High liquid limit soil soil improvement system based on recycled micro-powder-based solid waste modification

Abstract

The application belongs to the technical field of soil improvement, and particularly relates to a high liquid limit soil soil improvement system based on regenerated micro-powder-based solid waste modification. The system comprises: a pretreatment and addressable site construction subsystem, which is used for forming type labels and state labels on the surface of regenerated micro-powder particles of a regenerated micro-powder-based soil improvement slurry; a grid imaging and execution subsystem, which comprises a magnetic-sound-light execution table and a characterization camera; an addressable microsphere supply and recovery subsystem, which realizes reversible combination and dissociation of the aforementioned microspheres and the aforementioned action sites; a controller, which is configured to execute a grid sequencing-interlocking bridge execution algorithm; and a partition injection and in-situ solidification subsystem, which is used for triggering a multi-phase collaborative solidification program according to a soil improvement target output by the aforementioned controller. The application can significantly improve the stability and durability of high liquid limit soil in complex environments, and efficiently utilize regenerated micro-powder-based solid waste materials to reduce the use of traditional cementing agents.

Description

Technical Field

[0001] This invention belongs to the field of soil improvement technology and relates to soil pollution improvement and remediation, specifically a soil improvement system for high liquid limit soil based on recycled micro-powder-based solid waste modification. Background Technology

[0002] Currently, in the fields of high liquid limit soil modification and solid waste resource utilization, there are various research and engineering practices both domestically and internationally targeting soil stability improvement and waste mineral reuse. Conventional high liquid limit soil modification mainly relies on traditional inorganic solidifying agents such as cement, lime, and fly ash, improving soil structure through processes such as ion exchange, pozzolanic reaction, and cementation. These methods have the advantages of readily available raw materials and simple construction, but they suffer from problems such as a single solidification system structure, insufficient long-term stability, and imbalance between volume expansion and shrinkage. Especially under cyclical wet-drying or freeze-thaw environments, the microscopic bridging structure of the solidified body is prone to degradation, leading to a decrease in macroscopic strength. Furthermore, these traditional solidifying agent systems often consume large amounts of calcium carbonate-type raw materials and clinker cement, resulting in high energy consumption and significant carbon emissions, which contradicts the trend of high-value utilization of solid waste.

[0003] In recent years, research has focused on utilizing recycled micropowders and industrial solid waste (such as steel slag powder, blast furnace slag powder, and construction waste mortar powder) as active components in soil modifiers through methods such as refining, activation, and co-blending. These technologies reduce the carbon footprint by increasing the specific surface area and surface activity of solid waste, thus partially replacing traditional solidifying agents. However, these methods share a common characteristic: they rely on the average reactivity of the material and lack controllable distribution of active centers at the particle scale. This leads to random reaction pathways, uneven structures, and uncontrollable formation of bridging networks in the mixed system. Existing micropowder-modified slurries still exhibit characteristics of "local density and overall dispersion" during solidification, with low interfacial bonding and a tendency for debonding or slippage at multiphase interfaces, making precise structural reconstruction impossible.

[0004] Some researchers have attempted to introduce surface modifiers or dispersants to imbue the surface of regenerated microparticles with chemical groups, thereby improving slurry dispersibility and particle bonding. However, these methods are generally limited by the irreversibility of chemical reactions and the constraints of the reaction environment. For example, organosilane surface modification can improve differences in hydrophilicity, but the bonds formed are irreversible, and the structure cannot be self-adjusted according to environmental requirements after curing. Ionic modification systems, on the other hand, exhibit poor stability under high ionic strength conditions, easily leading to gelation failure. Furthermore, existing technologies cannot identify and control the microstructural evolution between particles in real time within the curing system, lacking an intelligent control mechanism capable of actively adjusting bridging orientation and distribution. Summary of the Invention

[0005] The main objective of this invention is to provide a soil improvement system for high liquid limit soil based on recycled micronized solid waste modification. This invention can significantly improve the stability and durability of high liquid limit soil in complex environments, while efficiently utilizing recycled micronized solid waste materials and reducing the use of traditional binders.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A soil improvement system for high liquid limit soil based on recycled micronized solid waste modification includes: a pretreatment and addressable site construction subsystem for forming action sites with a type tag and a status tag on the surface of recycled micronized particles in a recycled micronized soil amendment slurry; a grid imaging and execution subsystem, including a magneto-acoustic-optical stage and a characterization camera, for locating, pulling, aligning, and stabilizing the aforementioned action sites within a grid cell; and an addressable microsphere supply and recovery subsystem, which provides A microspheres corresponding to the aforementioned type tags. The system comprises: B microspheres and neutral microspheres, enabling reversible binding and dissociation of the aforementioned microspheres with the aforementioned action sites; a controller configured to execute a grid-sequential-interlocked bridging execution algorithm to generate and issue bridging action commands within each grid cell, thereby forming preferentially distributed directional bridging contacts in the aforementioned regenerated micro-powder-based soil amendment slurry; and a partitioned injection and in-situ solidification subsystem for injecting the reconstructed aforementioned regenerated micro-powder-based soil amendment slurry into a high liquid limit soil layer in partitions according to a soil improvement target output by the aforementioned controller, and triggering a multiphase synergistic solidification procedure.

[0007] Furthermore, the pretreatment and addressable site construction subsystem includes: a closed mixing chamber, a continuous flow window channel, an initiation zone, a capping zone, and a functionalized spray head; wherein, the initiation zone activates the particle surface with an area array light source, the capping zone terminates unreacted sites by shading, and the functionalized spray head sequentially introduces A ligand and B ligand to assign the aforementioned action site type labels, and the status labels of all newly generated action sites are initially set to pending activation.

[0008] Furthermore, the addressable microsphere supply and recovery subsystem includes a microsphere storage tank, a quantitative injection component, a dissociation eluent channel, and a recovery filtration component. Microspheres A and B reversibly bind to action sites that match their type labels and have an open status label, while neutral microspheres reversibly attach to neutral action sites. After bridging, the dissociation eluent supplied through the aforementioned dissociation eluent channel causes microspheres A, B, and neutral microspheres to reversibly dissociate from their action sites and be collected by the aforementioned recovery filtration component.

[0009] Furthermore, the grid imaging and execution subsystem includes a magnetic field coil, an ultrasonic transducer, an area array light source, and a characterization camera; the controller establishes grid cells in the observation window of the aforementioned continuous flow window channel through a grid definition unit, and images grid cells one by one according to the row and column scanning path, recording the visible positions of microsphere A, microsphere B, and neutral microsphere to generate a site list.

[0010] Furthermore, the controller is configured to execute a grid sequencing-interlocking bridging execution algorithm, which includes the following steps: Step 5-1, State confirmation and on-site activation: For entries in the aforementioned site list whose state label is to be activated, local illumination is performed in the corresponding grid cell using the aforementioned area array light source, causing their state labels to change to open or disabled; Step 5-2, Candidate generation: Within the same grid cell, taking a type A entry as a reference, starting from the first entry in the aforementioned site list, each entry is enumerated sequentially, and a type B entry with the closest spatial location and an open state label is found, forming a candidate pair. Neutral entries with open state labels are then searched on both sides of the line connecting these candidate pairs as optional auxiliary entries; Step 5-3, Conflict detection and resolution: When the same entry is referenced by multiple candidate pairs... At that time, candidate pairs are retained or removed and temporary locks are released according to a fixed order rule; Step 5-4, bridging action command generation and issuance: bridging action commands are generated for each non-conflicting candidate pair and issued to the magnetic-acoustic-optical execution station for execution; Step 5-5, immediate review and cancellation: after each bridging action command is executed, the status of the relevant entries is reviewed immediately, and the contact is released according to a fixed cancellation order if necessary; Step 5-6, backscan and interlock replacement: within the same grid cell, stable contacts are replaced and interlock replacement is performed when preset conditions are met; Step 5-7, grid switching and merging: all grid cells are processed sequentially according to the aforementioned matrix scanning path, and stable contacts are merged to generate soil improvement targets, while recording an unprocessed reason label.

[0011] Furthermore, the fixed order rule defined in step 5-3 is as follows: priority is given to retaining candidate pairs that simultaneously contain type A entries with the status label "open" and type B entries with the status label "open"; if multiple candidate pairs meet or do not meet the aforementioned priority conditions, the one with the smaller sequence number of its type A entries in the site list is retained; if the sequence numbers of type A entries are the same, the one with the smaller sequence number of its type B entries is retained; entries in removed candidate pairs are temporarily unlocked and returned to the site list.

[0012] Furthermore, the bridging action command includes fixed fields and is executed in a fixed order: a) a directional traction field, which controls the aforementioned magnetic field coil to apply a magnetic field gradient along the direction of the candidate pair connection, causing the A microspheres and B microspheres combined with the A and B entries to displace in the same direction and drive the corresponding particles to approach; b) an attitude integration field, which controls the aforementioned ultrasonic transducer to output pulses to eliminate lateral misalignment and complete alignment with short-arc correction; c) an optional auxiliary contact field, which applies short-term magnetic traction along the connection normal after the main contact is formed, so that the neutral microsphere and the corresponding action site form a laterally stable contact; d) a hold and release field, which maintains magnetic-acoustic co-holding until the characterization camera observes stable contact characteristics, then closes the magnetic field and opens the aforementioned dissociation eluent channel to achieve reversible dissociation and recovery of the A microspheres, B microspheres and neutral microspheres.

[0013] Furthermore, the candidate generation defined in step 5-2 is performed in a fixed order of the site list within each grid cell. The generation and traversal of the site list follow a matrix scanning path from top to bottom and then from left to right. The determination of the closest spatial location is based on the minimum calculated value representing the camera imaging coordinate difference, thereby avoiding the intersection of execution paths of different candidate pairs within the same grid cell.

[0014] Furthermore, after completing the processing of all grid cells, the controller outputs the soil improvement target and assigns an unprocessed reason label to any unprocessed item. The type of this unprocessed reason label is limited to one of the following: no complement, disabled, or path blocked. At the same time, the controller executes a pre-curing verification process, extracting several stable contacts in each grid cell for secondary independent scanning and verification. After the verification is passed, the defined partition injection and in-situ curing subsystems are allowed to perform curing of the corresponding partitions.

[0015] The soil improvement system for high liquid limit soil based on recycled micronized solid waste modification of the present invention has the following beneficial effects: This invention provides a soil improvement system for high liquid limit soils based on recycled micro-powder-based solid waste modification. It overcomes the limitations of traditional soil solidification materials that rely on inorganic cementation reactions, achieving a synergistic unity between programmable structural control at the micro-particle scale and macroscopic stable solidification. By constructing action sites with type and state tags on the surface of recycled micro-powder particles, the system can form clear reaction and binding channels at the particle level, transforming the bridging behavior between particles from random collisions to controlled addressing. This mechanism ensures the structural consistency of the recycled micro-powder-based soil improvement slurry under different environments, thereby significantly improving the shear strength and volume stability of the solidified body. The system's grid imaging and execution subsystem introduces magnetic, acoustic, and optical coupling to achieve real-time positioning and directional traction of particles. While ensuring high spatial resolution, it reduces interference between multiple particles, enabling directional bridging contact to be corrected and maintained within nanosecond response times. Through a grid sequencing-interlocking bridging execution algorithm executed by the controller, the system generates bridging action commands in a fixed order within each grid cell, ensuring the determinism and repeatability of the particle bridging process and effectively avoiding the structural randomness problems of previous probabilistic aggregation-based methods. The addressable microsphere supply and recovery subsystem provides reversibly combined A-microspheres, B-microspheres, and neutral microspheres, making the bridging formation and dissociation processes between particles controllable and repeatable. This allows for local structural reconstruction and adjustment without damaging the overall structure. The zoned injection and in-situ solidification subsystem injects the slurry into zones according to the soil improvement targets output by the controller. Combined with a multiphase synergistic solidification program, it achieves spatial mapping consistency between the microstructure and the macroscopic injection, resulting in uniform density distribution and continuous pore interfaces in the solidified body.

[0016] The overall system of this invention forms a closed-loop mechanism for microstructure identification, directional reconstruction, and on-site solidification, exhibiting high repeatability and strong environmental adaptability. Compared with existing technologies, this invention not only improves the stability of high liquid limit soils under extreme humidity and temperature variations, but also achieves efficient utilization of recycled micro-powder-based solid waste, reduces carbon emissions, and decreases the use of traditional cement-based materials, providing a scalable and intelligently controllable new technical approach for solid waste resource utilization and soil engineering. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the preprocessing and addressable site construction subsystem provided in an embodiment of the present invention. Figure 2 This is a schematic diagram illustrating the effect of magnetic field gradient intensity on the displacement velocity of a microsphere, provided in an embodiment of the present invention. Experimental conditions: channel gap 0.5 mm, microsphere diameter 2.0 mm. The temperature was 25℃, and the measurement time was 100-1000ms. Conclusion: When the magnetic field gradient is 1.5-3.0 mT / mm, the displacement velocity of the microsphere is moderate, which can be observed without overshooting. Figure 3 This is a schematic diagram illustrating the effect of channel gap on the density uniformity of the action site provided in an embodiment of the present invention. Experimental conditions: irradiation time 6s, flow rate 20m³ / min, area array light source 365nm, and mask opening spacing 300mm. Conclusion: With channel gaps ranging from 0.3 to 1.0 m, the consistency of light dose is high, and the uniformity of site density is >90%. Figure 4 This is a schematic diagram illustrating the effect of dissociation eluent concentration on microsphere recovery and bridging retention rate provided in this embodiment of the invention. Experimental conditions: spraying time 3s, flow rate 30mL / min, competitive binder type EDTA, temperature 25℃; Conclusion: at 30-50 mmol / L, microsphere recovery rate >60% and bridging retention rate >50%, balancing recovery and structural stability. Figure 5 This is a schematic diagram illustrating the effect of ultrasonic frequency on the porosity elimination rate of the solidified body according to an embodiment of the present invention. Experimental conditions: injection pressure 0.15MPa, partition side length 1.0m, temperature control temperature 28℃, measurement method is CT scan; Conclusion: the porosity elimination rate is optimal in the frequency range of 50-80kHz, and excessively high frequencies will lead to increased energy attenuation. Detailed Implementation

[0019] The method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0020] Example 1: Reference Figure 1A soil improvement system for high liquid limit soil based on recycled micronized solid waste modification includes: a pretreatment and addressable site construction subsystem for forming action sites with a type tag and a status tag on the surface of recycled micronized particles in a recycled micronized soil improvement slurry; a grid imaging and execution subsystem, including a magneto-acoustic-optical execution stage and a characterization camera, for locating, pulling, aligning, and stabilizing the aforementioned action sites within a grid cell; and an addressable microsphere supply and recovery subsystem, which provides A-microspheres corresponding to the aforementioned type tags. The system comprises: microspheres, B-microspheres, and neutral microspheres, enabling reversible binding and dissociation of the aforementioned microspheres with the aforementioned action sites; a controller configured to execute a grid-sequential-interlocked bridging execution algorithm to generate and issue bridging action commands within each grid cell, thereby forming preferentially distributed directional bridging contacts in the aforementioned regenerated micro-powder-based soil amendment slurry; and a partitioned injection and in-situ solidification subsystem for injecting the reconstructed aforementioned regenerated micro-powder-based soil amendment slurry into a high liquid limit soil layer in partitions according to a soil improvement target output by the aforementioned controller, and triggering a multiphase synergistic solidification procedure.

[0021] Example 2: In one specific implementation, the preprocessing and addressable site construction subsystem is an integrated floor-mounted device, consisting of the following components arranged in series according to the flow direction: Closed Mixing Chamber: Used for mixing and stably conveying regenerated micronized soil amendment slurry. The closed mixing chamber is a cylindrical pressure vessel with a volume of 10 to 50 liters. The inner wall is lined with an inert layer. A replenishment port and an exhaust port are located at the top, and a discharge port is located at the bottom. A three-bladed agitator and wall scraper are installed inside the chamber. The agitator shaft is magnetically coupled to a motor to achieve a seal-free structure, reducing leakage and contamination. A circulating temperature-controlled jacket is installed outside the chamber to maintain a stable temperature within the range of 15°C to 35°C, ensuring rheological stability. Inlet Filter and Pressure Stabilization Section: Located between the closed mixing chamber and the continuous flow viewing window channel, this section includes a replaceable filter cartridge and a pressure-stabilizing buffer chamber. The filter pore size is 50 to 100 microns, and the buffer chamber volume is 0.5 to 2 liters, used to eliminate pulsations and keep flow fluctuations below 5% of the total flow rate. Continuous flow window channel: This is a parallel plate thin-layer channel with a length of 300 mm to 600 mm, a width of 60 mm to 120 mm, and a channel gap of 0.3 mm to 1.0 mm. The upper plate is a quartz window plate, and the lower plate is an inert substrate. The optical transmittance of the quartz window plate is not less than 90% in the range of 350 nm to 450 nm, ensuring the transmission of the area array light source and forming uniform illumination in the initiation zone. Inlet manifolds and outlet manifolds are respectively provided at both ends of the channel to ensure that the flow velocity deviation in the lateral distribution does not exceed 10%. Initiation zone: Located in the upstream section of the continuous flow window channel, the center corresponds to the illumination area of ​​the area array light source. The area array light source is a planar light-emitting array with a working wavelength selectable at 365 nm or 405 nm, located 50 mm to 120 mm away from the quartz window plate, with an illuminance uniformity of not less than 90%, and the illumination area covering the effective width of the channel. A replaceable alignment mask is set below the initiation zone to form spatially selective activation in the channel thickness direction and flow direction. End-capping zone: Located 30 mm to 80 mm downstream of the initiation zone, it consists of a full light shield and a cooling baffle. The inner surface of the light shield has a matte black coating to immediately terminate un-irradiated areas and prevent further reaction. The cooling baffle is maintained 5 to 10 degrees Celsius lower than the initiation zone to slow down subsequent surface reactions. Functionalized spray head: Located 20 mm to 60 mm downstream of the end-capping zone, it has equidistant nozzles arranged along the channel width with a nozzle spacing of 5 mm to 10 mm. The functionalized spray head includes two independent liquid supply circuits, supplying solutions for ligand A and ligand B respectively. The spray sequence can be programmably switched to "A first, then B" or "B first, then A". An inert washing spray follows immediately after the spray to remove unbound residues. Status label setting and protection station: Located downstream of the functionalized spray head, it houses a controllable photosensitive barrier layer spray assembly and a gentle air dryer. The spray assembly deposits a removable photosensitive barrier layer, ensuring all newly generated sites are uniformly in an activated state. The gentle air dryer uses a low-speed airflow of 0.5 to 1.5 meters per second to remove free solvent from the surface within 5 to 20 seconds, preventing particle re-aggregation. The outlet collection and reflux section includes the outlet manifold, primary separation tank, and reflux pump.The primary separation tank has a volume of 2 to 5 liters and is used to settle large particle aggregates and achieve online disposal or reuse. A process monitoring camera and illuminance monitor are installed above the quartz window plate and around the area array light source. The process monitoring camera acquires process images at 30 to 120 frames per second, and the illuminance monitor has 3 to 5 monitoring points at the boundary between the initiation zone and the end-capping zone for closed-loop adjustment of illumination uniformity. Delivery and control components include a metering pump, a back pressure valve, a three-way switching valve, and a programmable human-machine interface. The metering pump flow rate is adjustable from 5 ml / min to 60 ml / min, and the back pressure valve has a stable operating pressure between 0.05 MPa and 0.2 MPa. These components are connected sequentially via stainless steel rigid pipes and solvent-resistant flexible hoses. All surfaces in contact with the regenerated micro-powder-based soil amendment slurry are coated with an inert coating to avoid introducing additional surface-active sites.

[0022] The bottom outlet of the closed mixing chamber is connected to the inlet manifold of the continuous flow window channel via the inlet filter and pressure stabilization section. A static rectifier is added to the inlet section, with the rectifier blade height consistent with the channel gap to reduce inlet shear difference. The relative position of the area array light source and the quartz window plate is precisely fixed by a three-dimensional displacement bracket to ensure that the parallelism error between the light source and the channel surface does not exceed 0.2 degrees. The alignment light-shielding mask is connected to the window frame by positioning pins, and the repeated installation position deviation does not exceed 0.2 mm. The initiation zone, end-sealing zone, functionalized spray heads, and status labels and protection stations are arranged sequentially along the flow direction, with the center distance adjustable from 20 mm to 80 mm to adapt to different flow rates and residence times. The two liquid supply circuits of the functionalized spray heads are each equipped with a check valve and a micro-backflow mechanism to avoid cross-contamination. The distance between the spray head and the inert washing spray does not exceed 30 mm to shorten the diffusion time of unbound materials. A bypass is set in the outlet collection and return section, allowing the filter cartridge to be replaced or the back pressure valve to be adjusted without stopping the upstream. The process observation camera is positioned at the transition section between the initiation zone and the functionalized spray head to simultaneously observe early images of the formation of the action site and the uniformity of deposition; illuminance monitors are distributed at the four corners and center of the initiation zone to trigger automatic compensation when the illuminance deviation exceeds 10 percent.

[0023] The regenerated micro-powder-based soil amendment slurry is added to the closed mixing chamber, and the agitator and scraper are activated. The mixing time is 5 to 15 minutes. The purpose of the closed structure is to stabilize volatile matter and temperature, ensuring a uniform pre-reaction state on the particle surface; the scraper reduces dead-angle accumulation, preventing the formation of uneven action sites later. After inlet filtration and pressure stabilization, the flow pulsation is reduced to less than 5% of the total flow, ensuring that the thin layer thickness fluctuation within the continuous flow window channel is less than 0.1 mm, thereby guaranteeing a consistent light dose in the initiation zone.

[0024] The area array light source is set to uniform illumination mode, and the alignment mask uses a regular or quasi-random array of openings. Light is incident from the quartz window near the process observation camera rather than from the lower plate because upper incident light reduces dose shading caused by scattering and obstruction near the lower wall of the thin layer, and ensures the image is consistent with the geometric reference of the illumination, facilitating online calibration. The wavelength of the area array light source is set to 365 nm or 405 nm. These two wavelengths are chosen because the quartz window has a transmittance of over 90% in this band, and most surface-initiated layers have significant absorption peaks in this band, enabling activation within 2 to 8 seconds of relatively short illumination time. The geometry and spacing of the mask openings are matched to the channel gap and the average particle size. For example, the opening diameter is set to 80 to 200 micrometers, and the opening spacing is set to 200 to 500 micrometers. This allows for the formation of discrete activation sites at the single-particle scale, rather than forming a continuous film between adjacent particles, facilitating subsequent alignment and traction.

[0025] A light-shielding cover with an matting inner wall is placed downstream of the initiation zone, allowing the thin layer leaving the illumination range to enter a fully shaded environment within 10 milliseconds. A cooling baffle 5 to 10 degrees Celsius lower than the initiation zone is set up to rapidly reduce residual reactions after leaving the illumination range, avoiding pattern blurring caused by edge diffusion. The significance of rapid end-capping is to maintain clear boundaries of the action sites, making the deposition area of ​​type tags controllable, thereby reducing the probability of overlap between different type tags.

[0026] The A-ligand solution and the B-ligand solution are sprayed sequentially through two supply circuits via a functionalized spray head. The order can be set to "A first, then B" or "B first, then A". The spraying time is 1 to 3 seconds, with an inert wash interval of 1 to 2 seconds. The reason for using sequential spraying instead of simultaneous spraying is to preserve the pattern differences formed in the initiation zone spatially, allowing A-ligands and B-ligands to preferentially bind within their respective coverage areas, reducing the probability of type conflict between adjacent sites on the same particle surface. When it is necessary to form neutral binding sites, an open area can be set in the light-shielding mask, or in some open areas, only inert washing is performed without spraying A-ligands and B-ligands. This results in a surface area that does not specifically bind to A-ligands or B-ligands, allowing reversible attachment of neutral binding sites to subsequent neutral microspheres.

[0027] After the functionalized spraying is completed, a photosensitive barrier layer of controllable thickness is sprayed separately, putting all newly generated action sites in an inactive state. A photosensitive barrier layer was chosen instead of a chemically irreversible sealing layer because subsequent on-site activation and immediate observation are required in the grid imaging and execution subsystem. The photosensitive barrier layer can be removed within hundreds of milliseconds to several seconds under the illumination of an area light source or a localized spot, without introducing new uncontrollable sites. A gentle airflow dryer completes surface drying within 5 to 20 seconds using a low-speed airflow of 0.5 to 1.5 meters per second. The choice of a low-speed airflow reduces secondary particle aggregation within the channel, thereby maintaining the spatial resolvability of discrete sites.

[0028] The thin-layer flow enters the primary separation tank through the outlet manifold. Within 2 to 5 minutes, the separation tank completes the gravity settling of large particle agglomerates. The settling material can be periodically discharged or recycled. The clarified liquid is returned to the closed mixing chamber by a reflux pump, achieving stable circulation. The purpose of the separation tank is to prevent agglomerates from blocking light during the next flow through the initiation zone, maintaining the formation efficiency of the active sites at a stable rate of over 80%.

[0029] The process monitoring camera captures a comparative image at the end of the initiation zone. Based on the brightness and darkness texture in the image and the illuminance monitoring data of the array light source, if the corner illuminance is detected to be less than 90% of the center illuminance, the system automatically extends the illumination time of the corner area or slightly increases the current drive in that area, ensuring that the final site density varies by no more than 10% in the channel width direction. This dual correction based on brightness and darkness texture and illuminance monitoring avoids overcompensation caused by single feedback lag, maintaining the spatial statistical characteristics of the action sites stable during continuous production.

[0030] Example 3: In some applications, a low-temperature plasma generator is used instead of an area array light source as the activation method for the initiation region. The low-temperature plasma operates at 20°C to 40°C, with a processing time of 0.5 to 2 seconds. The advantage of choosing low-temperature plasma is that it can also produce surface activation effects on dark or low-transmittance particles, making it suitable for scenarios with complex sources of regenerated micropowder. To maintain synergy with the subsequent end-capping region, a metal separator is placed between the plasma nozzle and the light shield to prevent charged particles from extending into the end-capping region. When it is necessary to form a higher density of discrete action sites on the same particle, the functionalized spray head can be replaced with a reciprocating scanning spray head. The reciprocating stroke is 5 mm to 15 mm, and the scanning speed is 50 mm / s to 150 mm / s. By synchronously controlling the scanning phase of the area array light source and the spray head, the site density can be increased by approximately 20% to 40% while maintaining a constant flow rate, without significantly increasing the overlap probability. In addition to forming a neutral region through a light shield, a neutral end-capping spray can be added after the functionalized spray, with a spray time of 0.5 to 1.5 seconds. The purpose of neutral end-capping spray is to occupy the remaining weak interaction sites, reduce the mutual interference between A ligand and B ligand in the edge region, so that the neutral interaction sites only play a lateral stabilizing role in subsequent traction, and reduce uncertainties on the main contact path.

[0031] Example 4: In the grid imaging and execution subsystem, the magneto-acoustic-optical execution stage is an integrated platform. A quartz window plate is mounted above the platform, and an inert substrate is below, both defining a continuous flow window channel. The following components are symmetrically mounted around the observation window on the platform: a pair of magnetic field coils are arranged along the long and short sides of the channel to generate a low-amplitude magnetic field gradient in any direction within the plane; another pair of coils is arranged in the normal direction of the channel to provide normal clamping when needed. The center distance between the long and short side coils is 80 mm to 180 mm, and the inner diameter of the coils is 60 mm to 120 mm. The coil frame is made of heat-resistant insulating material, and the winding is made of enameled copper wire. An ultrasonic transducer array is mounted below the inert substrate in a one-dimensional or two-dimensional array, with a center-to-center spacing of 10 mm to 20 mm and an operating frequency of 0.5 MHz to 2 MHz. An external sound-absorbing frame is provided to reduce standing waves caused by reflections. A surface array light source is mounted 50 mm to 120 mm above the quartz window plate, covering the observation window where the grid unit is located. The area array light source has two operating channels: one for activation (e.g., 365 nm or 405 nm) and one for imaging illumination (e.g., 520 nm to 560 nm). These two channels can be opened and closed independently to avoid mutual interference. The mechanical base and vibration-damping platform are integrally cast, with vibration-damping elements installed at the four corners of the base, ensuring that the platform's environmental vibration transmissibility is less than 10% in the 10 Hz to 200 Hz range. Translation and fine-tuning guides are provided on the base for precise alignment of the characterization camera and the area array light source. The characterization camera is mounted perpendicular to the quartz window plate and equipped with a telecentric lens to reduce perspective errors at the field of view edges. The sensor resolution ranges from 4 megapixels to 16 megapixels, with each pixel corresponding to an actual channel length of 0.5 μm to 2.0 μm. The frame rate is 30 to 120 frames per second. A bandpass filter is placed in front of the lens, allowing only the imaging illumination band to pass through, thereby suppressing stray light from the activation band. The focusing plane of the camera is set on the side closest to the quartz window plate, and the depth from the quartz interface is preferably selected to be the upper half of the channel thickness (for example, when the channel gap is 0.5 mm, the focusing depth is set in the range of 0.1 mm to 0.2 mm from the upper interface). Choosing to focus closer to the camera side has three advantages: First, the scattering in the upper half of the thin layer is smaller, resulting in higher contrast of the microsphere highlights, which is beneficial for accurate segmentation; second, the refraction path is shorter closer to the camera side, resulting in smaller geometric distortion, which is beneficial for accurately calculating the connection direction within the grid cell; third, the effect of small axial drift on the image plane size is more predictable, making it easier to set criteria for stable contact features.

[0032] The area array light source and characterization camera operate in a time-multiplexed manner: when activation is required, only the activation band is activated, and the imaging illumination is turned off; when imaging is required, only the imaging illumination band is activated, and the activation band is turned off. The switching time between the two is no more than 20 milliseconds to facilitate rapid re-inspection. The magnetic field coil power supply and the ultrasonic transducer power supply are powered separately and connected to the control cabinet through independent shielded wiring harnesses. The rise and fall edges of the coil drive current employ a slope-limiting strategy, ensuring that magnetic field changes are completed within 5 to 20 milliseconds, thereby avoiding magneto-induced interference to the characterization camera imaging. A light-shielding barrier is installed between the characterization camera and the area array light source, with the inner wall of the barrier made of matte black material to reduce spot drift caused by specular reflection. A sound-guiding layer with a thickness of 0.5 mm to 1.5 mm is used between the ultrasonic transducer array and the inert substrate. The sound-guiding layer covers the entire observation window area, and its edges are fitted with the sound-absorbing frame to reduce edge diffraction.

[0033] The grid cells are defined as overlay layers within the characterization camera's field of view. The side length of a single grid cell ranges from 200 to 600 micrometers, with the specific value matching the site density and microsphere size. The scanning order follows a top-down, then left-to-right row-and-column sequence. The direct benefit of this fixed order is that it aligns the selection of the connection direction with the coil arrangement, minimizes the amplitude of magnetic field vector switching, thereby reducing the transition time between each grid cell and avoiding intersections of execution paths between different grid cells. The grid coordinates and the characterization camera pixel coordinates are calibrated once after assembly: a transparent scribe plate is adhered to the quartz window plate, 3 to 5 calibration images are captured, the ratio of pixels to actual lengths is calculated, and the angular deviation is recorded. The calibration data is written to the control cabinet for direct retrieval during connection generation.

[0034] Example 5: Imaging illumination is turned on, and the camera acquires continuous images, accumulating 3 to 5 frames to obtain a background reference image, used to eliminate slow illumination unevenness. A real-time image is acquired and pixel-level differencing is performed with the background reference image to obtain candidate bright spots. Microspheres A, B, and neutral microspheres exhibit three distinguishable color or brightness responses under the imaging illumination band, because the recognition coatings on the microsphere surfaces have different reflection or emission characteristics. By setting three discrete transmission windows on the filter, the three types of responses can be directly separated in the same frame without switching filters. Connected component extraction is performed on the differential bright spots, and spots with an area less than 4 pixels or greater than 400 pixels are removed to eliminate noise and adhesion. The geometric center is calculated for the remaining spots to obtain the location sets of microspheres A, B, and neutral microspheres. Within the same grid cell, according to the fixed order of the site list, the current type A entry is selected, and the entry with the closest in-plane distance to it in the location set of microspheres B is found to obtain candidate pairs. The rationale for choosing "nearest" is that, in a thin-layer environment, a shorter connection means a smaller magnetic field switching amplitude required for the approach path, a shorter ultrasonic integration time, and less disturbance to surrounding untreated items.

[0035] Based on the connection direction of the candidate pairs, the long and short side coils are driven in the same direction to establish a synthetic magnetic field gradient along the connection direction. The gradient amplitude is selected within a fixed low range to produce observable but not overshooting displacements between microspheres A and B along the connection direction. The characterization camera acquires the displacement process at 60 to 120 frames per second, calculating the distance change between the two geometric centers in real time. When the distance converges to near a preset contact threshold, the coil drive is automatically reduced, entering the alignment stage. The reason for using magnetic field gradient traction instead of a simple constant magnetic field is that the gradient can provide directional thrust in the plane, causing the microspheres to move closer along the target connection line, while having a smaller impact in the normal direction, preventing the particles from being pressed against the lower or upper wall and reducing accidental jamming caused by wall friction.

[0036] The ultrasonic transducer array is activated with short pulses, with pulse durations ranging from 10 to 50 milliseconds, intervals from 50 to 200 milliseconds, and pulse counts from 3 to 7. These short pulses generate a gentle localized microfluidic flow at the approaching ends, eliminating minor lateral misalignments and aiding particle alignment. Simultaneous short-arc correction is performed: coils along the normal to the connecting line are alternately energized with small amplitudes, causing the synthetic magnetic field to sweep across a small angle range near the connecting line, ranging from 2 to 8 degrees. The effect of short-arc correction is to gradually correct the contact posture of the particles without significantly altering the distance between them, resulting in a more complete contact surface. During alignment, the characterization camera continuously observes the contrast changes at the contact edges. Alignment is considered complete when the width of the transition zone between light and dark areas at the contact edges is less than one pixel, and the relative drift between the two geometric centers does not exceed two pixels within a one-second time window, allowing for stable holding.

[0037] The stage maintains a low-amplitude magnetic field and low-amplitude ultrasound for 1 to 3 seconds, allowing the newly formed particle-particle contacts to fully adhere at the point of action. The significance of this holding phase lies in suppressing rebound and minor slippage, especially when there is a slow background flow within the channel, which can significantly improve the pass rate of subsequent re-inspections.

[0038] After the holding period, gradually reduce the magnetic field and turn off the ultrasound. If a dissociation eluent is needed to release microspheres A, B, and neutral microspheres, a short rinse is performed at this stage using the upstream supply fluid. After rinsing, acquire another image frame to confirm that stable contact characteristics still exist, thus completing the processing of this candidate pair.

[0039] Two imaging metrics are defined: first, the width of the light-dark transition band of the contact edge on the camera image plane is no greater than 1 pixel and remains unchanged within a 1-second time window; second, the relative drift between the two geometric centers within a 1-second time window is no greater than 2 pixels. Meeting both metrics is sufficient to determine a stable contact feature. These metrics are used because they are directly related to geometric alignment and mechanical stability. If either metric exceeds the threshold during the holding phase, the stage first releases the auxiliary contact (if present), then releases the contact on side B, and finally releases the contact on side A, withdrawing sequentially to avoid adhesion or secondary aggregation. After withdrawal, the entry for that grid cell is returned to the site list for rescanning.

[0040] When stronger directionality is required, the long and short side coils can be replaced with an octagonal coil group distributed in four quadrants. The octagonal coil group can complete the magnetic field direction switching in a shorter time and is suitable for applications where the grid cell side length is less than 300 micrometers.

[0041] Example 6: When suspended matter within the channel causes a decrease in bright-field imaging contrast, the imaging illumination can be switched to dark-field illumination: light is introduced at a shallow angle at the edge of the quartz window, so that the characterization camera only receives light scattered by the microspheres and particle edges. Dark-field mode significantly improves edge contrast, facilitating the identification of the geometric center against complex backgrounds. In applications requiring more precise control of focus depth, a second characterization camera can be added to observe the same grid cell at a symmetrical angle. Both characterization cameras focus on their respective quartz sides, thus achieving higher axial positioning accuracy without changing the channel thickness. For systems with high particle surface roughness or strong adhesion, the ultrasonic short-pulse sequence can be replaced with a stepped short-pulse sequence: first, outputting two to three consecutive pulses at a lower amplitude, then one to two pulses at a higher amplitude, and finally returning to a lower amplitude for one pulse. The stepped approach more easily overcomes initial static friction without excessively disturbing the aligned contacts without significantly increasing overall energy.

[0042] Example 7: In the addressable microsphere supply and recovery subsystem, microsphere storage tanks are used to store dispersions of microspheres A, B, and neutral microspheres, respectively. Each type of microsphere has an independent cylindrical storage tank with an effective volume of 1 liter to 5 liters. The inner wall is coated with an inert coating to reduce non-specific adsorption. A conical collection area with a collection angle of 60 to 90 degrees is provided at the bottom of each storage tank to facilitate particle concentration and complete evacuation. An exhaust cap is provided at the top of the storage tank to expel trapped air during replenishment and prevent air bubbles from forming in the subsequent injection channel. To prevent sedimentation, a low-speed paddle stirrer is provided in each storage tank with a rotation speed of 30 to 120 revolutions per minute. At the same time, an annular temperature control jacket is provided on the outside of the bottom of the storage tank to maintain the temperature in the range of 18 degrees Celsius to 25 degrees Celsius, so that the viscosity of the dispersion and the Brownian motion of the particles are within a stable range. The recommended volumetric dispersion concentrations for microspheres A and B are between 0.01% and 0.10%, while those for neutral microspheres are between 0.005% and 0.05%. Choosing lower volumetric dispersion concentrations reduces the probability of mutual occlusion within the channel, thereby improving the success rate of single-application identification. Each microsphere is equipped with an independent metering injection assembly, including a solvent-resistant metering pump, a back pressure valve, an in-line deaerator, and disposable injection tubing. The metering pump has an adjustable flow rate range of 1 mL / min to 30 mL / min, with flow fluctuations less than 5% of the set value. The back pressure valve is set between 0.05 MPa and 0.20 MPa to maintain stable laminar flow in the continuous flow window channel. The in-line deaerator discharges trace amounts of gas through a venting membrane, with a single deaerator time not exceeding 60 seconds. The injection tubing uses a parallel tee near the inlet of the continuous flow window channel, with a tee angle of 15° to 30° to reduce shear abrupt changes and avoid disturbing established action sites. The injection branches for microspheres A and B are completely physically isolated, with the tee no more than 30 mm from the main channel inlet. Each branch is equipped with a check valve and a backflow mechanism to prevent backflow and cross-contamination. The dissociation eluent channel is used to reversibly dissociate microspheres A, B, and neutral microspheres from the action site after bridging. This channel consists of an independent storage tank, a metering pump, a temperature control section, and spray heads. The effective volume of the storage tank is 2 to 10 liters. The length of the temperature control section is 200 to 500 mm, ensuring that the eluent reaches the set temperature before entering the spray heads. The spray heads are installed in the release area downstream of the continuous flow viewing window channel, with a nozzle spacing of 5 to 10 mm, a spray angle of 30 to 60 degrees, and a spray distance of 10 to 25 mm, ensuring coverage of the observation window without generating strong turbulence. The dissociation eluent formulation uses a combination of competitive binders, ionic strength modifiers, or reversible bond stability depressants. The advantage of using an independent channel is that it avoids accidental triggering of dissociation during the traction and alignment phases, while allowing precise control of the dissociation duration within the range of 1 to 5 seconds. The recovery filter assembly is installed in the downstream loop of the release zone and includes two parallel filter cartridges and a backwashable structure.The first-stage filtration has a pore size of 10 to 20 micrometers to intercept any potential agglomerates; the second-stage filtration has a pore size of 1 to 5 micrometers to capture individual microspheres. Pressure gauges are installed before and after each filter cartridge. When the pressure difference exceeds 0.05 MPa, filter cartridge replacement or backflushing is triggered. Backflushing involves passing an inert washing solution in reverse through the filter cartridge at a flow rate of 50 to 200 ml per minute for 30 to 120 seconds, while simultaneously activating a pulsating valve to detach deposited microspheres from the filter media surface and collect them in a recovery bottle. The parallel dual-path design allows for switching of filtration channels without shutting down the entire system, thus maintaining continuous production.

[0043] A transparent observation window and a short-stroke sampling valve are installed at the three-way inlet and release area, respectively, and an auxiliary camera is fixed above the observation window. The auxiliary camera acquires grayscale images at 30 to 60 frames per second. The instantaneous occupancy of microspheres in the channel is estimated by counting the number of bright scattering points per unit area, and compared with the number of valid entries identified by the characterization camera. If the deviation exceeds 15%, the flow rate of the quantitative injection component is automatically adjusted or the injection time is temporarily extended. Setting up dual observation points can cover the "post-injection" and "post-dissociation" stages respectively, thereby completing closed-loop correction without interrupting the main process.

[0044] The geometric diameters of microspheres A, B, and neutral microspheres range from 1.0 μm to 3.0 μm, with a geometric standard deviation controlled within 20%. Each microsphere contains a magnetically responsive core with a core volume fraction of 10% to 30%, enabling observable displacement under a magnetic field gradient without significant sedimentation. This size range and core volume fraction were chosen because: too small a diameter makes it difficult to achieve high-contrast recognition on the characterization camera, while too large a diameter increases fluid resistance and the probability of collision; too weak a magnetic response reduces traction efficiency, while too strong a response easily leads to chain aggregation within the channel. Identification groups matching the type label are fixed to the surfaces of microspheres A and B, while the surface of the neutral microsphere is an inert layer. The construction of the identification layer follows the principle of "stronger than Brownian perturbation but weaker than magnetoacoustic coupling retention," meaning that without a magnetoacoustic-optical stage for retention, the affinity of the identification layer is sufficient to resist slight shearing from the flow field, thus achieving visual positioning. In the release zone, the dissociation eluent reduces the stability of the identification layer through competitive binding or environmental changes, causing the binding to dissolve within 1 to 5 seconds. The advantage of choosing this reversible mechanism is that it satisfies the requirements of traction, alignment, and short-term retention, while also enabling point-to-point release and retrieval without leaving any residual microspheres. Each type of microsphere is coated with an imaging recognition layer outside the recognition layer, causing it to produce a distinguishable response in the imaging illumination band characterizing the camera. For example, microsphere A exhibits high brightness, microsphere B exhibits medium brightness, and neutral microspheres exhibit edge-enhanced dark-field scattering. The reason for using three different responses is that the three types of microspheres can be directly separated within a single frame image, reducing filter switching and improving the speed of localization and candidate generation.

[0045] The T-junction of the metering injection component is located 10 to 30 mm after the channel inlet manifold. The injection angle is set to 15 to 30 degrees, allowing the dispersion to enter the mainstream field in a wall-adhering manner. The direct effect of wall-adhering injection is to reduce the turbulent core in the central region, avoiding disruption of the action site pattern of the upstream construction; at the same time, it allows the microspheres to complete velocity matching with the mainstream before entering the observation window, reducing the grazing deviation at the grid cell boundaries. The back pressure valve and the metering pump work together to maintain stable laminar flow, keeping the Reynolds number in the channel in a low range. The benefit of laminar flow is that the lateral diffusion of microspheres is dominated by Brownian motion, and the spatial distribution is predictable, which is beneficial for grid cell-by-grid processing in a row-and-determinant sequence. Microspheres A and B are injected in a time-sharing manner, while neutral microspheres are injected briefly after the two or injected as needed near the converging end. The advantage of using "time-sharing injection" is that it reduces invalid collisions and erroneous attachments between microspheres A and B, making the "closest distance" judgment in candidate generation more stable. The significance of delayed injection of neutral microspheres lies in preserving their role as laterally stable contacts and reducing their interference with the main contact path due to premature placement.

[0046] Once microsphere A or microsphere B enters the observation window and is identified by the characterization camera, the magneto-acoustic-optical stage applies a low-amplitude magnetic field gradient along the line connecting the site list and candidate pairs. Microspheres A and B approach the target site along the connecting line, where their type labels match and their status labels are open. At this point, the recognition layer specifically binds to the target site, typically within 2 seconds. Applying a magnetic field gradient along the connecting line, rather than omnidirectional magnetization, aims to reduce attraction to surrounding unprocessed items, thereby minimizing false binding.

[0047] After confirming stable contact characteristics, the magneto-acoustic-optical actuator lowers the magnetic hold and shuts off the ultrasound, then activates the dissociation eluent channel. The dissociation eluent is sprayed into the release area at a flow rate of 10 to 50 ml / min for 1 to 5 seconds, releasing the microspheres from the contact site. The main stream then transports the free microspheres to the recovery filter assembly. The second-stage filter captures the individual microspheres, while the first-stage filter intercepts a small amount of agglomerates. The filtrate is returned to the closed mixing chamber or enters the waste tank, depending on the on-site process. The spray duration and flow rate of the dissociation eluent are limited within the above range because: too short a time leads to incomplete dissociation, while too long a time increases hydraulic disturbance to newly formed contacts; too low a flow rate makes it difficult to achieve coverage, while too high a flow rate generates localized high shear within the channel.

[0048] If microspheres are still locally retained on the surface after dissociation, a short-arc ultrasonic pulse can be applied to the release area, with a pulse duration of 20 to 50 milliseconds, while simultaneously sweeping the magnetic field direction at a small angle of 2 to 5 degrees. This will cause the microspheres to detach from the surface and be carried away by the mainstream. The combination of "short arc plus short pulse" can handle a small number of retention points without disrupting stable contact characteristics.

[0049] An auxiliary camera acquires grayscale images 5 to 10 millimeters upstream of the T-junction point, calculating the number of bright scattering points per unit area as an indicator of injection quality. If the indicator is low, the system increases the flow rate of the quantitative injection component by 5% to 10% and maintains it for 10 to 30 seconds; if the indicator is high, the opposite adjustment is made. Image counting is used instead of simple flow meter readings because image counting more closely reflects the actual occupancy within the channel and can reflect deviations caused by microsphere aggregation or pipeline stagnation. In the release zone observation window, the auxiliary camera counts the decrease in the number of bright scattering points per unit area after passing through the release zone. If the decrease is less than 80%, the system automatically extends the dissociation eluent spray by 1 to 2 seconds or increases the flow rate by 10% to 20%. This threshold is derived from an empirical upper limit of single dissociation efficiency, which can improve the recovery rate without increasing the risk of over-rinsing. When the differential pressure of any filter cartridge exceeds 0.05 MPa, it automatically switches to the parallel backup channel and initiates backwashing. After backflushing, perform a blank channel flush for 10 to 20 seconds to confirm that the differential pressure has recovered to within ±10% of the initial value before switching back to the main channel. The purpose of setting up a parallel backup channel is to achieve maintenance without shutting down the system and to avoid affecting the upstream traction cycle due to filter shutdown.

[0050] In one alternative implementation, for batches with low site density, alternating pulse injection can be used: microsphere A is pulsed for 2 to 4 seconds, paused for 2 to 4 seconds, and then microsphere B is pulsed again; neutral microspheres are injected as needed at the converging end for 1 to 2 seconds. Alternating pulses can increase the effective contact probability under low occupancy while avoiding excessively high microsphere density throughout the viewing window. When more uniform coverage of the release area is required, the dissociation eluent spray head can be replaced with a slow-speed rotating spray head with a rotation speed of 30 to 90 revolutions per minute and 6 to 12 nozzles. Rotary spraying can cover corner areas at the same total flow rate, reducing local residue. A short straight pipe is connected in series before filtration, and a permanent magnet ring clamp is fitted on the outside, causing the microspheres to accumulate on the pipe wall for a short time, and then the magnetic field is instantly released and they enter the filter cartridge. This method helps reduce the instantaneous load on the filter cartridge and delays filter clogging.

[0051] Example 8: The controller includes a processor, memory, timing generator, image acquisition interface, coil drive interface, ultrasonic drive interface, light source control interface, and external communication interface, and is installed in a closed cabinet. The processor is a multi-core general-purpose processor with a main frequency of 2.0 GHz to 3.5 GHz, and is equipped with 16 Gigabytes to 64 Gigabytes of memory. The timing generator provides a 1 microsecond-level time reference for synchronization with the characterization camera, area array light source, magnetic field coil, and ultrasonic transducer. The image acquisition interface is directly connected to the characterization camera via a high-speed data cable, with a single-frame transmission delay of no more than 10 milliseconds. The coil drive interface is connected to the power driver of the magnetic field coil via a shielded cable, and the rise and fall times of the drive current can be set within the range of 5 milliseconds to 20 milliseconds. The ultrasonic drive interface and the power supply of the ultrasonic transducer array use an independent power supply channel to avoid crosstalk. The light source control interface is connected to the area array light source with a low-latency signal, and the switching time is no more than 20 milliseconds. The external communication interface shares Ethernet communication with the partition injection and in-situ solidification subsystems, with a bandwidth of 100 Mbps to 1000 Mbps. The cable bundles between the controller cabinet and the magneto-acoustic-optical actuator are routed using independent cable trays. The coil drive cable bundles and image data cable bundles are spaced at least 100 mm apart, and a 90-degree crossover is used at intersections to reduce inductive crosstalk. Temperature-controlled air ducts are installed within the cabinet to maintain the core components at 20 to 30 degrees Celsius.

[0052] The controller generates and issues bridging action commands in a closed-loop manner of "positioning - traction - alignment - stability maintenance - re-inspection - retracement" within each grid cell, and processes them sequentially in all grid cells according to the row and column scanning path to form the soil improvement target.

[0053] The system comprises the following components: **Site List:** Stored according to the grid cells representing the camera's field of view. Each record includes a grid number, item number, type label, status label, and image coordinates. **Candidate Pair List:** A set of paired items obtained from the site list. Each record includes a type A item number, a type B item number, an optional neutral item number, the coordinates of the connection endpoints, and a temporary lock flag. **Stable Contact List:** Records item pairs that have passed the stable contact feature determination and optional auxiliary items. **Unprocessed Reason List:** Records items that did not enter the candidate list or did not complete contact within the current grid cell, along with their unprocessed reason labels. Unprocessed reason labels are limited to no complementarity, disabled, or path blocked.

[0054] The controller defines grid cells with side lengths ranging from 200 to 600 micrometers within the characterization camera's field of view. The scanning order follows a top-down, then left-to-right row-column sequence. The direct advantage of this fixed order is that the target direction switching amplitude of the magnetic field coils is smaller, the steady-state time of coil drive is shortened to the 5-20 ms range, and the execution paths of different grid cells are avoided from intersecting. Entries in the alignment point list are marked with a status label indicating they are to be activated. The controller instructs the area array light source to perform local illumination within that grid cell for a duration of 300-2000 ms. Immediately after illumination, an image is captured using imaging illumination. If the entry can achieve observable attachment to the corresponding type of addressable microsphere, the status label is set to "on"; otherwise, it is set to "disabled." Local illumination has two practical effects: first, it removes the blocking layer only at the necessary locations, reducing the activation probability of irrelevant areas; second, immediate image capture after illumination directly verifies successful activation, reducing the number of subsequent undoings.

[0055] Starting with the first entry in the site list, each entry in the A-type list is enumerated sequentially. For each A-type entry, within the same raster cell, B-type entries are searched in ascending order of Euclidean distance from image coordinates, provided both entries are in an open state. Upon finding a B-type entry, a candidate pair is generated, and a temporary lock flag is set for both entries. Then, neutral-type entries are searched on both sides of the line connecting the candidate pairs in ascending order of proximity. If an open neutral-type entry exists, the closest one is registered as a selectable auxiliary entry. Choosing the "closest distance" strategy shortens the approach path and coil direction switching time, reducing disturbance to nearby unprocessed entries.

[0056] When the same entry appears in multiple candidate pairs, the controller retains or removes candidate pairs according to a fixed order rule: Candidate pairs that contain both open type A and open type B entries are retained first; if multiple candidate pairs meet or do not meet this condition, the one with the smaller type A entry number in the site list is retained; if type A entry numbers are the same, the one with the smaller type B entry number is retained. Entries in removed candidate pairs are temporarily unlocked and returned to the site list. This fixed order avoids weight-based randomness, ensuring consistent results for the same data across different batches.

[0057] For candidate pairs without conflicts, bridging action commands are generated. These commands contain the following fixed fields and are executed in a fixed order: 6.1 Orientation Traction Field: The control magnetic field coil establishes a low-amplitude magnetic field gradient along the candidate pair connection direction, lasting 100 to 1000 milliseconds. 6.2 Attitude Integration Field: At the approach end, the control ultrasonic transducer outputs short pulses and performs short-arc correction, with an arc angle of 2 to 8 degrees and 3 to 7 pulses. 6.3 Auxiliary Contact Field (Optional): If a neutral entry exists, a short-term magnetic traction is applied along the connection normal for 50 to 300 milliseconds after the main contact is formed. 6.4 Hold and Release Field: The low-amplitude magnetic field and low-amplitude ultrasound are maintained together for 1 to 3 seconds, after which the upstream command releases the addressed microsphere and performs a short flush. The bridging action commands are issued using framed timing, with each frame containing a timestamp, grid number, and entry number. The round-trip delay between the controller and the magneto-acoustic-optical actuator is no greater than 10 milliseconds.

[0058] After each bridging action command is executed, the controller instructs the camera to acquire 1 to 3 frames of images and determine stable contact characteristics: the width of the light-dark transition zone at the contact edge is no more than 1 pixel, and the relative drift of the two geometric centers within a 1-second time window is no more than 2 pixels. If any condition is not met, the process is undone in the order of "first release the auxiliary contact, then release side B, and finally release side A," and the entry is returned to the site list. When there are still unprocessed entries in the site list and no new candidate pairs are generated, the controller performs a backscan along a matrix path within the grid cell. If a local replacement of a stable contact is found that can release a key entry and will not generate a new conflict after the replacement, the stable contact is undone, and a bridging action command is generated for the new combination. The backscan terminates when there are no unprocessed entries in the grid cell, or when there are unprocessed entries but no complementary relationship or replaceable path exists. After completing the current grid cell, the controller switches to the next grid cell, with the order fixed as top to bottom and then left to right. After all grid cells have been processed, the controller merges the stable contact table, generates soil improvement targets, and records the untreated reason labels for entries in the untreated reason table. The soil improvement targets include stable contact counts, distribution locations, and corresponding grid numbers, which are used for subsequent mapping to field injection zones.

[0059] When the same candidate pair is run in different batches, repeatability is guaranteed as long as the order of the site list entries, imaging coordinates, and status labels are consistent, and the field order, duration, and short arc angle of the bridging action commands generated by the controller remain consistent. Soil improvement targets are recorded using the grid number as the primary key, noting the location and number of stable contacts within each grid cell, and labeling the reason for non-treatment. This list is transmitted to the partitioned injection and in-situ solidification subsystem in the form of a data file, with a transmission delay of no more than 200 milliseconds.

[0060] Example 9: The device structure and connection relationship of the zoned injection and in-situ solidification subsystem are described in detail below. Injection zones are established on the target high liquid limit soil layer, corresponding one-to-one with or proportionally mapped to grid units. The side length of a single injection zone is 0.5 meters to 1.5 meters, selected based on site workability and the stable contact density in the soil improvement target. Choosing a smaller zone side length allows for improved injection precision in areas with lower stable contact density, thereby enhancing the consistency of subsequent solidification. The injection manifold is a multi-port distribution structure with 8 to 24 ports, connecting the injection hose to the zoned injection interface. The zoned injection interface can be a shallowly buried injection port or a borehole packer injection port, with a common burial depth of 0.5 meters to 2.0 meters. Each injection port is equipped with a one-way valve and a pressure display device; the one-way valve opening pressure is 0.05 MPa to 0.15 MPa. The injection pump is a variable-speed metering pump with a single-channel flow rate ranging from 0.5 liters per minute to 3.0 liters per minute. An outlet pressure stabilizer maintains the pressure between 0.10 MPa and 0.30 MPa. A return branch is used to recover excess slurry after saturation in the injection zone and to prevent excessive seepage. Each injection zone is equipped with a surface-mounted or shallowly buried ultrasonic transducer operating at a frequency of 20 kHz to 80 kHz for a duration of 2 to 10 minutes. The temperature control channel utilizes a surface heating strip or a warm water circulation plate to maintain the soil temperature between 20°C and 35°C. This temperature range is chosen to reduce pore water viscosity and promote the stabilization of existing particle-particle contacts at the microscopic interface. Echo probes and surface displacement markers are deployed at the zone boundaries. The echo probes emit and receive short bursts at a frequency of 1 to 4 times per minute, and the displacement markers are read using a dial indicator or optical scale. Both types of readings are used to confirm consistency during the curing process.

[0061] The controller maps the grid numbers in the soil improvement target to the field injection zone numbers. There are two mapping methods: one is proportional mapping, where if the grid cell side length and the injection zone side length are in a fixed ratio, the mapping is done directly in a row-determinant order; the other is aggregation mapping, where several adjacent grid cells are aggregated into one injection zone, with the aggregation number ranging from 2 to 9. After mapping, injection is performed in a row-determinant order to maintain consistency with the upstream processing order and facilitate comparison and verification. In each injection zone, the injection pump injects the reconstructed regenerated micronized soil amendment slurry at a flow rate of 0.5 to 3.0 liters per minute, with a single zone injection volume of 2 to 12 liters. If stable backflow occurs in the backflow branch during injection and does not decrease for 30 consecutive seconds, the zone is considered saturated. The advantage of using "backflow stability judgment" is that it avoids excessive pressurization and reduces secondary disturbance to the pore structure. The ultrasonic and temperature control channels are activated immediately after injection. The ultrasonic treatment time is 2 to 10 minutes, with temperature control maintained at 20 to 35 degrees Celsius for a duration of 10 to 60 minutes. Adopting the sequence of "ultrasonic treatment first, temperature control throughout" first eliminates small gaps through localized microfluidics, and then stabilizes the interface through temperature, thereby improving the uniformity of the cured body. In each injection zone, the echo probe performs two independent measurements before and after curing. If the difference in the arrival time of the characteristic echoes from the two measurements does not exceed 10 percent, and the surface displacement marker shows no continuous increase within 10 minutes, the zone is considered passed, and the process proceeds to the next zone; if it fails, the ultrasonic treatment time is extended by 1 to 3 minutes or the temperature control is increased by 2 to 5 degrees Celsius, and two more independent measurements are performed.

[0062] After all partitions are completed, the partition injection and in-situ solidification subsystem records the passage status of each partition and sends it back to the controller. If a partition cannot reach saturation due to geological obstacles, the reason for not handling it is recorded as "path blocked" and marked on-site for subsequent handling.

[0063] On-site, the equipment is advanced sequentially from top to bottom and then from left to right, minimizing movement paths and pipeline switching, reducing the probability of misoperation caused by pipeline intersections, and facilitating the comparison of on-site inspection data with soil improvement targets. Maintaining pressure between 0.10 MPa and 0.30 MPa and flow rate between 0.5 L / min and 3.0 L / min allows the regenerated micro-powder-based soil amendment slurry to enter existing pores without damaging the pore skeleton. Excessive pressure can cause pore expansion and mud-water surge, while insufficient pressure cannot overcome capillary resistance. The combination of ultrasonic treatment followed by continuous temperature control dissipates micro-layering generated during injection near the injection port and maintains particle-particle contact for a certain period, thereby improving the predictability of stable contact characteristics at the on-site scale.

[0064] refer to Figure 1The schematic diagram of the pretreatment and addressable site construction subsystem illustrates a continuous flow processing device. This device features multiple functional sections arranged horizontally to precisely construct contact sites on the surface of regenerated micronized powder particles. At the front end of the system is a closed mixing chamber, a cylindrical pressure vessel with a volume ranging from 10 to 50 liters. Inside the chamber is a three-bladed agitator for thorough mixing of the regenerated micronized powder-based soil amendment slurry, with mixing time controlled between 5 and 15 minutes. An external temperature-controlled circulation jacket maintains the temperature stably between 15 and 35 degrees Celsius, ensuring the consistency of the particle surface state before reaction. The bottom outlet of the closed mixing chamber is connected to the inlet filtration and pressure stabilization section via a pipe. The filtration and pressure stabilization section includes replaceable filter cartridges with pore sizes ranging from 50 to 100 micrometers. This section also includes a pressure stabilization buffer chamber to eliminate flow pulsations, reducing flow fluctuations to below 5% of the total flow rate. After filtration and pressure stabilization, the slurry enters a continuous flow window channel. This channel has a parallel plate thin-layer structure, with a length of 300 mm to 600 mm and a channel gap precisely controlled between 0.3 mm and 1.0 mm. The upper plate of the channel is a quartz window plate with an optical transmittance of no less than 90% in the 350 nm to 450 nm wavelength range, while the lower plate is an inert substrate. A surface array light source is installed above the channel at the location corresponding to the initiation region, with a working wavelength selectable at 365 nm or 405 nm, at a distance of 50 mm to 120 mm from the quartz window plate. A alignment mask is placed between the surface array light source and the quartz window plate, with a regularly or quasi-randomly distributed array of openings on the mask. The opening diameter is 80 μm to 200 μm, and the opening spacing is 200 μm to 500 μm. Light passes through the mask openings and irradiates the thin layer of slurry flowing through the initiation region for 2 to 8 seconds, forming discrete activation sites on the particle surface. A capping zone is positioned 30 to 80 millimeters downstream of the initiation zone, equipped with a full light-shielding shield and a cooling baffle. The inner surface of the light-shielding shield has a matte black coating, allowing the thin layer leaving the light-illuminated area to enter the fully light-shielded environment within 10 milliseconds. The cooling baffle is 5 to 10 degrees Celsius cooler than the initiation zone, rapidly terminating residual reactions and maintaining clear boundaries at the activation sites. A functionalized spray head is positioned 20 to 60 millimeters downstream of the capping zone. This spray head has equidistant nozzles spaced 5 to 10 millimeters apart along the channel width. The spray head contains two independent liquid supply circuits, supplying solutions for ligand A and ligand B respectively, with each solution sprayed for 1 to 3 seconds, followed by a 1 to 2 second inert wash interval. This sequential spraying method ensures preferential binding of ligands A and B within their respective coverage areas, forming type tags at the activation sites. Downstream of the functionalized spray head is a status label setting and protection station, which is equipped with a photosensitive barrier spraying assembly. A removable photosensitive barrier layer is sprayed to ensure that all newly generated action sites are uniformly in an activated state. Subsequently, a gentle air dryer completes surface drying within 5 to 20 seconds with a low airflow of 0.5 to 1.5 meters per second, preventing particle re-aggregation.Finally, the thin-layer flow passes through the outlet manifold into the outlet collection section, completing the collection of particles with addressable sites. The entire system achieves precise control of site density and spatial distribution through continuous flow, and the site formation efficiency can be stably maintained above 80%.

[0065] refer to Figure 2 , Figure 2 The graph shows the relationship between magnetic field gradient intensity and microsphere displacement velocity. The horizontal axis represents the magnetic field gradient intensity in millitalas per millimeter (mT / mm), ranging from 0.5 to 5.0 mT / mm. The vertical axis represents the microsphere displacement velocity in micrometers per second (µS), ranging from 5 to 55 µS / mm. The graph includes two experimental curves: the solid curve represents microsphere A with a magnetic response core volume fraction of 10%, and the dashed curve represents microsphere B with a magnetic response core volume fraction of 30%. Both curves show a rapid increase followed by a flattening trend. Within the magnetic field gradient intensity range of 0.5 to 3.0 mT / mm, the displacement velocity of both types of microspheres increases significantly with increasing magnetic field gradient, from 5 µS / mm to approximately 30 µS / mm for microsphere A and from 5 µS / mm to approximately 33 µS / mm for microsphere B. When the magnetic field gradient intensity exceeds 3.0 mT / mm, the growth trend of displacement velocity slows down significantly; at 5.0 mT / mm, the displacement velocity of microsphere A reaches approximately 52 µS / mm, and that of microsphere B reaches approximately 54 µS / mm. The optimal operating range, as indicated in the figure, is 1.5 to 3.0 millitas per millimeter. Within this range, the microsphere displacement velocity is moderate, allowing for clear observation by the characterization camera without causing positioning errors or overshoot due to excessive speed. The experimental conditions were: a channel gap of 0.5 mm, a microsphere diameter of 2.0 μm, a temperature of 25 degrees Celsius, and a measurement time of 100 to 1000 milliseconds.

[0066] refer to Figure 3 and Figure 4 , Figure 3The relationship between the channel gap of the continuous flow window and the uniformity coefficient of the site density is shown. The horizontal axis represents the channel gap in millimeters, ranging from 0.2 to 1.8 mm. The vertical axis represents the site density uniformity coefficient, expressed as a percentage, ranging from 50% to 100%. The curve exhibits a distinct bell-shaped distribution, rising first and then falling. At a channel gap of 0.2 mm, the uniformity coefficient is approximately 55%. As the gap increases, the uniformity coefficient rapidly increases, reaching approximately 80% at 0.6 mm, and peaking at 96.5% at 1.0 mm. This is because within this gap range, the thin layer thickness is moderate, resulting in a uniform irradiation dose distribution in the initiation region after the area array light source penetrates the quartz window plate, avoiding boundary effects caused by an excessively thin layer or light intensity attenuation caused by an excessively thick layer. When the channel gap exceeds 1.0 mm, the uniformity coefficient begins to decrease, dropping to approximately 45% at 1.8 mm. This is due to the increased difference in irradiation dose along the depth direction caused by an excessively thick layer, resulting in inconsistent irradiation energy received by particles near the upper and lower surfaces. The recommended working range, as indicated in the figure, is 0.3 to 1.0 mm, within which the site density uniformity remains above 90%. Experimental conditions were: irradiation time 6 seconds, flow rate 20 mL / min, array light source wavelength 365 nm, and mask aperture spacing 300 μm.

[0067] Figure 4 This graph illustrates the inverse effects of dissociation eluent concentration on two key performance indicators. The horizontal axis represents the dissociation eluent concentration in millimoles per liter (mmol / L), ranging from 10 to 80 mmol / L. The vertical axis represents the recovery or retention rate, expressed as a percentage, ranging from 10% to 100%. The graph contains two curves: the solid curve represents the microsphere recovery rate, showing a continuous upward trend; the dashed curve represents the bridging retention rate, showing a continuous downward trend. At a concentration of 10 mmol / L, the microsphere recovery rate is approximately 25%, and the bridging retention rate is approximately 95%. As the dissociation eluent concentration increases, the disruptive effect of the competitive binder on the binding between the recognition layer and the action site intensifies, making it easier for the microspheres to dissociate from the particle surface, resulting in a continuous increase in recovery rate, reaching approximately 92% at 80 mmol / L. Simultaneously, higher eluent concentrations also disturb the established interparticle bridging contacts, leading to a decrease in bridging retention rate, which drops to approximately 20% at 80 mmol / L. The two curves intersect at a concentration of approximately 40 mmol / L, which is marked as the equilibrium point. The optimal operating range is indicated in the figure as 30–50 mmol / L. Within this concentration range, the microsphere recovery rate can reach over 60%, while the bridging retention rate remains above 50%, achieving a balance between recovery efficiency and structural stability. Experimental conditions were: spray time 3 seconds, flow rate 30 mL / min, EDTA as the competing binder, and temperature 25°C.

[0068] Figure 5This study demonstrates the effect of ultrasonic frequency on the elimination of porosity within a solidified body and compares the differences between different treatment times. The horizontal axis represents ultrasonic frequency in kilohertz (kHz), ranging from 20 to 100 kHz. The vertical axis represents porosity elimination rate as a percentage, ranging from 30% to 100%. The graph includes two curves: the solid curve represents the case with an ultrasonic treatment time of 5 minutes, and the dashed curve represents the case with an treatment time of 10 minutes. Both curves show a trend of first rising and then falling, forming a peak range. At a frequency of 20 kHz, the porosity elimination rate is approximately 40% after 5 minutes of treatment and approximately 48% after 10 minutes. As the frequency increases to 70 kHz, both curves reach their peaks, with an elimination rate of approximately 78% after 5 minutes and approximately 85% after 10 minutes. This is because the cavitation and microfluidic effects generated by ultrasound at the slurry-soil interface are most significant in the 50-80 kHz frequency range, effectively eliminating the small voids formed during the injection process. When the frequency exceeds 80 kHz, the elimination rate begins to decrease. This is because the attenuation of high-frequency ultrasound in the medium increases, resulting in a decrease in penetration depth. The optimal operating range is marked as 50 to 80 kHz in the figure. Comparing the two curves, it can be seen that extending the treatment time to 10 minutes can increase the elimination rate by approximately 7 to 10 percentage points. The experimental conditions were: injection pressure of 0.15 MPa, partition side length of 1.0 meter, temperature control of 28 degrees Celsius, and measurement method of CT scan.

[0069] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result according to substantially the same method falls within the scope of the present invention. Therefore, the scope of the present invention is defined only by the appended claims.

Claims

1. A soil improvement system for high liquid limit soil based on recycled micronized solid waste modification, characterized in that, The system includes: a pre-processing and addressable site construction subsystem for forming action sites with a type tag and a status tag on the surface of regenerated micronized particles in a regenerated micronized soil amendment slurry; a grid imaging and execution subsystem, including a magneto-acoustic-optical stage and a characterization camera, for locating, pulling, aligning, and stabilizing the aforementioned action sites within a grid cell; and an addressable microsphere supply and recovery subsystem, which provides A microspheres, B microspheres, and neutral microspheres corresponding to the aforementioned type tags. It achieves reversible binding and dissociation between the aforementioned microspheres and the aforementioned action sites; a controller configured to execute a grid-sequential-interlocked bridging execution algorithm to generate and issue bridging action commands in each grid cell, thereby forming preferentially distributed directional bridging contacts in the aforementioned regenerated micro-powder-based soil amendment slurry; a partitioned injection and in-situ solidification subsystem, used to inject the reconstructed aforementioned regenerated micro-powder-based soil amendment slurry into the high liquid limit soil layer in partitions according to a soil improvement target output by the aforementioned controller, and trigger a multiphase synergistic solidification procedure.

2. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 1, characterized in that, The pretreatment and addressable site construction subsystem includes: a closed mixing chamber, a continuous flow window channel, an initiation zone, a capping zone, and a functionalized spray head; wherein, the initiation zone activates the particle surface with an area array light source, the capping zone terminates unreacted sites by shading, and the functionalized spray head sequentially introduces A ligand and B ligand to assign the aforementioned action site type labels, and the status labels of all newly generated action sites are initially set to pending activation.

3. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 2, characterized in that, The addressable microsphere supply and recovery subsystem includes a microsphere storage tank, a metering injection component, a dissociation eluent channel, and a recovery filtration component. Microspheres A and B reversibly bind to action sites that match their type labels and have an open status label, while neutral microspheres reversibly attach to neutral action sites. After bridging, the dissociation eluent supplied through the aforementioned dissociation eluent channel causes microspheres A, B, and neutral microspheres to reversibly dissociate from their action sites and be collected by the aforementioned recovery filtration component.

4. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 3, characterized in that, The grid imaging and execution subsystem includes a magnetic field coil, an ultrasonic transducer, an area array light source, and a characterization camera. The controller establishes grid cells in the observation window of the aforementioned continuous flow window channel through a grid definition unit, and images grid cells one by one according to the row and column scanning path, recording the visible positions of microsphere A, microsphere B, and neutral microsphere to generate a site list.

5. The high liquid limit soil improvement system based on recycled micro-powder-based solid waste modification as described in claim 4, characterized in that, The controller is configured to execute a grid sequencing-interlocking bridged execution algorithm, which includes the following steps: Step 5-1, State confirmation and on-site activation: For entries in the aforementioned site list whose state label is to be activated, local illumination is performed within the corresponding grid cell using the aforementioned area array light source, causing their state labels to change to open or disabled; Step 5-2, Candidate generation: Within the same grid cell, taking a type A entry as a reference, starting from the first entry in the aforementioned site list, each entry is enumerated sequentially, and the closest type B entry with an open state label is found to form a candidate pair. Neutral type entries with open state labels are then searched on both sides of the line connecting these candidate pairs as optional auxiliary entries; Step 5-3, Conflict detection and resolution: When the same entry is referenced by multiple candidate pairs... Candidate pairs are retained or removed and temporary locks are released according to a fixed order rule; Step 5-4, Bridging action command generation and issuance: a bridging action command is generated for each non-conflicting candidate pair and issued to the magnetic-acoustic-optical execution station for execution; Step 5-5, Immediate review and cancellation: after each bridging action command is executed, the status of the relevant entries is reviewed immediately, and the contact is released according to a fixed cancellation order if necessary; Step 5-6, Backscan and interlock replacement: within the same grid cell, the stable contact is replaced, and interlock replacement is performed when the preset conditions are met; Step 5-7, Grid switching and merging: all grid cells are processed sequentially according to the aforementioned matrix scanning path, and stable contacts are merged to generate soil improvement targets, while recording an unprocessed reason label.

6. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 5, characterized in that, The fixed order rule defined in step 5-3 is as follows: candidate pairs that simultaneously contain both type A entries with the status label "open" and type B entries with the status label "open" are preferentially retained; if multiple candidate pairs meet or do not meet the aforementioned priority conditions, the type A entry with the smaller sequence number in the site list is retained; if the type A entries have the same sequence number, the type B entry with the smaller sequence number is retained; entries in removed candidate pairs are temporarily unlocked and returned to the site list.

7. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 6, characterized in that, The bridging action command includes fixed fields and is executed in a fixed order: a) a directional traction field, which controls the aforementioned magnetic field coil to apply a magnetic field gradient along the direction of the candidate pair connection line, causing the A microspheres and B microspheres combined with the A and B entries to displace in the same direction and drive the corresponding particles to move closer; b) an attitude integration field, which controls the aforementioned ultrasonic transducer to output pulses to eliminate lateral misalignment and complete alignment with short-arc correction; c) an optional auxiliary contact field, which, after the main contact is formed, applies short-term magnetic traction along the connection line normal, so that the neutral microsphere and the corresponding action site form a laterally stable contact. d) A hold and release field that maintains magnetic-acoustic co-holding until a stable contact characteristic is observed by the camera, then the magnetic field is turned off and the aforementioned dissociation eluent channel is opened to achieve reversible dissociation and recovery of microspheres A, B and neutral microspheres.

8. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 7, characterized in that, The candidate generation defined in step 5-2 is performed in a fixed order of the site list within each grid cell. The generation and traversal of the site list follow a row-and-column scanning path from top to bottom and then from left to right. The determination of the closest spatial location is based on the minimum calculated value representing the camera imaging coordinate difference, thereby avoiding the intersection of execution paths of different candidate pairs within the same grid cell.

9. The high liquid limit soil improvement system based on recycled micronized solid waste modification as described in claim 8, characterized in that, After completing the processing of all grid cells, the controller outputs the soil improvement target and assigns an unprocessed reason label to any unprocessed item. The type of this unprocessed reason label is limited to one of the following: no complement, disabled, or path blocked. At the same time, the controller executes a pre-curing verification process, extracting several stable contacts in each grid cell for secondary independent scanning and verification. After the verification is passed, the defined partition injection and in-situ curing subsystems are allowed to perform curing of the corresponding partitions.

Images