next pulse. For example, if the neutron yield of the previous pulse is higher than the threshold, it indicates that the tritium combustion is complete, and the external tritium injection amount for the next pulse can be reduced by 20%-50%; if the neutron yield is lower than the threshold, the tritium injection amount is appropriately increased. Simultaneously, the helium-3 reinjection amount is adjusted based on the liquid helium-3 recovery amount from the cryogenic condenser 18.

[0036] The auxiliary stabilizing coil 8 is upgraded to a superconducting coil or a water-cooled copper coil, extending its continuous operating time to several hours to several days, maintaining a stable weak magnetic cage environment. The superconducting coil uses REBCO high-temperature superconducting tape with a critical temperature >77K, which can significantly reduce cooling power consumption.

[0037] The partitioned vacuum pumping system 16 incorporates pressure sensors and feedback control, adjusting the pumping speed based on the pressure in the transition section to optimize impurity removal efficiency. The cryogenic condenser 18 employs multi-stage refrigeration: a first stage of 70K pre-cooling (liquid nitrogen), a second stage of 10K cryogenic cooling (GM refrigerator stage 1), and a third stage of <4K ultra-low temperature cooling (GM refrigerator stage 2 or dilution refrigerator) to improve helium-3 recovery rate.

[0038] The core verification objectives of this embodiment are: to verify whether the system tritium inventory tends to reach dynamic equilibrium after multiple pulse superpositions, and whether the annual external tritium consumption can be controlled at the gram level; at the same time, to verify the pulse power accumulation effect of the magnetic mirror coil group 6 being reused as a DEC induction coil, and the continuous operation reliability of helium-3 cryogenic condensation recovery. Example 3: Steady-state power generation system

[0039] like Figure 5 As shown, the energy released by fusion is released in pulses. For the DD reaction, the energy released in each pulse includes: • The kinetic energy of charged particles (tritium, helium-3, protons, accounting for about 60% of the total energy): The expansion of plasma drives the change of the magnetic field, which induces a pulse current in the magnetic mirror coil group 6 (reused as a DEC induction coil), and converts it into DC power through the rectifier module in the pulse power control system 7; • Neutron energy (approximately 40% of total energy): absorbed and converted into thermal energy by the cladding or shielding layer surrounding the collision-fusion region 3.

[0040] The pulse-induced electrical energy is converted into DC power by the pulse rectifier 10 (high-voltage silicon stack or silicon carbide rectifier module) and fed into the DC bus. Since the FRC pulse operating frequency is on the order of 1-100Hz, the DC power output by the pulse rectifier 10 fluctuates significantly.

[0041] The energy storage and damping array 11 is connected in parallel to the DC bus and includes a supercapacitor bank (capacity approximately 1-10F, withstand voltage 1-10kV) or a high-speed flywheel energy storage system (moment of inertia 10-100kg·m², rotational speed 10000-50000rpm). It is used to absorb pulse power peaks and release energy during pulse gaps, suppressing fluctuating DC levels to a stable DC bus voltage (fluctuation rate <5%).

[0042] It should be noted that the energy storage smoothing array 11 in this embodiment is not a general-purpose energy storage device, but is specifically designed to match the characteristics of FRC pulses: its capacity must meet the characteristics of absorbing microsecond-level pulse width and megawatt-level peak power steep rise and fall within the pulse repetition frequency range of 1-100Hz; its power density must match the millisecond-level decay timescale of the DEC induced current to ensure that the charge and discharge cycle is completed within the interval between adjacent pulses (10ms-1s). Conventional photovoltaic or wind power energy storage systems are designed for continuous smooth power, and their charge and discharge rates and cycle frequencies cannot meet the FRC pulse conditions. Therefore, the energy storage smoothing array 11 of this invention is a dedicated power conversion link tightly coupled to the FRC fusion reactor.

[0043] The grid-connected inverter 12 converts stable DC power into 50Hz / 60Hz AC power (voltage 380V / 10kV / 35kV selectable), which is then connected to the power grid 14 via the step-up transformer 13.

[0044] The control system 15 adjusts the FRC pulse repetition frequency (within the range of 1-100Hz) in real time according to the load command of the power grid 14 to achieve the load tracking capability of the fusion power station. The switching sequence of the grid-connected inverter 12 is synchronously matched with the magnetic field decay timescale of the recovery decay mode to complete the commutation between pulses and avoid conflict with the rising edge of the current of the next pulse compression.

[0045] Key timing sequence: The magnetic field decay timescale τ (milliseconds) of the recovery decay mode (T3-T4) is precisely matched with the current rise time t_rise (microseconds) of the fast compression coil 5 in the next pulse, satisfying τ > 100·t_rise. Specifically, in the early stage of the recovery decay mode, the current of the magnetic mirror coil group 6 decays rapidly from its peak, and the plasma expansion cuts the magnetic field lines, generating DEC induced current; in the late stage of the recovery decay mode, the magnetic field of the magnetic bottle has dropped to a level that is insufficient to confine high-energy products (about 0.01-0.1T). At this time, the fusion products have been fully recovered into the formation region 2, the DEC induced power generation process has been completed, and it does not affect the start of the next pulse. Example 4: Timing Relationship between Inter-Pulse Zoned Pumping and Two-Component Recovery

[0046] like Figure 6 , Figure 8 As shown, the timing relationship between inter-pulse pumping and two-component recovery is as follows: T3-T4a (early stage of recovery decay, approximately 0.2-2 ms): Magnetic mirror coil group 6 switches to recovery decay mode, and the magnetic field decays to form a weak gradient pointing towards formation region 2. The enriched plasma (containing tritium, helium-3, and unreacted deuterium) diffuses towards formation region 2 under the drive of the magnetic field gradient. During this stage, the partitioned vacuum pumping system 16 is shut down to ensure that tritium and helium-3 are not disturbed by the pumping.

[0047] T4a-T4b (Bicomponent Selective Recovery Stage, approximately 0.3-3 ms): Residual gas (mainly protons, a small amount of unreacted deuterium, and impurities) diffused into the conical transition section 4 reaches the extraction port. During this stage, the zoned vacuum extraction system 16 is activated (at a low pumping speed) to selectively remove light waste gas. Simultaneously, the cryogenic condenser 18 is activated to liquefy and recover helium-3 diffused into this region at below 4 K; a temporary tritium adsorption trap 17 at the end of zone 2 is formed to selectively adsorb tritium at room temperature.

[0048] T4b-T0' (Preparation Stage): The partitioned vacuum pumping system 16 is shut down, the tritium temporary adsorption trap 17 is heated to 200-400°C to release the adsorbed tritium, and the liquid helium-3 recovered by the cryogenic condenser trap 18 is vaporized and reinjected, mixed with the newly injected deuterium, in preparation for the next pulse.

[0049] • Physical safeguards for two-component recovery: • Magnetic bottle confinement priority: During the T2-T3 stage, tritium and helium-3 are jointly confined in the fusion region and do not participate in the pumping; • Recycling attenuation orientation: The magnetic field gradient in the T3-T4a stage points towards the formation region, and the two components preferentially flow to the formation region rather than the transition section; • Formation zone "two-component trap": Tritium adsorption trap 17 at the end adsorbs tritium at low temperature and releases it by heating before the pulse; Helium-3 is liquefied and recovered in the low-temperature condensation trap 18 in the transition section due to its chemical inertness; • Zoned air extraction: The distance between the air extraction port and the formation zone is greater than 1 / 2 of the transition section length, and the air extraction effect does not cover the formation zone. Example 5: Physical Mechanism of Multiplexing Magnetic Mirror Coil Groups into DEC Induction Coils

[0050] During the recovery decay mode, the current of the magnetic mirror coil group 6 decays from its peak value I_max to zero according to a preset curve. Let the number of turns of the magnetic mirror coil group 6 be N, and the effective area of ​​a single turn be S (average cone area). Assuming that plasma expansion causes the magnetic flux to change from Φ_max to Φ_min, the induced electromotive force is: ε = -N · dΦ / dt = -N·S·dB / dt.

[0051] Because the current decay timescale τ (milliseconds) of the recovery decay mode is much larger than the rise timescale (microseconds) of the fast compression mode, the absolute value of dB / dt is small, but the product of N and S is large. In addition, the tapered coils are closely arranged to increase the effective magnetic flux change area. Therefore, the total induced electromotive force can reach hundreds of volts to thousands of volts, which is sufficient to drive the rectifier module.

[0052] The direction of the induced current is opposite to the direction of the original current of the magnetic mirror coil group 6 (Lenz's law). Therefore, the rectifier module in the pulse power control system 7 is equipped with a reverse current isolation diode to prevent the induced current from flowing back into the discharge capacitor of the magnetic mirror coil group 6.

[0053] Advantages of using magnetic mirror coil group 6 as a DEC induction coil: • Simplest structure: No additional coils required, compact axial space of the vacuum chamber; • Shortest energy path: The induced current is generated directly from the constraint coil, reducing transmission losses; • Natural timing matching: DEC power generation and dual-component recovery share the same timing window, requiring no additional control logic; • Conical bonding enhances efficiency: The conical coil frame increases the effective magnetic flux cutting area during plasma expansion, thereby improving the induced electromotive force. Example 6: Quasi-pulse self-sustaining two-component material balance and energy relay

[0054] like Figure 4 , Figure 8 As shown, taking a single pulse consumption of 1 gram of deuterium and an initial tritium igniter of 10 mg as an example: First pulse (P1): 1g of deuterium + 10mg of tritium were injected, and after rapid compression, DD fusion was achieved in collision fusion zone 3. Based on a combustion depth of 5%, approximately 0.5mg of tritium and approximately 0.5mg of helium-3 were produced.

[0055] Magnetic bottle confinement recovery: Magnetic mirror coil group 6 switches to magnetic bottle confinement mode and maintains it for 50-200μs. Assuming the recovery efficiency of charged products is 30%-60%, approximately 0.15-0.3mg of tritium and approximately 0.15-0.3mg of helium-3 can be recovered between pulses.

[0056] Selective recovery of two components: tritium is retained by adsorption trap 17, and helium-3 is liquefied and recovered by cryogenic condensation trap 18.

[0057] DEC Induction Power Generation: During the recovery decay mode, the magnetic mirror coil group 6 is reused as a DEC induction coil, and the induced electrical energy generated by plasma expansion is rectified and fed into the DC bus.

[0058] Second pulse (P2): 1g of deuterium is injected into formation region 2 (via deuterium feeder 9a). The newly injected deuterium, along with the recovered tritium (0.15-0.3mg) and helium-3 (0.15-0.3mg), are rapidly compressed into collision fusion region 3. Due to the presence of residual tritium, the deuterium-deuterium fusion ignition threshold of the second pulse is reduced, and the external tritium supply can be reduced to 7mg. Due to the participation of helium-3, the D+³He reaction contributes additional energy.

[0059] Repeating the above process, after a startup phase of about 50-100 pulses, the system's tritium inventory tends to reach dynamic equilibrium, and the annual net consumption of external tritium can be controlled at the level of 1-10 grams; helium-3 achieves internal circulation, and the annual net consumption approaches zero.

[0060] Energy relay mechanism: Unlike the traditional pure pulse operation mode that involves complete emptying and refilling after each pulse, the quasi-pulse operation of this invention uses the recovery and decay mode of the magnetic mirror coil group 6 to use the fusion products (tritium, helium-3) of the previous pulse (Pn) as the initial plasma component of the subsequent pulse (Pn+1), while converting the expansion magnetic energy of the previous pulse into electrical energy via DEC. This energy relay not only reduces the external tritium igniter requirement of the subsequent pulse, but also enhances the energy output of the subsequent pulse through the secondary reactions of tritium, helium-3, and deuterium (D+T, D+³He). Example 7: Comparison with Existing Technologies

[0061] The core difference between the quasi-pulse FRC deuterium-deuterium fusion reactor of this invention and the existing FRC technology route lies in: In existing technologies, the two-stage cascaded magnetic compression route directly compresses deuterium to fusion conditions, requiring the retention of a second-stage active slow compression coil and a supporting high-power power supply system, resulting in high engineering complexity. Achieving fusion by directly compressing deuterium to fusion is extremely challenging in terms of technical implementation. This invention adopts a single-stage fast compression combined with the time-multiplexing of magnetic mirror coil groups, and achieves engineering-feasible fusion operation through a tritium catalytic ignition mechanism, reducing the difficulty of technical implementation.

[0062] Existing FRC technologies such as Helion Energy employ a first-stage DT and later-stage D-³He fuel route, which is essentially an "open-loop" operation. Fusion products (especially tritium and helium-3) are discharged or sold after combustion, requiring the device to be refueled. Tritium resources are not utilized, and the source of helium-3 is limited. This invention adopts a DD fuel route, which achieves magnetic confinement recovery and inter-pulse recycling of fusion products within the same device through the time-series multiplexing of magnetic mirror coil groups. This transforms tritium from a "disposable consumable" into a "recyclable igniter." Helium-3, as a byproduct, is recovered through cryogenic condensation and reinjected to participate in the D+³He reaction, significantly reducing dependence on external fuels.

[0063] Existing FRC technology uses direct induction recovery to recover pulse energy into capacitor banks for release in the next pulse within the device. This is an internal energy loop at the experimental device level, physically isolated from the external power grid. This invention reuses a magnetic mirror coil group as a DEC induction coil, converting the magnetic energy changes from plasma expansion into electrical energy. This electrical energy is then converted into steady-state AC power via a rectification-storage-inverter stage tightly coupled with the FRC pulse characteristics, forming a complete industrial closed loop from fusion reaction to power system. This upgrades the FRC device from an experimental prototype to a grid-connected baseload power source. The capacity and power density of the energy storage smoothing array are specifically matched to the pulse repetition frequency (1-100Hz) and microsecond-level pulse width, overcoming the technical obstacle of conventional grid-connected systems being unable to adapt to the steep rise and fall characteristics of pulse power.

[0064] In existing FRC devices, the DEC induction coil and magnetic confinement coil are arranged independently, resulting in redundant device structure and low axial space utilization. This invention reuses the magnetic mirror coil group as the DEC induction coil and adopts a conical fit arrangement, eliminating the need for a separate DEC coil arrangement, reducing one set of coils and its power supply system, and achieving the simplest structure.

[0065] Existing FRC devices lack a pulse-to-pulse dual-component selective recovery mechanism, making it impossible to simultaneously address residual impurity removal and tritium and helium-3 resource retention. This invention resolves the contradiction between tritium and helium-3 self-sustaining operation and residual gas removal through a zoned vacuum pumping system that selectively removes light waste gas from the sidewall of the conical transition section, while simultaneously utilizing a dual-component recovery mechanism of tritium adsorption traps and helium-3 condensation traps.

Claims

1. A quasi-pulse FRC deuterium-deuterium fusion reactor, characterized in that, include: A linear vacuum chamber has FRC plasma formation regions symmetrically arranged at both ends and a straight cylindrical collision fusion region in the middle. The formation region and the collision fusion region are connected by a conical transition section. The diameter of the formation region is larger than the diameter of the collision fusion region. An FRC plasma forming coil is arranged inside the forming region to pre-ionize the injected deuterium gas and establish a reverse magnetic field to form FRC plasma. A fast compression coil, arranged outside the formation region, is used to generate a pulsed magnetic field to accelerate the FRC plasma axially toward the collision fusion region; The magnetic mirror coil assembly is symmetrically arranged at the small-diameter end of the conical transition section near the collision fusion zone. The coil frame of the magnetic mirror coil assembly is conical, and its inner diameter gradually changes from the outer diameter of the forming zone to the outer diameter of the collision fusion zone along the axial direction and closely fits the outer wall of the conical transition section. The magnetic mirror coil assembly sequentially executes the following four functional modes within a single pulse cycle: • Compression-assisted mode: In the fast compression phase, it works in conjunction with the fast compression coil to form a magnetic field gradient pointing towards the collision fusion region, assisting in the acceleration of FRC plasma; • Magnetic mirror reflection mode: A peak magnetic field intensity is formed during the collision and fusion stage, reflecting high-energy charged fusion products; • Magnetic bottle confinement mode: Maintains a weak magnetic field confinement region between two magnetic field peaks, prolonging the residence time of tritium nuclei, helium-3 nuclei, and protons in the fusion region; • Recovery attenuation mode: The magnetic field attenuates controllably according to a preset time sequence, guiding the confined enriched plasma to the formation region. At the same time, the magnetic mirror coil group is reused as a direct energy conversion induction coil, and the plasma expansion drives the change of magnetic flux to induce a pulse current in the magnetic mirror coil group. An auxiliary stabilizing coil is arranged close to the periphery of the collision fusion zone to provide a continuous or quasi-steady-state weak background magnetic field, forming a magnetic cage to dampen the violent movement of the plasma after the collision and to confine the charged fusion products to reduce wall losses. A pulse power supply control system is used to provide timing pulse current to the magnetic mirror coil group to realize the switching of the four functional modes, and includes a rectifier module for converting the pulse current generated by the direct energy conversion induction coil into direct current. The deuterium main fuel feeder and the tritium igniter injector are symmetrically arranged at the ends of the two forming areas. The deuterium main fuel feeder is used to replenish the main fuel deuterium between pulses, and the tritium igniter injector is used to quantitatively inject a trace amount of tritium igniter. A temporary tritium adsorption trap is arranged at the end of the formation region for adsorbing and recovering tritium between pulses and releasing it by heating before the next pulse; A partitioned vacuum pumping system is arranged on the side wall of the conical transition section. The partitioned vacuum pumping system includes a cryogenic condensation and recovery branch. The cryogenic condensation and recovery branch is equipped with a cryogenic condensation trap for liquefying and separating the helium-3 that diffuses into the conical transition section between pulses in the cryogenic condensation trap for recovery. The pumping port of the partitioned vacuum pumping system is far away from the formation zone to avoid pumping out the tritium recovered to the formation zone.

2. The quasi-pulse FRC deuterium-deuterium fusion reactor according to claim 1, characterized in that, The FRC plasma forming coil is a θ-pinch type or coaxial gun type coil with an inductance of 0.1μH to 1μH, and is powered by a low-voltage, high-current pulse power supply; the weak background magnetic field strength provided by the auxiliary stabilizing coil is 1 / 10 to 1 / 20 of the peak magnetic field strength of the magnetic mirror coil group in magnetic mirror reflection mode; the duration of the magnetic mirror coil group in magnetic bottle confinement mode is more than an order of magnitude greater than the pulse duration of the fast compression coil.

3. The quasi-pulse FRC deuterium-deuterium fusion reactor according to claim 1, characterized in that, The magnetic mirror coil group is a low-inductance hollow coil with an inductance value of 0.5μH to 2μH and 20 to 100 turns; the fast compression coil is a θ-pinch type coil with an inductance value of 1μH to 5μH and is powered by a high-voltage pulse capacitor bank; the rectifier module in the pulse power control system is equipped with a reverse current isolation diode to prevent the direct energy conversion induced current from flowing back to the discharge capacitor of the magnetic mirror coil group.

4. The quasi-pulse FRC deuterium-deuterium fusion reactor according to claim 1, characterized in that, The partitioned vacuum pumping system includes a turbomolecular pump or a cryogenic pump, with the pumping port located on the side wall of the conical transition section, and the axial distance between the pumping port and the formation zone is greater than 1 / 2 of the length of the conical transition section. The tritium temporary adsorption trap uses palladium film or uranium bed material, and the cryogenic condensation trap operates at a temperature below 4K. The liquefaction and separation of helium-3 is achieved by utilizing the difference between the boiling points of helium-3 and deuterium and tritium.

5. A quasi-pulse FRC deuterium-deuterium fusion reactor device based on any one of claims 1 to 4 and a method for steady-state power generation of its nuclear fuel self-sustaining cycle, characterized in that, Includes the following steps: First pulse operation: Deuterium, the main fuel, and a small amount of tritium, the igniter, are injected into the two-end formation regions. The FRC plasma formation coil is activated to generate pre-ionized plasma and establish a reverse magnetic field. Then, the fast compression coil is activated to drive the two FRC plasmas to collide and compress in the collision fusion region. The deuterium-deuterium fusion reaction produces tritium, helium-3, protons, and neutrons. Combustion confinement stage: Switch the magnetic mirror coil group to magnetic mirror reflection mode and magnetic bottle confinement mode to confine the charged particles in the fusion products to the collision fusion region. Utilize the difference in cyclotron frequency between tritium nuclei and helium-3 nuclei to make them stay together in the weak field region of the magnetic bottle and undergo secondary reactions with the background deuterium plasma to amplify the energy output. Inter-pulse recovery phase: Switch the magnetic mirror coil group to recovery attenuation mode. The magnetic field attenuation forms a weak gradient pointing towards the formation region, guiding the plasma enriched with tritium, helium-3 and unreacted deuterium to the formation region. Two-component selective recovery stage: Start the zoned vacuum pumping system to remove residual protons and impurity gases that have diffused into the conical transition section; at the same time, use the tritium temporary adsorption trap to selectively adsorb tritium at room temperature to 100°C, and use the low temperature condenser trap to liquefy and separate helium-3 at below 4K, retaining tritium and helium-3 in the formation zone. Direct energy conversion stage: During the recovery decay mode, the magnetic mirror coil group is reused as a direct energy conversion induction coil, and the induced current generated by the change in magnetic flux driven by plasma expansion is converted into direct current by the rectifier module; Quasi-pulse self-sustaining operation: The above pulse process is repeated. In subsequent pulses, deuterium is replenished through the deuterium main fuel feeder. The tritium igniter injector dynamically adjusts the external tritium injection amount according to the recovery amount. The tritium catalytic cycle is maintained by the recovered products, forming a quasi-pulse continuous operation. The tritium and helium-3 recovered in the previous pulse are used as igniters and energy amplification media in the subsequent pulses to participate in the secondary reaction. Steady-state power generation stage: The pulse power and direct energy released by fusion are collected and converted into induced electrical energy. The rectifier converts the induced electrical energy into DC power, the energy storage and smoothing array absorbs the pulse fluctuations, and then the grid-connected inverter converts the DC power into AC power for output to the grid. The capacity and power density of the energy storage and smoothing array are designed to match the pulse repetition frequency of the fusion reaction device (1-100Hz) and the peak pulse power, so as to convert the pulse power with pulse widths of Hertz to 100Hz and microseconds into a smooth DC bus.

6. The method for steady-state power generation using nuclear fuel self-sustaining cycle according to claim 5, characterized in that, The energy storage and stabilization array includes a supercapacitor bank or a high-speed flywheel energy storage system. The supercapacitor bank has a capacity of 1-10F and a withstand voltage of 1-10kV, or the high-speed flywheel energy storage system has a moment of inertia of 10-100kg·m² and a rotational speed of 10000-50000rpm. The external tritium injection amount of the subsequent pulse is adjusted according to the feedback of the neutron yield diagnostic signal of the previous pulse, so that the tritium inventory of the system approaches dynamic equilibrium.

7. The method for steady-state power generation using nuclear fuel self-sustaining cycle according to claim 5, characterized in that, The magnetic field decay timescale τ of the recovery decay mode and the rising edge t_rise of the current of the next pulse fast compression coil satisfy τ > 100·t_rise, so that the fusion products are fully recovered without affecting their efficiency of expansion and power generation; after the quasi-pulse self-sustaining operation has gone through the start-up phase of 50-100 pulses, the system tritium inventory tends to stabilize, and the annual net consumption of external tritium is controlled at the level of 1-10 grams.

8. The method for steady-state power generation using nuclear fuel self-sustaining cycle according to claim 5, characterized in that, The timing sequence between the dual-component selective recovery stage and the inter-pulse recovery stage is as follows: In the early stage of recovery attenuation, the current of the magnetic mirror coil group decays to form a weak gradient pointing towards the formation region, and the enriched plasma diffuses into the formation region. During this stage, the partitioned vacuum pumping system is shut down. In the late stage of recovery attenuation, the residual gas that has diffused to the conical transition section reaches the pumping port. During this stage, the partitioned vacuum pumping system is turned on and operates at a low pumping speed to selectively remove light waste gas. At the same time, the cryogenic condenser is started to recover helium-3.

9. The method for steady-state power generation using nuclear fuel self-sustaining cycle according to claim 5, characterized in that, The tritium temporarily adsorbed in the tritium trap at the end of the formation region is adsorbed and recovered at low temperature between pulses. It is then heated to 200-400°C before the next pulse and released, mixing with the newly injected deuterium to form the initial plasma. The liquid helium-3 recovered in the low-temperature condensation trap is vaporized and quantitatively reinjected into the formation region as secondary fuel to participate in the D-³He reaction in the next pulse.

10. The method for steady-state power generation using nuclear fuel self-sustaining cycle according to claim 5, characterized in that, The pulse repetition frequency during the steady-state power generation phase is adjusted in real time within the range of 1-100Hz according to the grid load command, realizing load tracking of the fusion power station; the direct energy conversion induced electrical energy is fed into the DC bus through the rectifier module in the pulse power control system, and after being smoothed by the energy storage smoothing array together with the DC power output from the pulse rectifier, it is converted into 50Hz / 60Hz power frequency AC power by the grid-connected inverter; the switching sequence of the grid-connected inverter is synchronously matched with the magnetic field decay time scale of the recovery decay mode, so as to complete the inverter commutation during the pulse gap.

Images